wikimedia/mediawiki-extensions-DiscussionTools
1
0
Fork 0
mirror of https://gerrit.wikimedia.org/r/mediawiki/extensions/DiscussionTools synced 2024-12-19 03:21:42 +00:00
Code Issues Packages Projects Releases Wiki Activity
mediawiki-extensions-Discus.../tests/cases/pl-big-oldparser/pl-big-oldparser.json

5113 lines
109 KiB
JSON
Raw Normal View History

Integration tests for parsing Tests that handle a specific case and describe what they are testing would be nice… but tests that just document the current status to avoid regressions are also okay and easier to add. Change-Id: I0b3530ae0e77de70932aaf623f5290d1876699a0
2019-10-24 12:59:42 +00:00
[
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
{
"placeholderHeading": false,
"type": "heading",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"4/0/0",
"4/0/1"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"headingLevel": 2,
"level": 0,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "h-Stok-2018-09-29T06:23:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "h-Kalendarium-2018-09-29T06:23:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-09-29T06:23:00.000Z",
"author": "Stok",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"6/0",
"7/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"7/1",
"7/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"7/4/2",
"7/4/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 1,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Stok-2018-09-29T06:23:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Stok-2018-09-29T06:23:00.000Z-Kalendarium",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-09-29T07:06:00.000Z",
"author": "Cybularny",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"9/0/0",
"9/0/3/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"9/0/1",
"9/0/3/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"9/0/3/1",
"9/0/3/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Cybularny-2018-09-29T07:06:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Cybularny-2018-09-29T07:06:00.000Z-Stok-2018-09-29T06:23:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-01T18:20:00.000Z",
"author": "Zwiadowca21",
CommentParser: Output display name if different to username The only normalisation we apply for comparison is lowercasing. Change-Id: Id3d57c2066429fcedc7dcc091e74ed46e17060f1
2022-02-04 18:16:24 +00:00
"displayName": "Zwiadowca",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"11/0/0/0/0/0",
"11/0/0/0/12/25"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"11/0/0/0/10",
"11/0/0/0/12/25"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"11/0/0/0/12/1",
"11/0/0/0/12/25"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 3,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Zwiadowca21-2018-10-01T18:20:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Zwiadowca21-2018-10-01T18:20:00.000Z-Cybularny-2018-09-29T07:06:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
}
]
}
]
}
]
},
{
"placeholderHeading": false,
"type": "heading",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"13/0/0",
"13/0/1"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"headingLevel": 2,
"level": 0,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "h-Wostr-2018-09-30T01:22:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "h-Zmiany_interwiki_w_Wikidanych-2018-09-30T01:22:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-09-30T01:22:00.000Z",
"author": "Wostr",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"15/0",
"19/3/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"19/0",
"19/3/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"19/3/1",
"19/3/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 1,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Wostr-2018-09-30T01:22:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Wostr-2018-09-30T01:22:00.000Z-Zmiany_interwiki_w_Wikidanych",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-09-30T07:43:00.000Z",
"author": "Eurohunter",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"21/0/0/0/0/0",
"21/0/0/0/7/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"21/0/0/0/4",
"21/0/0/0/7/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"21/0/0/0/7/2",
"21/0/0/0/7/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 3,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Eurohunter-2018-09-30T07:43:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Eurohunter-2018-09-30T07:43:00.000Z-Wostr-2018-09-30T01:22:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [
"Comment skips indentation level"
],
"replies": [
{
"type": "comment",
"timestamp": "2018-09-30T07:52:00.000Z",
"author": "Eurohunter",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"21/0/0/0/8/0/0",
"21/0/0/0/8/0/8/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"21/0/0/0/8/0/5",
"21/0/0/0/8/0/8/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"21/0/0/0/8/0/8/2",
"21/0/0/0/8/0/8/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 4,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Eurohunter-2018-09-30T07:52:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Eurohunter-2018-09-30T07:52:00.000Z-Eurohunter-2018-09-30T07:43:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-09-30T07:55:00.000Z",
"author": "Eurohunter",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"21/0/0/0/8/0/9/0/0",
"21/0/0/0/8/0/9/0/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"21/0/0/0/8/0/9/0/1",
"21/0/0/0/8/0/9/0/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"21/0/0/0/8/0/9/0/4/2",
"21/0/0/0/8/0/9/0/4/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 5,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Eurohunter-2018-09-30T07:55:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Eurohunter-2018-09-30T07:55:00.000Z-Eurohunter-2018-09-30T07:52:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
}
]
}
]
}
]
},
{
"type": "comment",
"timestamp": "2018-09-30T10:42:00.000Z",
"author": "WTM",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"23/0",
"23/16/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"23/7",
"23/10/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"23/13",
"23/16/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"23/10/2",
"23/10/27"
],
[
"23/16/2",
"23/16/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 1,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-WTM-2018-09-30T10:42:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-WTM-2018-09-30T10:42:00.000Z-Zmiany_interwiki_w_Wikidanych",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-09-30T11:09:00.000Z",
"author": "Eurohunter",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"25/0/0/0",
"25/0/5/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"25/0/2",
"25/0/5/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"25/0/5/2",
"25/0/5/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Eurohunter-2018-09-30T11:09:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Eurohunter-2018-09-30T11:09:00.000Z-WTM-2018-09-30T10:42:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-09-30T16:32:00.000Z",
"author": "Yashaa",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"25/0/6/0/0/0",
"25/0/6/0/5/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"25/0/6/0/2",
"25/0/6/0/5/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"25/0/6/0/5/2",
"25/0/6/0/5/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 3,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Yashaa-2018-09-30T16:32:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Yashaa-2018-09-30T16:32:00.000Z-Eurohunter-2018-09-30T11:09:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-09-30T21:24:00.000Z",
"author": "Eurohunter",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"25/0/6/0/6/0/0",
"25/0/6/0/6/0/8/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"25/0/6/0/6/0/5",
"25/0/6/0/6/0/8/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"25/0/6/0/6/0/8/2",
"25/0/6/0/6/0/8/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 4,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Eurohunter-2018-09-30T21:24:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Eurohunter-2018-09-30T21:24:00.000Z-Yashaa-2018-09-30T16:32:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-01T10:18:00.000Z",
"author": "Yashaa",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"25/0/6/0/6/0/9/0/0",
"25/0/6/0/6/0/9/0/4/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"25/0/6/0/6/0/9/0/1",
"25/0/6/0/6/0/9/0/4/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"25/0/6/0/6/0/9/0/4/2",
"25/0/6/0/6/0/9/0/4/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 5,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Yashaa-2018-10-01T10:18:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Yashaa-2018-10-01T10:18:00.000Z-Eurohunter-2018-09-30T21:24:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
}
]
}
]
}
]
}
]
},
{
"type": "comment",
"timestamp": "2018-09-30T22:17:00.000Z",
"author": "WTM",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"27/0",
"27/18/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"27/15",
"27/18/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"27/18/2",
"27/18/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 1,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-WTM-2018-09-30T22:17:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-WTM-2018-09-30T22:17:00.000Z-Zmiany_interwiki_w_Wikidanych",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-01T06:39:00.000Z",
"author": "Panek",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"29/0/0",
"29/0/4/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"29/0/1",
"29/0/4/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"29/0/4/2",
"29/0/4/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Panek-2018-10-01T06:39:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Panek-2018-10-01T06:39:00.000Z-WTM-2018-09-30T22:17:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-01T09:52:00.000Z",
"author": "Panek",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"29/0/5/0/0",
"29/0/5/2/4/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"29/0/5/2/1",
"29/0/5/2/4/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"29/0/5/2/4/2",
"29/0/5/2/4/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 3,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Panek-2018-10-01T09:52:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Panek-2018-10-01T09:52:00.000Z-Panek-2018-10-01T06:39:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
}
]
},
{
"type": "comment",
"timestamp": "2018-10-01T20:23:00.000Z",
"author": "Barcival",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"29/2/0",
"29/2/8/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"29/2/5",
"29/2/8/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"29/2/8/2",
"29/2/8/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Barcival-2018-10-01T20:23:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Barcival-2018-10-01T20:23:00.000Z-WTM-2018-09-30T22:17:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
}
]
}
]
},
{
"placeholderHeading": false,
"type": "heading",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"31/1/0",
"31/1/3"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"headingLevel": 2,
"level": 0,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "h-Zwiadowca21-2018-10-01T13:54:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "h-Hymn_Polski_a_Mazurek_Dąbrowskiego-2018-10-01T13:54:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-01T13:54:00.000Z",
"author": "Zwiadowca21",
CommentParser: Output display name if different to username The only normalisation we apply for comparison is lowercasing. Change-Id: Id3d57c2066429fcedc7dcc091e74ed46e17060f1
2022-02-04 18:16:24 +00:00
"displayName": "Zwiadowca",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"33/0",
"33/15/25"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"33/13",
"33/15/25"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"33/15/1",
"33/15/25"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 1,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Zwiadowca21-2018-10-01T13:54:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Zwiadowca21-2018-10-01T13:54:00.000Z-Hymn_Polski_a_Mazurek_Dąbrowskiego",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
},
{
"type": "comment",
"timestamp": "2018-10-01T17:32:00.000Z",
"author": "Piolodiusz",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"34/0",
"34/6/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"34/3",
"34/6/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"34/6/2",
"34/6/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 1,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Piolodiusz-2018-10-01T17:32:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Piolodiusz-2018-10-01T17:32:00.000Z-Hymn_Polski_a_Mazurek_Dąbrowskiego",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-02T16:02:00.000Z",
"author": "Eurohunter",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"36/0/0",
"36/0/4/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"36/0/1",
"36/0/4/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"36/0/4/2",
"36/0/4/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Eurohunter-2018-10-02T16:02:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Eurohunter-2018-10-02T16:02:00.000Z-Piolodiusz-2018-10-01T17:32:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
}
]
}
]
},
{
"placeholderHeading": false,
"type": "heading",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"38/0/0",
"38/0/1"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"headingLevel": 2,
"level": 0,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "h-MOs810-2018-10-03T12:51:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "h-Elx-2018-10-03T12:51:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-03T12:51:00.000Z",
"author": "MOs810",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"40/0",
"40/6/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"40/3",
"40/6/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"40/6/2",
"40/6/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 1,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-MOs810-2018-10-03T12:51:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-MOs810-2018-10-03T12:51:00.000Z-Elx",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-03T13:16:00.000Z",
"author": "Teukros",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"42/0/0",
"42/0/6/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"42/0/3",
"42/0/6/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"42/0/6/2",
"42/0/6/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Teukros-2018-10-03T13:16:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Teukros-2018-10-03T13:16:00.000Z-MOs810-2018-10-03T12:51:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-03T13:59:00.000Z",
"author": "MOs810",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"42/0/7/0/0",
"42/0/7/0/6/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"42/0/7/0/3",
"42/0/7/0/6/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"42/0/7/0/6/2",
"42/0/7/0/6/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 3,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-MOs810-2018-10-03T13:59:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-MOs810-2018-10-03T13:59:00.000Z-Teukros-2018-10-03T13:16:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
}
]
},
{
"type": "comment",
"timestamp": "2018-10-03T17:27:00.000Z",
"author": "Aotearoa",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"42/2/0",
"42/2/4/25"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"42/2/1",
"42/2/4/25"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"42/2/4/1",
"42/2/4/25"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Aotearoa-2018-10-03T17:27:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Aotearoa-2018-10-03T17:27:00.000Z-MOs810-2018-10-03T12:51:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
}
]
}
]
},
{
"placeholderHeading": false,
"type": "heading",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"44/0/0",
"44/0/1"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"headingLevel": 2,
"level": 0,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "h-GregX21-2018-10-04T17:49:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "h-Mozilla_a_Mozilla_Foundation-2018-10-04T17:49:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-04T17:49:00.000Z",
"author": "GregX21",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"46/0",
"48/4/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"48/1",
"48/4/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"48/4/2",
"48/4/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 1,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-GregX21-2018-10-04T17:49:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-GregX21-2018-10-04T17:49:00.000Z-Mozilla_a_Mozilla_Foundation",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-05T15:32:00.000Z",
"author": "Piolodiusz",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"50/0/0",
"50/0/6/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"50/0/3",
"50/0/6/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"50/0/6/2",
"50/0/6/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Piolodiusz-2018-10-05T15:32:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Piolodiusz-2018-10-05T15:32:00.000Z-GregX21-2018-10-04T17:49:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-06T07:05:00.000Z",
"author": "GregX21",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"52/0/0/0/0",
"52/0/0/0/4/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"52/0/0/0/1",
"52/0/0/0/4/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"52/0/0/0/4/2",
"52/0/0/0/4/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 3,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-GregX21-2018-10-06T07:05:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-GregX21-2018-10-06T07:05:00.000Z-Piolodiusz-2018-10-05T15:32:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
}
]
},
{
"type": "comment",
"timestamp": "2018-10-06T12:12:00.000Z",
"author": "Witia",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"54/0/0",
"54/0/12/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"54/0/9",
"54/0/12/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"54/0/12/2",
"54/0/12/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Witia-2018-10-06T12:12:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Witia-2018-10-06T12:12:00.000Z-GregX21-2018-10-04T17:49:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-06T16:18:00.000Z",
"author": "GregX21",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"56/0/0/0/0",
"58/0/0/0/4/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"58/0/0/0/1",
"58/0/0/0/4/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"58/0/0/0/4/2",
"58/0/0/0/4/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 3,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-GregX21-2018-10-06T16:18:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-GregX21-2018-10-06T16:18:00.000Z-Witia-2018-10-06T12:12:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
}
]
}
]
}
]
},
{
"placeholderHeading": false,
"type": "heading",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"60/1/0",
"60/1/1"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"headingLevel": 2,
"level": 0,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "h-MOs810-2018-07-29T10:14:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "h-Autoencyklopedyczność_kościołów_-_propozycja_głosowania-2018-07-29T10:14:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-07-29T10:14:00.000Z",
"author": "MOs810",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"62/0",
"62/14/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"62/11",
"62/14/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"62/14/2",
"62/14/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 1,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-MOs810-2018-07-29T10:14:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-MOs810-2018-07-29T10:14:00.000Z-Autoencyklopedyczność_kościołów_-_propozycja_głosowania",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-07-29T10:31:00.000Z",
"author": "Gdarin",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"64/0/0",
"64/0/6/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"64/0/3",
"64/0/6/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"64/0/6/1",
"64/0/6/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Gdarin-2018-07-29T10:31:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Gdarin-2018-07-29T10:31:00.000Z-MOs810-2018-07-29T10:14:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
},
{
"type": "comment",
"timestamp": "2018-07-29T10:34:00.000Z",
"author": "Teukros",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"64/2/0",
"64/2/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"64/2/1",
"64/2/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"64/2/4/2",
"64/2/4/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Teukros-2018-07-29T10:34:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Teukros-2018-07-29T10:34:00.000Z-MOs810-2018-07-29T10:14:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
},
{
"type": "comment",
"timestamp": "2018-07-29T11:27:00.000Z",
"author": "Kenraiz",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"64/4/0",
"64/4/4/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"64/4/1",
"64/4/4/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"64/4/4/1",
"64/4/4/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Kenraiz-2018-07-29T11:27:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Kenraiz-2018-07-29T11:27:00.000Z-MOs810-2018-07-29T10:14:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-07-29T11:34:00.000Z",
"author": "Gruzin",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"66/0/0/0/0",
"66/0/0/0/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"66/0/0/0/1",
"66/0/0/0/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"66/0/0/0/4/2",
"66/0/0/0/4/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 3,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Gruzin-2018-07-29T11:34:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Gruzin-2018-07-29T11:34:00.000Z-Kenraiz-2018-07-29T11:27:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-07-29T12:55:00.000Z",
"author": "Anagram16",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"66/0/2/0/0/0/0",
"66/0/2/0/0/0/4/28"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"66/0/2/0/0/0/1",
"66/0/2/0/0/0/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"66/0/2/0/0/0/4/2",
"66/0/2/0/0/0/4/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 4,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Anagram16-2018-07-29T12:55:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Anagram16-2018-07-29T12:55:00.000Z-Gruzin-2018-07-29T11:34:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
}
]
}
]
},
{
"type": "comment",
"timestamp": "2018-07-29T11:52:00.000Z",
"author": "Adamt",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"68/0/0",
"68/0/4/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"68/0/1",
"68/0/4/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"68/0/4/1",
"68/0/4/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Adamt-2018-07-29T11:52:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Adamt-2018-07-29T11:52:00.000Z-MOs810-2018-07-29T10:14:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
},
{
"type": "comment",
"timestamp": "2018-07-29T12:12:00.000Z",
"author": "Anagram16",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"68/2/0",
"68/2/4/28"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"68/2/1",
"68/2/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"68/2/4/2",
"68/2/4/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Anagram16-2018-07-29T12:12:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Anagram16-2018-07-29T12:12:00.000Z-MOs810-2018-07-29T10:14:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
},
{
"type": "comment",
"timestamp": "2018-07-29T13:37:00.000Z",
"author": "Hoa binh",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"68/4/0",
"68/4/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"68/4/1",
"68/4/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"68/4/4/2",
"68/4/4/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Hoa_binh-2018-07-29T13:37:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Hoa_binh-2018-07-29T13:37:00.000Z-MOs810-2018-07-29T10:14:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-07-29T13:45:00.000Z",
"author": "Anagram16",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"70/0/0/0/0",
"70/0/0/0/4/28"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"70/0/0/0/1",
"70/0/0/0/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"70/0/0/0/4/2",
"70/0/0/0/4/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 3,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Anagram16-2018-07-29T13:45:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Anagram16-2018-07-29T13:45:00.000Z-Hoa_binh-2018-07-29T13:37:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
},
{
"type": "comment",
"timestamp": "2018-07-29T16:11:00.000Z",
"author": "Gruzin",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"70/0/0/2/0",
"70/0/0/2/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"70/0/0/2/1",
"70/0/0/2/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"70/0/0/2/4/2",
"70/0/0/2/4/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 3,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Gruzin-2018-07-29T16:11:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Gruzin-2018-07-29T16:11:00.000Z-Hoa_binh-2018-07-29T13:37:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
}
]
},
{
"type": "comment",
"timestamp": "2018-07-30T17:54:00.000Z",
"author": "Zetpe0202",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"72/0/0",
"72/0/20/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"72/0/17",
"72/0/20/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"72/0/20/2",
"72/0/20/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Zetpe0202-2018-07-30T17:54:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Zetpe0202-2018-07-30T17:54:00.000Z-MOs810-2018-07-29T10:14:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-07-30T18:47:00.000Z",
"author": "Gruzin",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"74/0/0/0/0",
"74/0/0/0/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"74/0/0/0/1",
"74/0/0/0/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"74/0/0/0/4/2",
"74/0/0/0/4/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 3,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Gruzin-2018-07-30T18:47:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Gruzin-2018-07-30T18:47:00.000Z-Zetpe0202-2018-07-30T17:54:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
}
]
},
{
"type": "comment",
"timestamp": "2018-07-30T19:08:00.000Z",
"author": "John Belushi",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"76/0/0",
"76/0/4/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"76/0/1",
"76/0/4/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"76/0/4/1",
"76/0/4/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-John_Belushi-2018-07-30T19:08:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-John_Belushi-2018-07-30T19:08:00.000Z-MOs810-2018-07-29T10:14:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
},
{
"type": "comment",
"timestamp": "2018-07-30T19:30:00.000Z",
"author": "Gruzin",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"76/2/0",
"76/2/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"76/2/1",
"76/2/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"76/2/4/2",
"76/2/4/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Gruzin-2018-07-30T19:30:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Gruzin-2018-07-30T19:30:00.000Z-MOs810-2018-07-29T10:14:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
},
{
"type": "comment",
"timestamp": "2018-07-31T08:26:00.000Z",
"author": "MOs810",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"76/4/0",
"76/4/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"76/4/1",
"76/4/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"76/4/4/2",
"76/4/4/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-MOs810-2018-07-31T08:26:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-MOs810-2018-07-31T08:26:00.000Z-MOs810-2018-07-29T10:14:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-07-31T08:36:00.000Z",
"author": "John Belushi",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"78/0/0/0/0",
"78/0/0/0/4/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"78/0/0/0/1",
"78/0/0/0/4/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"78/0/0/0/4/1",
"78/0/0/0/4/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 3,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-John_Belushi-2018-07-31T08:36:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-John_Belushi-2018-07-31T08:36:00.000Z-MOs810-2018-07-31T08:26:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
},
{
"type": "comment",
"timestamp": "2018-07-31T14:56:00.000Z",
"author": "Gruzin",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"78/0/0/2/0",
"78/0/0/2/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"78/0/0/2/1",
"78/0/0/2/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"78/0/0/2/4/2",
"78/0/0/2/4/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 3,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Gruzin-2018-07-31T14:56:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Gruzin-2018-07-31T14:56:00.000Z-MOs810-2018-07-31T08:26:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
},
{
"type": "comment",
"timestamp": "2018-07-31T15:21:00.000Z",
"author": "Teukros",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"78/0/0/4/0",
"78/0/0/4/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"78/0/0/4/1",
"78/0/0/4/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"78/0/0/4/4/2",
"78/0/0/4/4/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 3,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Teukros-2018-07-31T15:21:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Teukros-2018-07-31T15:21:00.000Z-MOs810-2018-07-31T08:26:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
},
{
"type": "comment",
"timestamp": "2018-07-31T16:44:00.000Z",
"author": "Malarz pl",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"78/0/0/6/0",
"78/0/0/6/4/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"78/0/0/6/1",
"78/0/0/6/4/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"78/0/0/6/4/1",
"78/0/0/6/4/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 3,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Malarz_pl-2018-07-31T16:44:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Malarz_pl-2018-07-31T16:44:00.000Z-MOs810-2018-07-31T08:26:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-07-31T17:50:00.000Z",
"author": "Zala",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"78/0/0/6/5/0/0",
"78/0/0/6/5/0/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"78/0/0/6/5/0/1",
"78/0/0/6/5/0/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"78/0/0/6/5/0/4/2",
"78/0/0/6/5/0/4/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 4,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Zala-2018-07-31T17:50:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Zala-2018-07-31T17:50:00.000Z-Malarz_pl-2018-07-31T16:44:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
}
]
},
{
"type": "comment",
"timestamp": "2018-07-31T22:05:00.000Z",
"author": "Adamt",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"78/0/0/8/0",
"78/0/0/8/4/25"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"78/0/0/8/1",
"78/0/0/8/4/25"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"78/0/0/8/4/1",
"78/0/0/8/4/25"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 3,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Adamt-2018-07-31T22:05:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Adamt-2018-07-31T22:05:00.000Z-MOs810-2018-07-31T08:26:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-08-01T18:20:00.000Z",
"author": "John Belushi",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"78/0/2/0/0/0/0",
"78/0/2/0/0/0/10/25"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"78/0/2/0/0/0/7",
"78/0/2/0/0/0/10/25"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"78/0/2/0/0/0/10/1",
"78/0/2/0/0/0/10/25"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 4,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-John_Belushi-2018-08-01T18:20:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-John_Belushi-2018-08-01T18:20:00.000Z-Adamt-2018-07-31T22:05:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
}
]
}
]
},
{
"type": "comment",
"timestamp": "2018-08-06T14:08:00.000Z",
"author": "KamilK7",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"80/0/0",
"80/0/4/25"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"80/0/1",
"80/0/4/25"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"80/0/4/1",
"80/0/4/25"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-KamilK7-2018-08-06T14:08:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-KamilK7-2018-08-06T14:08:00.000Z-MOs810-2018-07-29T10:14:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
}
]
},
{
"type": "comment",
"timestamp": "2018-08-13T22:43:00.000Z",
"author": "Cz ja",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"82/0",
"82/8/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"82/1",
"82/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"82/5",
"82/8/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"82/4/2",
"82/4/27"
],
[
"82/8/2",
"82/8/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 1,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Cz_ja-2018-08-13T22:43:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Cz_ja-2018-08-13T22:43:00.000Z-Autoencyklopedyczność_kościołów_-_propozycja_głosowania",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-08-14T07:07:00.000Z",
"author": "Malarz pl",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"84/0/0",
"84/0/6/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"84/0/3",
"84/0/6/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"84/0/6/1",
"84/0/6/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Malarz_pl-2018-08-14T07:07:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Malarz_pl-2018-08-14T07:07:00.000Z-Cz_ja-2018-08-13T22:43:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
}
]
},
{
"type": "comment",
"timestamp": "2018-08-31T15:59:00.000Z",
"author": "BasileusAutokratorPL",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"86/0",
"86/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"86/1",
"86/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"86/4/2",
"86/4/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 1,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-BasileusAutokratorPL-2018-08-31T15:59:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-BasileusAutokratorPL-2018-08-31T15:59:00.000Z-Autoencyklopedyczność_kościołów_-_propozycja_głosowania",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-09-08T16:10:00.000Z",
"author": "Ciacho5",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"88/0/0/0",
"88/0/7/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"88/0/4",
"88/0/7/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"88/0/7/2",
"88/0/7/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Ciacho5-2018-09-08T16:10:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Ciacho5-2018-09-08T16:10:00.000Z-BasileusAutokratorPL-2018-08-31T15:59:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
}
]
},
{
"type": "comment",
"timestamp": "2018-09-09T15:15:00.000Z",
"author": "BasileusAutokratorPL",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"90/0",
"90/4/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"90/1",
"90/4/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"90/4/2",
"90/4/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 1,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-BasileusAutokratorPL-2018-09-09T15:15:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-BasileusAutokratorPL-2018-09-09T15:15:00.000Z-Autoencyklopedyczność_kościołów_-_propozycja_głosowania",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-08T09:34:00.000Z",
"author": "Alan ffm",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"92/0/0",
"92/0/4/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"92/0/1",
"92/0/4/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"92/0/4/2",
"92/0/4/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Alan_ffm-2018-10-08T09:34:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Alan_ffm-2018-10-08T09:34:00.000Z-BasileusAutokratorPL-2018-09-09T15:15:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
}
]
}
]
},
{
"placeholderHeading": false,
"type": "heading",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"94/1/0",
"94/1/1"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"headingLevel": 2,
"level": 0,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "h-Wostr-2018-10-09T20:26:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "h-Zbrodnie_oddziałów_Ludowej_Straży_Bezpieczeństwa-2018-10-09T20:26:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-09T20:26:00.000Z",
"author": "Wostr",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"96/0",
"96/8/25"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"96/5",
"96/8/25"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"96/8/1",
"96/8/25"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 1,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Wostr-2018-10-09T20:26:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Wostr-2018-10-09T20:26:00.000Z-Zbrodnie_oddziałów_Ludowej_Straży_Bezpieczeństwa",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-11T12:15:00.000Z",
"author": "Kriis bis",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"98/0/0",
"98/0/14/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"98/0/11",
"98/0/14/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"98/0/14/2",
"98/0/14/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Kriis_bis-2018-10-11T12:15:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Kriis_bis-2018-10-11T12:15:00.000Z-Wostr-2018-10-09T20:26:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
}
]
}
]
},
{
"placeholderHeading": false,
"type": "heading",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"100/0/0",
"100/0/2"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"headingLevel": 2,
"level": 0,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "h-Ciacho5-2018-10-10T19:51:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "h-Sekcja_Literatura-2018-10-10T19:51:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-10T19:51:00.000Z",
"author": "Ciacho5",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"102/0",
"102/8/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"102/5",
"102/8/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"102/8/2",
"102/8/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 1,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Ciacho5-2018-10-10T19:51:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Ciacho5-2018-10-10T19:51:00.000Z-Sekcja_Literatura",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-10T21:52:00.000Z",
"author": "Piotr967",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"104/0/0",
"104/0/4/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"104/0/1",
"104/0/4/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"104/0/4/1",
"104/0/4/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Piotr967-2018-10-10T21:52:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Piotr967-2018-10-10T21:52:00.000Z-Ciacho5-2018-10-10T19:51:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
},
{
"type": "comment",
"timestamp": "2018-10-11T08:12:00.000Z",
"author": "Ciacho5",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"106/0/0",
"106/0/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"106/0/1",
"106/0/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"106/0/4/2",
"106/0/4/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Ciacho5-2018-10-11T08:12:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Ciacho5-2018-10-11T08:12:00.000Z-Ciacho5-2018-10-10T19:51:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
},
{
"type": "comment",
"timestamp": "2018-10-11T12:44:00.000Z",
"author": "Henryk Tannhäuser",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"106/2/0",
"106/2/8/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"106/2/5",
"106/2/8/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"106/2/8/2",
"106/2/8/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Henryk_Tannhäuser-2018-10-11T12:44:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Henryk_Tannhäuser-2018-10-11T12:44:00.000Z-Ciacho5-2018-10-10T19:51:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
}
]
}
]
},
{
"placeholderHeading": false,
"type": "heading",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"108/0/0",
"108/0/1"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"headingLevel": 2,
"level": 0,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "h-WTM-2018-10-11T20:46:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "h-Jerzy_Mrozowicki-2018-10-11T20:46:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-11T20:46:00.000Z",
"author": "WTM",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"110/0",
"110/6/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"110/3",
"110/6/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"110/6/2",
"110/6/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 1,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-WTM-2018-10-11T20:46:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-WTM-2018-10-11T20:46:00.000Z-Jerzy_Mrozowicki",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-12T07:11:00.000Z",
"author": "Kriis bis",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"112/0/0",
"112/0/8/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"112/0/5",
"112/0/8/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"112/0/8/2",
"112/0/8/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Kriis_bis-2018-10-12T07:11:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Kriis_bis-2018-10-12T07:11:00.000Z-WTM-2018-10-11T20:46:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
}
]
}
]
},
{
"placeholderHeading": false,
"type": "heading",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"114/1/0",
"114/1/1"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"headingLevel": 2,
"level": 0,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "h-Jacek555-2018-10-06T12:50:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "h-„Cyklon_B”_czy_„cyklon_B”?-2018-10-06T12:50:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-06T12:50:00.000Z",
"author": "Jacek555",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"116/0",
"116/6/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"116/3",
"116/6/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"116/6/2",
"116/6/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 1,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Jacek555-2018-10-06T12:50:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Jacek555-2018-10-06T12:50:00.000Z-„Cyklon_B”_czy_„cyklon_B”?",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-06T16:35:00.000Z",
"author": "Vinsinna",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"118/0/0",
"118/0/4/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"118/0/1",
"118/0/4/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"118/0/4/2",
"118/0/4/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Vinsinna-2018-10-06T16:35:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Vinsinna-2018-10-06T16:35:00.000Z-Jacek555-2018-10-06T12:50:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
},
{
"type": "comment",
"timestamp": "2018-10-06T17:18:00.000Z",
"author": "Wostr",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"120/0/0",
"120/0/8/25"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"120/0/5",
"120/0/8/25"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"120/0/8/1",
"120/0/8/25"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Wostr-2018-10-06T17:18:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Wostr-2018-10-06T17:18:00.000Z-Jacek555-2018-10-06T12:50:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
},
{
"type": "comment",
"timestamp": "2018-10-06T17:36:00.000Z",
"author": "Matrek",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"120/2/0",
"120/2/4/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"120/2/1",
"120/2/4/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"120/2/4/2",
"120/2/4/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Matrek-2018-10-06T17:36:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Matrek-2018-10-06T17:36:00.000Z-Jacek555-2018-10-06T12:50:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
},
{
"type": "comment",
"timestamp": "2018-10-07T08:20:00.000Z",
"author": "Dreamcatcher25",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"120/4/0",
"120/4/6/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"120/4/3",
"120/4/6/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"120/4/6/2",
"120/4/6/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Dreamcatcher25-2018-10-07T08:20:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Dreamcatcher25-2018-10-07T08:20:00.000Z-Jacek555-2018-10-06T12:50:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-07T16:04:00.000Z",
"author": "Matrek",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"120/4/7/0/0/0/0",
"120/4/7/0/0/0/6/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"120/4/7/0/0/0/3",
"120/4/7/0/0/0/6/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"120/4/7/0/0/0/6/2",
"120/4/7/0/0/0/6/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 4,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Matrek-2018-10-07T16:04:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Matrek-2018-10-07T16:04:00.000Z-Dreamcatcher25-2018-10-07T08:20:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [
"Comment skips indentation level"
],
"replies": []
},
{
"type": "comment",
"timestamp": "2018-10-07T11:43:00.000Z",
"author": "Wostr",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"120/4/7/2/0",
"120/4/7/2/4/25"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"120/4/7/2/1",
"120/4/7/2/4/25"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"120/4/7/2/4/1",
"120/4/7/2/4/25"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 3,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Wostr-2018-10-07T11:43:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Wostr-2018-10-07T11:43:00.000Z-Dreamcatcher25-2018-10-07T08:20:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
}
]
},
{
"type": "comment",
"timestamp": "2018-10-07T16:27:00.000Z",
"author": "Wostr",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"120/6/0",
"120/6/8/25"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"120/6/5",
"120/6/8/25"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"120/6/8/1",
"120/6/8/25"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Wostr-2018-10-07T16:27:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Wostr-2018-10-07T16:27:00.000Z-Jacek555-2018-10-06T12:50:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-07T16:34:00.000Z",
"author": "Matrek",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"120/6/9/0/0",
"120/6/9/0/8/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"120/6/9/0/1",
"120/6/9/0/4/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"120/6/9/0/5",
"120/6/9/0/8/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"120/6/9/0/4/2",
"120/6/9/0/4/26"
],
[
"120/6/9/0/8/2",
"120/6/9/0/8/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 3,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Matrek-2018-10-07T16:34:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Matrek-2018-10-07T16:34:00.000Z-Wostr-2018-10-07T16:27:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-07T16:53:00.000Z",
"author": "Wostr",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"120/6/9/0/9/0/0",
"120/6/9/0/9/0/8/25"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"120/6/9/0/9/0/5",
"120/6/9/0/9/0/8/25"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"120/6/9/0/9/0/8/1",
"120/6/9/0/9/0/8/25"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 4,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Wostr-2018-10-07T16:53:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Wostr-2018-10-07T16:53:00.000Z-Matrek-2018-10-07T16:34:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-07T18:49:00.000Z",
"author": "Vinsinna",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"120/6/9/0/9/0/9/0/0",
"120/6/9/0/9/0/9/0/6/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"120/6/9/0/9/0/9/0/3",
"120/6/9/0/9/0/9/0/6/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"120/6/9/0/9/0/9/0/6/2",
"120/6/9/0/9/0/9/0/6/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 5,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Vinsinna-2018-10-07T18:49:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Vinsinna-2018-10-07T18:49:00.000Z-Wostr-2018-10-07T16:53:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
}
]
}
]
},
{
"type": "comment",
"timestamp": "2018-10-08T16:33:00.000Z",
"author": "Jacek555",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"120/6/9/2/0",
"120/6/9/2/6/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"120/6/9/2/3",
"120/6/9/2/6/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"120/6/9/2/6/2",
"120/6/9/2/6/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 3,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Jacek555-2018-10-08T16:33:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Jacek555-2018-10-08T16:33:00.000Z-Wostr-2018-10-07T16:27:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
}
]
},
{
"type": "comment",
"timestamp": "2018-10-07T16:32:00.000Z",
"author": "Gruzin",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"120/8/0",
"120/8/4/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"120/8/1",
"120/8/4/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"120/8/4/2",
"120/8/4/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Gruzin-2018-10-07T16:32:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Gruzin-2018-10-07T16:32:00.000Z-Jacek555-2018-10-06T12:50:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
},
{
"type": "comment",
"timestamp": "2018-10-07T18:00:00.000Z",
"author": "Lesław Zimny",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"120/10/0",
"120/10/6/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"120/10/3",
"120/10/6/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"120/10/6/2",
"120/10/6/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Lesław_Zimny-2018-10-07T18:00:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Lesław_Zimny-2018-10-07T18:00:00.000Z-Jacek555-2018-10-06T12:50:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-08T08:01:00.000Z",
"author": "Alan ffm",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"122/0/0/0/0",
"122/0/0/0/6/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"122/0/0/0/3",
"122/0/0/0/6/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"122/0/0/0/6/2",
"122/0/0/0/6/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 3,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Alan_ffm-2018-10-08T08:01:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Alan_ffm-2018-10-08T08:01:00.000Z-Lesław_Zimny-2018-10-07T18:00:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-08T16:33:00.000Z",
"author": "Jacek555",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"122/0/2/0/0/0/0",
"122/0/2/0/0/0/4/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"122/0/2/0/0/0/1",
"122/0/2/0/0/0/4/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"122/0/2/0/0/0/4/2",
"122/0/2/0/0/0/4/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 4,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Jacek555-2018-10-08T16:33:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Jacek555-2018-10-08T16:33:00.000Z-Alan_ffm-2018-10-08T08:01:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
}
]
}
]
},
{
"type": "comment",
"timestamp": "2018-10-08T16:33:00.000Z",
"author": "Jacek555",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"124/0/0",
"124/0/8/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"124/0/5",
"124/0/8/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"124/0/8/2",
"124/0/8/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Jacek555-2018-10-08T16:33:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Jacek555-2018-10-08T16:33:00.000Z-Jacek555-2018-10-06T12:50:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
},
{
"type": "comment",
"timestamp": "2018-10-16T17:40:00.000Z",
"author": "Wostr",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"124/2/0",
"124/2/8/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"124/2/5",
"124/2/8/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"124/2/8/1",
"124/2/8/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Wostr-2018-10-16T17:40:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Wostr-2018-10-16T17:40:00.000Z-Jacek555-2018-10-06T12:50:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
},
{
"type": "comment",
"timestamp": "2018-10-16T17:53:00.000Z",
"author": "Jacek555",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"126/0/0",
"126/0/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"126/0/1",
"126/0/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"126/0/4/2",
"126/0/4/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Jacek555-2018-10-16T17:53:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Jacek555-2018-10-16T17:53:00.000Z-Jacek555-2018-10-06T12:50:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
}
]
}
]
},
{
"placeholderHeading": false,
"type": "heading",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"128/1/0",
"128/1/1"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"headingLevel": 2,
"level": 0,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "h-Pawelmhm-2018-10-16T10:55:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "h-Przypisy_do_ebooków._Co_z_numerami_stron?-2018-10-16T10:55:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-16T10:55:00.000Z",
"author": "Pawelmhm",
CommentParser: Output display name if different to username The only normalisation we apply for comparison is lowercasing. Change-Id: Id3d57c2066429fcedc7dcc091e74ed46e17060f1
2022-02-04 18:16:24 +00:00
"displayName": "pawelmxm",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"130/0",
"130/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"130/1",
"130/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"130/4/2",
"130/4/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 1,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Pawelmhm-2018-10-16T10:55:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Pawelmhm-2018-10-16T10:55:00.000Z-Przypisy_do_ebooków._Co_z_numerami_stron?",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-16T11:30:00.000Z",
"author": "Ironupiwada",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"132/0/0",
"132/0/6/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"132/0/3",
"132/0/6/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"132/0/6/2",
"132/0/6/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Ironupiwada-2018-10-16T11:30:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Ironupiwada-2018-10-16T11:30:00.000Z-Pawelmhm-2018-10-16T10:55:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
},
{
"type": "comment",
"timestamp": "2018-10-16T12:01:00.000Z",
"author": "Pawelmhm",
CommentParser: Output display name if different to username The only normalisation we apply for comparison is lowercasing. Change-Id: Id3d57c2066429fcedc7dcc091e74ed46e17060f1
2022-02-04 18:16:24 +00:00
"displayName": "pawelmxm",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"132/2/0",
"132/2/6/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"132/2/3",
"132/2/6/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"132/2/6/2",
"132/2/6/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Pawelmhm-2018-10-16T12:01:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Pawelmhm-2018-10-16T12:01:00.000Z-Pawelmhm-2018-10-16T10:55:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
},
{
"type": "comment",
"timestamp": "2018-10-16T12:14:00.000Z",
"author": "Tomasz Raburski",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"132/4/0",
"132/4/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"132/4/1",
"132/4/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"132/4/4/2",
"132/4/4/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Tomasz_Raburski-2018-10-16T12:14:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Tomasz_Raburski-2018-10-16T12:14:00.000Z-Pawelmhm-2018-10-16T10:55:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
},
{
"type": "comment",
"timestamp": "2018-10-16T13:48:00.000Z",
"author": "Ciacho5",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"132/6/0",
"132/6/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"132/6/1",
"132/6/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"132/6/4/2",
"132/6/4/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Ciacho5-2018-10-16T13:48:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Ciacho5-2018-10-16T13:48:00.000Z-Pawelmhm-2018-10-16T10:55:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-16T14:57:00.000Z",
"author": "Ironupiwada",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"132/6/5/0/0",
"132/6/5/0/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"132/6/5/0/1",
"132/6/5/0/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"132/6/5/0/4/2",
"132/6/5/0/4/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 3,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Ironupiwada-2018-10-16T14:57:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Ironupiwada-2018-10-16T14:57:00.000Z-Ciacho5-2018-10-16T13:48:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-17T06:56:00.000Z",
"author": "Ciacho5",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"132/6/5/0/5/0/0",
"132/6/5/0/5/0/6/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"132/6/5/0/5/0/3",
"132/6/5/0/5/0/6/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"132/6/5/0/5/0/6/2",
"132/6/5/0/5/0/6/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 4,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Ciacho5-2018-10-17T06:56:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Ciacho5-2018-10-17T06:56:00.000Z-Ironupiwada-2018-10-16T14:57:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
}
]
}
]
}
]
}
]
},
{
"placeholderHeading": false,
"type": "heading",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"134/0/0",
"134/0/1"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"headingLevel": 2,
"level": 0,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "h-Gżdacz-2018-10-17T06:39:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "h-Nazewnictwo_planetoid-2018-10-17T06:39:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-17T06:39:00.000Z",
"author": "Gżdacz",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"136/0",
"136/8/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"136/5",
"136/8/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"136/8/2",
"136/8/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 1,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Gżdacz-2018-10-17T06:39:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Gżdacz-2018-10-17T06:39:00.000Z-Nazewnictwo_planetoid",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-17T07:16:00.000Z",
"author": "Michał Rosa",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"138/0/0",
"138/0/6/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"138/0/3",
"138/0/6/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"138/0/6/2",
"138/0/6/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Michał_Rosa-2018-10-17T07:16:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Michał_Rosa-2018-10-17T07:16:00.000Z-Gżdacz-2018-10-17T06:39:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
}
]
}
]
},
{
"placeholderHeading": false,
"type": "heading",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"140/1/0",
"140/1/1"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"headingLevel": 2,
"level": 0,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "h-Piotr967-2018-10-11T17:25:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "h-Czy_ten_szablon_powinien_istnieć?-2018-10-11T17:25:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-11T17:25:00.000Z",
"author": "Piotr967",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"142/0/0",
"142/9/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"142/6",
"142/9/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"142/9/1",
"142/9/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 1,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Piotr967-2018-10-11T17:25:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Piotr967-2018-10-11T17:25:00.000Z-Czy_ten_szablon_powinien_istnieć?",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-12T11:17:00.000Z",
"author": "Bobek321",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"144/0/0",
"144/0/6/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"144/0/3",
"144/0/6/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"144/0/6/2",
"144/0/6/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Bobek321-2018-10-12T11:17:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Bobek321-2018-10-12T11:17:00.000Z-Piotr967-2018-10-11T17:25:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-12T18:22:00.000Z",
"author": "Piotr967",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"146/0/0/0/0",
"146/0/0/0/8/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"146/0/0/0/5",
"146/0/0/0/8/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"146/0/0/0/8/1",
"146/0/0/0/8/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 3,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Piotr967-2018-10-12T18:22:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Piotr967-2018-10-12T18:22:00.000Z-Bobek321-2018-10-12T11:17:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-12T18:51:00.000Z",
"author": "Kenraiz",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"146/0/0/0/9/0/0",
"146/0/0/0/9/0/6/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"146/0/0/0/9/0/3",
"146/0/0/0/9/0/6/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"146/0/0/0/9/0/6/1",
"146/0/0/0/9/0/6/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 4,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Kenraiz-2018-10-12T18:51:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Kenraiz-2018-10-12T18:51:00.000Z-Piotr967-2018-10-12T18:22:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
}
]
}
]
},
{
"type": "comment",
"timestamp": "2018-10-15T22:47:00.000Z",
"author": "Ented",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"148/0/0",
"148/0/6/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"148/0/3",
"148/0/6/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"148/0/6/2",
"148/0/6/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Ented-2018-10-15T22:47:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Ented-2018-10-15T22:47:00.000Z-Piotr967-2018-10-11T17:25:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
},
{
"type": "comment",
"timestamp": "2018-10-17T19:08:00.000Z",
"author": "Bobek321",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"148/2/0",
"148/2/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"148/2/1",
"148/2/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"148/2/4/2",
"148/2/4/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Bobek321-2018-10-17T19:08:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Bobek321-2018-10-17T19:08:00.000Z-Piotr967-2018-10-11T17:25:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-17T19:25:00.000Z",
"author": "Piotr967",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"150/0/0/0/0",
"150/0/0/0/4/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"150/0/0/0/1",
"150/0/0/0/4/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"150/0/0/0/4/1",
"150/0/0/0/4/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 3,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Piotr967-2018-10-17T19:25:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Piotr967-2018-10-17T19:25:00.000Z-Bobek321-2018-10-17T19:08:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
}
]
},
{
"type": "comment",
"timestamp": "2018-10-17T19:49:00.000Z",
"author": "Malarz pl",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"150/2/0",
"150/2/6/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"150/2/3",
"150/2/6/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"150/2/6/1",
"150/2/6/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Malarz_pl-2018-10-17T19:49:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Malarz_pl-2018-10-17T19:49:00.000Z-Piotr967-2018-10-11T17:25:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
}
]
}
]
},
{
"placeholderHeading": false,
"type": "heading",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"152/1/0",
"152/1/1"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"headingLevel": 2,
"level": 0,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "h-Pawelmhm-2018-10-18T10:29:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "h-Pytanie_dotyczące_artykułów_o_czasopismach._Czy_ta_sama_nazwa_czasopisma_wyda-2018-10-18T10:29:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-18T10:29:00.000Z",
"author": "Pawelmhm",
CommentParser: Output display name if different to username The only normalisation we apply for comparison is lowercasing. Change-Id: Id3d57c2066429fcedc7dcc091e74ed46e17060f1
2022-02-04 18:16:24 +00:00
"displayName": "pawelmxm",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"154/0",
"154/14/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"154/11",
"154/14/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"154/14/2",
"154/14/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 1,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Pawelmhm-2018-10-18T10:29:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Pawelmhm-2018-10-18T10:29:00.000Z-Pytanie_dotyczące_artykułów_o_czasopismach._Czy_ta_sama_nazwa_czasopisma_wyda",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-18T11:05:00.000Z",
"author": "Tomasz Raburski",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"156/0/0",
"156/0/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"156/0/1",
"156/0/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"156/0/4/2",
"156/0/4/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Tomasz_Raburski-2018-10-18T11:05:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Tomasz_Raburski-2018-10-18T11:05:00.000Z-Pawelmhm-2018-10-18T10:29:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
},
{
"type": "comment",
"timestamp": "2018-10-18T12:04:00.000Z",
"author": "Ciacho5",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"156/2/0",
"156/2/6/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"156/2/3",
"156/2/6/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"156/2/6/2",
"156/2/6/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Ciacho5-2018-10-18T12:04:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Ciacho5-2018-10-18T12:04:00.000Z-Pawelmhm-2018-10-18T10:29:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
}
]
}
]
},
{
"placeholderHeading": false,
"type": "heading",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"158/1/0",
"158/1/1"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"headingLevel": 2,
"level": 0,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "h-Demkoff-2018-10-19T01:41:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "h-Tło_w_komórce-2018-10-19T01:41:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-19T01:41:00.000Z",
"author": "Demkoff",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"160/0/0/0",
"164/0/10/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"164/0/7",
"164/0/10/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"164/0/10/1",
"164/0/10/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Demkoff-2018-10-19T01:41:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Demkoff-2018-10-19T01:41:00.000Z-Tło_w_komórce",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [
"Comment skips indentation level"
],
"replies": []
}
]
},
{
"placeholderHeading": false,
"type": "heading",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"166/1/0",
"166/1/1"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"headingLevel": 2,
"level": 0,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "h-Paweł_Ziemian-2018-10-02T20:26:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "h-Ułamki-2018-10-02T20:26:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-02T20:26:00.000Z",
"author": "Paweł Ziemian",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"168/0",
"168/18/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"168/15",
"168/18/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"168/18/2",
"168/18/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 1,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Paweł_Ziemian-2018-10-02T20:26:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Paweł_Ziemian-2018-10-02T20:26:00.000Z-Ułamki",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-02T22:16:00.000Z",
"author": "Matma Rex",
CommentParser: Output display name if different to username The only normalisation we apply for comparison is lowercasing. Change-Id: Id3d57c2066429fcedc7dcc091e74ed46e17060f1
2022-02-04 18:16:24 +00:00
"displayName": "Matma Rex",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"170/0/0",
"170/0/6/25"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"170/0/3",
"170/0/6/25"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"170/0/6/1",
"170/0/6/25"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Matma_Rex-2018-10-02T22:16:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Matma_Rex-2018-10-02T22:16:00.000Z-Paweł_Ziemian-2018-10-02T20:26:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
}
]
},
{
"type": "comment",
"timestamp": "2018-10-19T12:09:00.000Z",
"author": "Ashaio",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"172/0",
"172/8/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"172/5",
"172/8/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"172/8/2",
"172/8/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 1,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Ashaio-2018-10-19T12:09:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Ashaio-2018-10-19T12:09:00.000Z-Ułamki",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-19T20:23:00.000Z",
"author": "Paweł Ziemian",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"174/0/0",
"174/0/6/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"174/0/3",
"174/0/6/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"174/0/6/2",
"174/0/6/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Paweł_Ziemian-2018-10-19T20:23:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Paweł_Ziemian-2018-10-19T20:23:00.000Z-Ashaio-2018-10-19T12:09:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
}
]
}
]
},
{
"placeholderHeading": false,
"type": "heading",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"176/1/0",
"176/1/1"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"headingLevel": 2,
"level": 0,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "h-Vinsinna-2018-10-18T13:45:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "h-\"To_źródło_nie_zostało_użyte_do_napisania_tych_treści\"-2018-10-18T13:45:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-18T13:45:00.000Z",
"author": "Vinsinna",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"178/0",
"178/10/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"178/7",
"178/10/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"178/10/2",
"178/10/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 1,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Vinsinna-2018-10-18T13:45:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Vinsinna-2018-10-18T13:45:00.000Z-\"To_źródło_nie_zostało_użyte_do_napisania_tych_treści\"",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-18T15:07:00.000Z",
"author": "Kenraiz",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"180/0/0",
"180/0/4/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"signatureRanges": [
[
"180/0/1",
"180/0/4/26"
]
],
"timestampRanges": [
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
[
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"180/0/4/1",
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"180/0/4/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Kenraiz-2018-10-18T15:07:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Kenraiz-2018-10-18T15:07:00.000Z-Vinsinna-2018-10-18T13:45:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-18T17:12:00.000Z",
"author": "Vinsinna",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"180/0/5/0/0",
"180/0/5/0/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"180/0/5/0/1",
"180/0/5/0/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"180/0/5/0/4/2",
"180/0/5/0/4/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 3,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Vinsinna-2018-10-18T17:12:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Vinsinna-2018-10-18T17:12:00.000Z-Kenraiz-2018-10-18T15:07:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
}
]
},
{
"type": "comment",
"timestamp": "2018-10-18T15:24:00.000Z",
"author": "WTM",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"180/2/0",
"180/2/6/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"180/2/3",
"180/2/6/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"180/2/6/2",
"180/2/6/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-WTM-2018-10-18T15:24:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-WTM-2018-10-18T15:24:00.000Z-Vinsinna-2018-10-18T13:45:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-18T17:11:00.000Z",
"author": "Vinsinna",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"180/2/7/0/0",
"180/2/7/0/6/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"180/2/7/0/3",
"180/2/7/0/6/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"180/2/7/0/6/2",
"180/2/7/0/6/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 3,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Vinsinna-2018-10-18T17:11:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Vinsinna-2018-10-18T17:11:00.000Z-WTM-2018-10-18T15:24:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
}
]
},
{
"type": "comment",
"timestamp": "2018-10-18T18:39:00.000Z",
"author": "Ented",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"182/0/0",
"182/0/13/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"182/0/10",
"182/0/13/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"182/0/13/2",
"182/0/13/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Ented-2018-10-18T18:39:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Ented-2018-10-18T18:39:00.000Z-Vinsinna-2018-10-18T13:45:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
},
{
"type": "comment",
"timestamp": "2018-10-18T18:45:00.000Z",
"author": "Kenraiz",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"182/2/0",
"182/2/11/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"182/2/8",
"182/2/11/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"182/2/11/1",
"182/2/11/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Kenraiz-2018-10-18T18:45:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Kenraiz-2018-10-18T18:45:00.000Z-Vinsinna-2018-10-18T13:45:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-19T19:10:00.000Z",
"author": "Vinsinna",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"182/2/12/0/0",
"182/2/12/0/12/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"182/2/12/0/9",
"182/2/12/0/12/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"182/2/12/0/12/2",
"182/2/12/0/12/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 3,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Vinsinna-2018-10-19T19:10:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Vinsinna-2018-10-19T19:10:00.000Z-Kenraiz-2018-10-18T18:45:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-19T19:21:00.000Z",
"author": "Kenraiz",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"182/2/12/0/13/0/0",
"182/2/12/0/13/0/4/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"182/2/12/0/13/0/1",
"182/2/12/0/13/0/4/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"182/2/12/0/13/0/4/1",
"182/2/12/0/13/0/4/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 4,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Kenraiz-2018-10-19T19:21:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Kenraiz-2018-10-19T19:21:00.000Z-Vinsinna-2018-10-19T19:10:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
},
{
"type": "comment",
"timestamp": "2018-10-21T12:40:00.000Z",
"author": "Vinsinna",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"182/2/12/0/13/2/0",
"182/2/12/0/13/2/6/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"182/2/12/0/13/2/3",
"182/2/12/0/13/2/6/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"182/2/12/0/13/2/6/2",
"182/2/12/0/13/2/6/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 4,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Vinsinna-2018-10-21T12:40:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Vinsinna-2018-10-21T12:40:00.000Z-Vinsinna-2018-10-19T19:10:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
}
]
}
]
}
]
}
]
},
{
"placeholderHeading": false,
"type": "heading",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"184/1/0",
"184/1/1"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"headingLevel": 2,
"level": 0,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "h-Czyz1-2018-10-21T12:45:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "h-Tytuł_filmu-2018-10-21T12:45:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-21T12:45:00.000Z",
"author": "Czyz1",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"186/0",
"186/12/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"186/9",
"186/12/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"186/12/2",
"186/12/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 1,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Czyz1-2018-10-21T12:45:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Czyz1-2018-10-21T12:45:00.000Z-Tytuł_filmu",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-21T17:12:00.000Z",
"author": "Teukros",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"188/0/0",
"188/0/6/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"188/0/3",
"188/0/6/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"188/0/6/2",
"188/0/6/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Teukros-2018-10-21T17:12:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Teukros-2018-10-21T17:12:00.000Z-Czyz1-2018-10-21T12:45:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-21T19:09:00.000Z",
"author": "Czyz1",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"188/0/7/0/0",
"188/0/7/0/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"188/0/7/0/1",
"188/0/7/0/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"188/0/7/0/4/2",
"188/0/7/0/4/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 3,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Czyz1-2018-10-21T19:09:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Czyz1-2018-10-21T19:09:00.000Z-Teukros-2018-10-21T17:12:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
}
]
},
{
"type": "comment",
"timestamp": "2018-10-21T19:25:00.000Z",
"author": "Mike210381",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"188/2/0",
"188/2/8/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"188/2/5",
"188/2/8/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"188/2/8/2",
"188/2/8/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Mike210381-2018-10-21T19:25:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Mike210381-2018-10-21T19:25:00.000Z-Czyz1-2018-10-21T12:45:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
},
{
"type": "comment",
"timestamp": "2018-10-22T15:26:00.000Z",
"author": "Aotearoa",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"188/4/0",
"188/4/8/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"188/4/5",
"188/4/8/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"188/4/8/1",
"188/4/8/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Aotearoa-2018-10-22T15:26:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Aotearoa-2018-10-22T15:26:00.000Z-Czyz1-2018-10-21T12:45:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-23T10:29:00.000Z",
"author": "Czyz1",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"188/4/9/0/0",
"188/4/9/0/6/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"188/4/9/0/3",
"188/4/9/0/6/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"188/4/9/0/6/2",
"188/4/9/0/6/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 3,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Czyz1-2018-10-23T10:29:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Czyz1-2018-10-23T10:29:00.000Z-Aotearoa-2018-10-22T15:26:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
}
]
},
{
"type": "comment",
"timestamp": "2018-10-23T16:44:00.000Z",
"author": "Eurohunter",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"188/6/0",
"188/6/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"188/6/1",
"188/6/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"188/6/4/2",
"188/6/4/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Eurohunter-2018-10-23T16:44:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Eurohunter-2018-10-23T16:44:00.000Z-Czyz1-2018-10-21T12:45:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
}
]
}
]
},
{
"placeholderHeading": false,
"type": "heading",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"190/0/0",
"190/0/1"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"headingLevel": 2,
"level": 0,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "h-Stok-2018-10-23T19:16:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "h-Osiedla-2018-10-23T19:16:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-23T19:16:00.000Z",
"author": "Stok",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"192/0",
"194/3/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"194/0",
"194/3/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"194/3/2",
"194/3/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 1,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Stok-2018-10-23T19:16:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Stok-2018-10-23T19:16:00.000Z-Osiedla",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
}
]
},
{
"placeholderHeading": false,
"type": "heading",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"196/0/0",
"196/0/1"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"headingLevel": 2,
"level": 0,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "h-Eurohunter-2018-10-23T21:59:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "h-Pokolenie_Z-2018-10-23T21:59:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-23T21:59:00.000Z",
"author": "Eurohunter",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"198/0",
"198/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"198/1",
"198/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"198/4/2",
"198/4/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 1,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Eurohunter-2018-10-23T21:59:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Eurohunter-2018-10-23T21:59:00.000Z-Pokolenie_Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
}
]
},
{
"placeholderHeading": false,
"type": "heading",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"200/1/0",
"200/1/1"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"headingLevel": 2,
"level": 0,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "h-Agrezt-2018-10-25T07:59:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "h-Zofia_Więcławówna_-_korekta_umiejscowienia_grafiki-2018-10-25T07:59:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-25T07:59:00.000Z",
"author": "Agrezt",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"202/0",
"203/3/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"203/0",
"203/3/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"203/3/2",
"203/3/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 1,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Agrezt-2018-10-25T07:59:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Agrezt-2018-10-25T07:59:00.000Z-Zofia_Więcławówna_-_korekta_umiejscowienia_grafiki",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-25T08:15:00.000Z",
"author": "Ciacho5",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"205/0/0",
"205/0/10/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"205/0/7",
"205/0/10/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"205/0/10/2",
"205/0/10/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Ciacho5-2018-10-25T08:15:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Ciacho5-2018-10-25T08:15:00.000Z-Agrezt-2018-10-25T07:59:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
}
]
}
]
},
{
"placeholderHeading": false,
"type": "heading",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"207/0/0",
"207/0/1"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"headingLevel": 2,
"level": 0,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "h-Andrzej1993-2018-10-14T19:29:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "h-Bitwa_pod_Brunete-2018-10-14T19:29:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-14T19:29:00.000Z",
"author": "Andrzej1993",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"209/0",
"215/3/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"215/0",
"215/3/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"215/3/2",
"215/3/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 1,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Andrzej1993-2018-10-14T19:29:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Andrzej1993-2018-10-14T19:29:00.000Z-Bitwa_pod_Brunete",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-25T12:59:00.000Z",
"author": "KamilK7",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"217/0/0",
"217/0/8/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"217/0/5",
"217/0/8/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"217/0/8/1",
"217/0/8/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-KamilK7-2018-10-25T12:59:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-KamilK7-2018-10-25T12:59:00.000Z-Andrzej1993-2018-10-14T19:29:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
}
]
}
]
},
{
"placeholderHeading": false,
"type": "heading",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"219/1/0",
"219/1/1"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"headingLevel": 2,
"level": 0,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "h-Karol_Szapsza-2018-10-25T20:40:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "h-Podwójne_nazwiska_w_infoboksach-2018-10-25T20:40:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-25T20:40:00.000Z",
"author": "Karol Szapsza",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"221/0",
"221/6/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"221/3",
"221/6/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"221/6/2",
"221/6/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 1,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Karol_Szapsza-2018-10-25T20:40:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Karol_Szapsza-2018-10-25T20:40:00.000Z-Podwójne_nazwiska_w_infoboksach",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
}
]
},
{
"placeholderHeading": false,
"type": "heading",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"224/1/0",
"224/1/1"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"headingLevel": 2,
"level": 0,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "h-Piotr967-2018-10-23T16:01:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "h-Facebook_jako_źródło-2018-10-23T16:01:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-23T16:01:00.000Z",
"author": "Piotr967",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"226/0",
"226/6/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"226/3",
"226/6/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"226/6/1",
"226/6/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 1,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Piotr967-2018-10-23T16:01:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Piotr967-2018-10-23T16:01:00.000Z-Facebook_jako_źródło",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-23T16:43:00.000Z",
"author": "Eurohunter",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"228/0/0",
"228/0/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"228/0/1",
"228/0/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"228/0/4/2",
"228/0/4/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Eurohunter-2018-10-23T16:43:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Eurohunter-2018-10-23T16:43:00.000Z-Piotr967-2018-10-23T16:01:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
},
{
"type": "comment",
"timestamp": "2018-10-23T16:46:00.000Z",
"author": "Tokyotown8",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"230/0/0",
"230/0/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"230/0/1",
"230/0/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"230/0/4/2",
"230/0/4/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Tokyotown8-2018-10-23T16:46:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Tokyotown8-2018-10-23T16:46:00.000Z-Piotr967-2018-10-23T16:01:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-23T16:52:00.000Z",
"author": "Piotr967",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"232/0/0/0/0",
"232/0/0/0/4/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"232/0/0/0/1",
"232/0/0/0/4/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"232/0/0/0/4/1",
"232/0/0/0/4/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 3,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Piotr967-2018-10-23T16:52:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Piotr967-2018-10-23T16:52:00.000Z-Tokyotown8-2018-10-23T16:46:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-23T17:16:00.000Z",
"author": "Tokyotown8",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"232/0/0/0/5/0/0",
"232/0/0/0/5/0/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"232/0/0/0/5/0/1",
"232/0/0/0/5/0/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"232/0/0/0/5/0/4/2",
"232/0/0/0/5/0/4/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 4,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Tokyotown8-2018-10-23T17:16:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Tokyotown8-2018-10-23T17:16:00.000Z-Piotr967-2018-10-23T16:52:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
},
{
"type": "comment",
"timestamp": "2018-10-23T17:40:00.000Z",
"author": "Mpn",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"232/0/0/0/5/2/0",
"232/0/0/0/5/2/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"232/0/0/0/5/2/1",
"232/0/0/0/5/2/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"232/0/0/0/5/2/4/2",
"232/0/0/0/5/2/4/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 4,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Mpn-2018-10-23T17:40:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Mpn-2018-10-23T17:40:00.000Z-Piotr967-2018-10-23T16:52:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-23T17:44:00.000Z",
"author": "Szoltys",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"232/0/0/0/5/2/5/0/0",
"232/0/0/0/5/2/5/0/2/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"232/0/0/0/5/2/5/0/1",
"232/0/0/0/5/2/5/0/2/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"232/0/0/0/5/2/5/0/2/1",
"232/0/0/0/5/2/5/0/2/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 5,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Szoltys-2018-10-23T17:44:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Szoltys-2018-10-23T17:44:00.000Z-Mpn-2018-10-23T17:40:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
}
]
}
]
}
]
},
{
"type": "comment",
"timestamp": "2018-10-23T17:56:00.000Z",
"author": "109.233.47.132",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"234/0/0",
"234/0/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"234/0/1",
"234/0/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"234/0/4/2",
"234/0/4/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-109.233.47.132-2018-10-23T17:56:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-109.233.47.132-2018-10-23T17:56:00.000Z-Piotr967-2018-10-23T16:01:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
},
{
"type": "comment",
"timestamp": "2018-10-23T20:50:00.000Z",
"author": "Panek",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"234/2/0",
"234/2/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"234/2/1",
"234/2/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"234/2/4/2",
"234/2/4/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Panek-2018-10-23T20:50:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Panek-2018-10-23T20:50:00.000Z-Piotr967-2018-10-23T16:01:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
},
{
"type": "comment",
"timestamp": "2018-10-23T21:36:00.000Z",
"author": "Kenraiz",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"234/4/0",
"234/4/4/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"234/4/1",
"234/4/4/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"234/4/4/1",
"234/4/4/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Kenraiz-2018-10-23T21:36:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Kenraiz-2018-10-23T21:36:00.000Z-Piotr967-2018-10-23T16:01:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
},
{
"type": "comment",
"timestamp": "2018-10-25T12:09:00.000Z",
"author": "KamilK7",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"234/6/0",
"234/6/4/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"234/6/1",
"234/6/4/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"234/6/4/1",
"234/6/4/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-KamilK7-2018-10-25T12:09:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-KamilK7-2018-10-25T12:09:00.000Z-Piotr967-2018-10-23T16:01:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
}
]
},
{
"type": "comment",
"timestamp": "2018-10-25T17:14:00.000Z",
"author": "BasileusAutokratorPL",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"236/0",
"236/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"236/1",
"236/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"236/4/2",
"236/4/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 1,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-BasileusAutokratorPL-2018-10-25T17:14:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-BasileusAutokratorPL-2018-10-25T17:14:00.000Z-Facebook_jako_źródło",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-25T18:43:00.000Z",
"author": "Żyrafał",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"238/0/0",
"238/0/10/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"238/0/7",
"238/0/10/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"238/0/10/1",
"238/0/10/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Żyrafał-2018-10-25T18:43:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Żyrafał-2018-10-25T18:43:00.000Z-BasileusAutokratorPL-2018-10-25T17:14:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
},
{
"type": "comment",
"timestamp": "2018-10-26T05:36:00.000Z",
"author": "Piolodiusz",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"240/0/0",
"240/0/8/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"240/0/5",
"240/0/8/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"240/0/8/2",
"240/0/8/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Piolodiusz-2018-10-26T05:36:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Piolodiusz-2018-10-26T05:36:00.000Z-BasileusAutokratorPL-2018-10-25T17:14:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-26T16:09:00.000Z",
"author": "Żyrafał",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"240/0/9/0/0",
"240/0/9/0/6/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"240/0/9/0/3",
"240/0/9/0/6/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"240/0/9/0/6/1",
"240/0/9/0/6/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 3,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Żyrafał-2018-10-26T16:09:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Żyrafał-2018-10-26T16:09:00.000Z-Piolodiusz-2018-10-26T05:36:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
}
]
},
{
"type": "comment",
"timestamp": "2018-10-26T16:13:00.000Z",
"author": "Michalwadas",
CommentParser: Output display name if different to username The only normalisation we apply for comparison is lowercasing. Change-Id: Id3d57c2066429fcedc7dcc091e74ed46e17060f1
2022-02-04 18:16:24 +00:00
"displayName": "Michał Wadas",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"242/0/0",
"242/0/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"242/0/1",
"242/0/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"242/0/4/2",
"242/0/4/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Michalwadas-2018-10-26T16:13:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Michalwadas-2018-10-26T16:13:00.000Z-BasileusAutokratorPL-2018-10-25T17:14:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
}
]
}
]
},
{
"placeholderHeading": false,
"type": "heading",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"244/1/0",
"244/1/1"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"headingLevel": 2,
"level": 0,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "h-WTM-2018-10-26T17:17:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "h-Wyśrodkowanie_tekstu-2018-10-26T17:17:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-26T17:17:00.000Z",
"author": "WTM",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"246/0",
"248/0/6/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"248/0/3",
"248/0/6/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"248/0/6/2",
"248/0/6/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 1,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-WTM-2018-10-26T17:17:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-WTM-2018-10-26T17:17:00.000Z-Wyśrodkowanie_tekstu",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [
"Comment starts and ends with different indentation"
],
"replies": []
}
]
},
{
"placeholderHeading": false,
"type": "heading",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"250/1/0",
"250/1/1"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"headingLevel": 2,
"level": 0,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "h-BasileusAutokratorPL-2018-10-26T17:07:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "h-Reprezentacja_Polski_w_piłce_nożnej_mężczyzn-2018-10-26T17:07:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-26T17:07:00.000Z",
"author": "BasileusAutokratorPL",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"251/0",
"254/8/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"254/5",
"254/8/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"254/8/2",
"254/8/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 1,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-BasileusAutokratorPL-2018-10-26T17:07:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-BasileusAutokratorPL-2018-10-26T17:07:00.000Z-Reprezentacja_Polski_w_piłce_nożnej_mężczyzn",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-26T17:27:00.000Z",
"author": "Ciacho5",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"256/0/0",
"256/0/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"256/0/1",
"256/0/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"256/0/4/2",
"256/0/4/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Ciacho5-2018-10-26T17:27:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Ciacho5-2018-10-26T17:27:00.000Z-BasileusAutokratorPL-2018-10-26T17:07:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
},
{
"type": "comment",
"timestamp": "2018-10-26T20:19:00.000Z",
"author": "Mitrovitz",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"256/2/0",
"256/2/12/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"256/2/9",
"256/2/12/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"256/2/12/1",
"256/2/12/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Mitrovitz-2018-10-26T20:19:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Mitrovitz-2018-10-26T20:19:00.000Z-BasileusAutokratorPL-2018-10-26T17:07:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-26T21:21:00.000Z",
"author": "Vinsinna",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"256/2/13/0/0",
"256/2/13/0/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"256/2/13/0/1",
"256/2/13/0/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"256/2/13/0/4/2",
"256/2/13/0/4/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 3,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Vinsinna-2018-10-26T21:21:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Vinsinna-2018-10-26T21:21:00.000Z-Mitrovitz-2018-10-26T20:19:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
}
]
},
{
"type": "comment",
"timestamp": "2018-10-26T22:01:00.000Z",
"author": "Piotr967",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"256/4/0",
"256/4/8/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"256/4/1",
"256/4/4/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"256/4/5",
"256/4/8/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"256/4/4/1",
"256/4/4/26"
],
[
"256/4/8/1",
"256/4/8/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Piotr967-2018-10-26T22:01:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Piotr967-2018-10-26T22:01:00.000Z-BasileusAutokratorPL-2018-10-26T17:07:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
},
{
"type": "comment",
"timestamp": "2018-10-26T22:10:00.000Z",
"author": "Ented",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"256/6/0",
"256/6/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"256/6/1",
"256/6/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"256/6/4/2",
"256/6/4/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Ented-2018-10-26T22:10:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Ented-2018-10-26T22:10:00.000Z-BasileusAutokratorPL-2018-10-26T17:07:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
}
]
}
]
},
{
"placeholderHeading": false,
"type": "heading",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"258/0/0",
"258/0/1"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"headingLevel": 2,
"level": 0,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "h-Wiktoryn-2018-10-20T18:23:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "h-Ikonki_w_infoboksach-2018-10-20T18:23:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-20T18:23:00.000Z",
"author": "Wiktoryn",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"260/0",
"260/6/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"260/3",
"260/6/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"260/6/1",
"260/6/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 1,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Wiktoryn-2018-10-20T18:23:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Wiktoryn-2018-10-20T18:23:00.000Z-Ikonki_w_infoboksach",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-28T17:53:00.000Z",
"author": "Malarz pl",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"262/0/0",
"262/0/6/25"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"262/0/3",
"262/0/6/25"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"262/0/6/1",
"262/0/6/25"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Malarz_pl-2018-10-28T17:53:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Malarz_pl-2018-10-28T17:53:00.000Z-Wiktoryn-2018-10-20T18:23:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
},
{
"type": "comment",
"timestamp": "2018-10-28T18:50:00.000Z",
"author": "Elfhelm",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"262/2/0",
"262/2/4/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"262/2/1",
"262/2/4/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"262/2/4/2",
"262/2/4/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Elfhelm-2018-10-28T18:50:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Elfhelm-2018-10-28T18:50:00.000Z-Wiktoryn-2018-10-20T18:23:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
}
]
}
]
},
{
"placeholderHeading": false,
"type": "heading",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"264/1/0",
"264/1/1"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"headingLevel": 2,
"level": 0,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "h-Arvedui89-2018-10-29T08:12:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "h-{{Tenisista_infobox}}-2018-10-29T08:12:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-29T08:12:00.000Z",
"author": "Arvedui89",
CommentParser: Output display name if different to username The only normalisation we apply for comparison is lowercasing. Change-Id: Id3d57c2066429fcedc7dcc091e74ed46e17060f1
2022-02-04 18:16:24 +00:00
"displayName": "Aʀvєδuι",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"266/0",
"266/6/25"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"266/3",
"266/6/25"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"266/6/1",
"266/6/25"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 1,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Arvedui89-2018-10-29T08:12:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Arvedui89-2018-10-29T08:12:00.000Z-{{Tenisista_infobox}}",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-29T10:20:00.000Z",
"author": "Ciacho5",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"268/0/0",
"268/0/4/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"268/0/1",
"268/0/4/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"268/0/4/2",
"268/0/4/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Ciacho5-2018-10-29T10:20:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Ciacho5-2018-10-29T10:20:00.000Z-Arvedui89-2018-10-29T08:12:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
},
{
"type": "comment",
"timestamp": "2018-10-29T10:49:00.000Z",
"author": "Malarz pl",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"268/2/0",
"268/2/7/25"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"268/2/4",
"268/2/7/25"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"268/2/7/1",
"268/2/7/25"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Malarz_pl-2018-10-29T10:49:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Malarz_pl-2018-10-29T10:49:00.000Z-Arvedui89-2018-10-29T08:12:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
}
]
}
]
},
{
"placeholderHeading": false,
"type": "heading",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"270/1/0",
"270/1/1"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"headingLevel": 2,
"level": 0,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "h-Smyrwa-2018-10-24T01:36:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "h-Encyklopedyczność_biogramów_harcerskich-2018-10-24T01:36:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-24T01:36:00.000Z",
"author": "Smyrwa",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"272/0",
"290/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"290/1",
"290/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"290/4/2",
"290/4/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 1,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Smyrwa-2018-10-24T01:36:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Smyrwa-2018-10-24T01:36:00.000Z-Encyklopedyczność_biogramów_harcerskich",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-24T06:35:00.000Z",
"author": "Gżdacz",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"292/0/0",
"292/0/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"292/0/1",
"292/0/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"292/0/4/2",
"292/0/4/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Gżdacz-2018-10-24T06:35:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Gżdacz-2018-10-24T06:35:00.000Z-Smyrwa-2018-10-24T01:36:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
}
]
},
{
"type": "comment",
"timestamp": "2018-10-24T15:14:00.000Z",
"author": "Pbk",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"292/2/0",
"296/10/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"296/1",
"296/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"296/7",
"296/10/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"296/4/2",
"296/4/27"
],
[
"296/10/2",
"296/10/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 1,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Pbk-2018-10-24T15:14:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Pbk-2018-10-24T15:14:00.000Z-Encyklopedyczność_biogramów_harcerskich",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [
"Comment starts and ends with different indentation"
],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-30T14:58:00.000Z",
"author": "Smyrwa",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"298/0/0/0/0",
"300/0/0/0/6/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"300/0/0/0/3",
"300/0/0/0/6/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"300/0/0/0/6/2",
"300/0/0/0/6/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 3,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Smyrwa-2018-10-30T14:58:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Smyrwa-2018-10-30T14:58:00.000Z-Pbk-2018-10-24T15:14:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [
"Comment skips indentation level"
],
"replies": []
},
{
"type": "comment",
"timestamp": "2018-10-25T10:50:00.000Z",
"author": "Andrzei111",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"302/0/0",
"302/0/6/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"302/0/3",
"302/0/6/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"302/0/6/2",
"302/0/6/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Andrzei111-2018-10-25T10:50:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Andrzei111-2018-10-25T10:50:00.000Z-Pbk-2018-10-24T15:14:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-30T14:58:00.000Z",
"author": "Smyrwa",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"304/0/0/0/0",
"304/0/0/0/6/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"304/0/0/0/3",
"304/0/0/0/6/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"304/0/0/0/6/2",
"304/0/0/0/6/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 3,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Smyrwa-2018-10-30T14:58:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Smyrwa-2018-10-30T14:58:00.000Z-Andrzei111-2018-10-25T10:50:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
}
]
},
{
"type": "comment",
"timestamp": "2018-10-25T10:59:00.000Z",
"author": "Felis domestica",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"306/0/0",
"306/0/6/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"306/0/3",
"306/0/6/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"306/0/6/2",
"306/0/6/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Felis_domestica-2018-10-25T10:59:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Felis_domestica-2018-10-25T10:59:00.000Z-Pbk-2018-10-24T15:14:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
}
]
},
{
"type": "comment",
"timestamp": "2018-10-25T20:03:00.000Z",
"author": "BasileusAutokratorPL",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"308/0",
"308/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"308/1",
"308/4/27"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"308/4/2",
"308/4/27"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 1,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-BasileusAutokratorPL-2018-10-25T20:03:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-BasileusAutokratorPL-2018-10-25T20:03:00.000Z-Encyklopedyczność_biogramów_harcerskich",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-30T14:58:00.000Z",
"author": "Smyrwa",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"310/0/0/0/0",
"310/0/0/0/4/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"310/0/0/0/1",
"310/0/0/0/4/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"310/0/0/0/4/2",
"310/0/0/0/4/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 3,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Smyrwa-2018-10-30T14:58:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Smyrwa-2018-10-30T14:58:00.000Z-BasileusAutokratorPL-2018-10-25T20:03:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [
"Comment skips indentation level"
],
"replies": []
}
]
}
]
},
{
"placeholderHeading": false,
"type": "heading",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"312/1/0",
"312/1/1"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"headingLevel": 2,
"level": 0,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "h-Ciacho5-2018-10-31T13:57:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "h-Szablon?_do_zrobienia?-2018-10-31T13:57:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-31T13:57:00.000Z",
"author": "Ciacho5",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"314/0",
"314/8/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"314/5",
"314/8/26"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"314/8/2",
"314/8/26"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 1,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Ciacho5-2018-10-31T13:57:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Ciacho5-2018-10-31T13:57:00.000Z-Szablon?_do_zrobienia?",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": [
{
"type": "comment",
"timestamp": "2018-10-31T14:55:00.000Z",
"author": "Malarz pl",
"range": [
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"316/0/0",
"316/0/4/25"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
],
"signatureRanges": [
[
tests: Fix root node / container handling Since times immemorial, and for reasons lost to history, our test code was adding an extra <div> wrapper before parsing the HTML used for tests. This wasn't a problem, until now, because I want to add some tests for T303396 that need to check that the *real* wrappers present in some test cases are handled correctly. Changes to test cases mostly remove a leading "0/" from serialized ranges, corresponding to removing the extra wrapper. Change-Id: Ia50e3590538c8cd274b02d2a937ba1a3fbb4ac89
2022-03-09 19:32:51 +00:00
"316/0/1",
"316/0/4/25"
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
]
],
Parser: Store timestamp ranges Change-Id: Ifcbe22011f11f4374f38b7aa346da5a96cac968c
2022-10-04 12:50:57 +00:00
"timestampRanges": [
[
"316/0/4/1",
"316/0/4/25"
]
],
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"level": 2,
Introduce comment "names" to identify comments across revisions/pages The existing comment IDs can't be used to find the same comment on a different revision or page (when it's transcluded), because they depend on the comment's parent and its position on the page. Comment names depend only on the author and timestamp. The trade-off is that they can't distinguish comments posted within the same minute, or in the same edit, so we will still need the IDs sometimes. Prefer using comment names when replying, if they're not ambiguous. This fixes T273413 and T275821. Heading names depend on the author and timestamp of the oldest comment. This way we don't have to detect changes to the heading text, but we can't distinguish headings without any comments. Bug: T274685 Bug: T273413 Bug: T275821 Change-Id: Id85c50ba38d1e532cec106708c077b908a3fcd49
2021-02-12 19:16:13 +00:00
"name": "c-Malarz_pl-2018-10-31T14:55:00.000Z",
Make IDs (to be used as URL hashes) wikitext safe * Use hyphens instead of pipes a separators * Use underscores for spaces in usernames Change-Id: I6efd9739fc73e45002e50e64c43ce0de1c2f1239
2021-03-17 14:39:41 +00:00
"id": "c-Malarz_pl-2018-10-31T14:55:00.000Z-Ciacho5-2018-10-31T13:57:00.000Z",
Use tabs for indentation in JSON test files Change-Id: I1d8f8b33b19bcff249ad08dfe687f87f5e5bf9bf
2021-01-26 23:55:15 +00:00
"warnings": [],
"replies": []
}
]
}
]
}
Integration tests for parsing Tests that handle a specific case and describe what they are testing would be nice… but tests that just document the current status to avoid regressions are also okay and easier to add. Change-Id: I0b3530ae0e77de70932aaf623f5290d1876699a0
2019-10-24 12:59:42 +00:00
]
Reference in a new issue Copy permalink