To use WANObjectCache correctly in a multi-DC-safe way, we need to use
getWithSetCallback() to read data, and call delete() when it changes.
NotifUser's caching of notification counts and timestamps relied
heavily on set() calls, and so wasn't multi-DC-safe.
Changes in this commit:
* Rather than caching counts/timestamps in separate cache keys, and
using separate cache keys for each section (alert/message/all), put
all this data in an array and store that in a single cache key.
This reduces the number of cache keys per user per wiki from 6 to 1.
* Similarly, use a single global cache key per user. The global check
key for the last updated timestamp is retained, so we now have
2 global cache keys per user (down from 7)
* Remove preloading using getMulti(), no longer needed
* Move computation of counts and timestamps into separate compute
functions (one for local, one for global), and wrap them with
a getter that uses getWithSetCallback().
* Use TS_MW strings instead of MWTimestamp objects internally, to
simplify comparisons and max() operations.
* Make existing getters wrap around this new getter. They now ignore
their $cached and $dbSource parameters, and we should deprecate/change
these function signatures.
* In resetNotificationCounts(), just delete the cache keys. In global
mode, also recompute the notification counts and put them in the
echo_unread_wikis table. We could also set() the data into the cache
at this point, but don't, because you're not supposed to mix set() and
getWithSetCallback() calls and I don't want to find out what happens
if you do.
Bug: T164860
Change-Id: I4f86aab11d50d20280a33e0504ba8ad0c6c01842
All files containing more than one PHP class were split into
multiple files.
extension.json was updated to match new class locations.
phpcs `OneObjectStructurePerFile.MultipleFound` rule was
re-enabled.
Bug: T177809
Change-Id: I6fc2ec9cc35e6bac5a7c44d94b0f1b1b40e6dba5
The continuous integration infrastructure for these is being removed
(or never existed), and our team decided not to invest the time to
convert them to node.js.
Bug: T171848
Change-Id: I0faeecb2635f24c40c83aa689b670e69aa381431
When intval() fails, the function returns a zero. We should remove
the failures from the blacklist.
Bug: T178512
Change-Id: I89ad680a287da16c2fbd6aa4d53a725142429144
Otherwise, if $list->getValues() contains the number 0,
any non-numerical string will match, because 'foo'==0 is true.
This, in combination with a broken maintenance script that had
inserted 0s into some users' blacklist, broke all notifications
for those users.
Bug: T177825
Change-Id: If8700b4d0de0fdba876eb9d5cc4997e185dfeb3c
With 349457, anytime a revision is created, a corresponding record
is created in the new ip_changes table. This may cause tests to fail
if they don't drop the ip_changes table between individual tests.
https://gerrit.wikimedia.org/r/#/c/349457/
Change-Id: I48f0b64f19f9582b40540fa1b42a39d281979625
The return value from the method is only suitable for passing to
$db->insert(). To get the inserted ID, you need to call $db->insertId()
even if $db->nextSequenceValue() returned non-null.
Bug: T164900
Change-Id: I466fd372804927b3ad72125c7a69d253bd7a24f8
Some browser tests were broken by 945fccf009.
The badge element is now technically rendered offscreen, with only the
:before and :after pseudoelements being onscreen. Because of this, Selenium
thinks that the badges are invisible, and this breaks various things in
totally unexpected ways.
* article_page.rb: Store references to the parent <li> elements of badges.
This might not be necessary but I don't know how to access them otherwise.
* badge_steps.rb: When clicking the badges, click the parent <li> element
rather than the not-really-invisible <a>. Effectively, the <a> gets
clicked anyway, since they overlap.
* no_javascript.feature/no_javascript.rb: Wait for page load before
checking that we're on the right page. The wait is no longer
implicit, since Selenium thinks we're clicking the <li> rather than
<a> (links are special-cased).
* notification_steps.rb:
* Check whether the badges exist on the page, rather than whether they
are visible.
* Use a weird hack to read badge text. Apparently you can't read the text
of elements that Selenium thinks are invisible.
http://stackoverflow.com/questions/20888592/gettext-method-of-selenium-chrome-driver-sometimes-returns-an-empty-string
Bug: T161941
Change-Id: Ic6bcd1088249109e49a47cc9007e6ee002d3d8ba
Other tests (possibly from other enabled extensions) could also,
by chance, create Echo notifications for the user MWEchoThankYouEditTests
uses. In order to those tests be reliable, they should make sure
there is no notification data in the database prior to running tests.
Bug: T161087
Change-Id: I870a50b1f831795731235fa8ec97477b3e470b50
When content is changed and the change contains the signature
of the user, the method checking for reasonable mentions in
that changes did not consider multiple signatures.
The patch fixes that and adds a test for it.
Bug: T154406
Change-Id: I86303f42e97d16c68e3235b0e2d13542ceedf1fe
This speeds up the tests from 2 mins to 50 seconds for me.
(with a slow setup, no caching etc.)
Bug: T158120
Change-Id: I8adb6c8fe783d1be8841a1139bb141da3b046f9d
