Commit graph

62 commits

Author SHA1 Message Date
Inez Korczynski 29a8ac77c3 Merge branch 'dmrewrite' of ssh://gerrit.wikimedia.org:29418/mediawiki/extensions/VisualEditor into dmrewrite 2012-06-13 16:23:35 -07:00
Inez Korczynski 81473a8119 Make method ve.dm.Document.addAnnotationsToData accept collection of annotations as a param - instead of an array 2012-06-13 16:23:19 -07:00
Catrope 0bc1693998 Fix partial undo of ab0c37 in 2ea103
Change-Id: I42225802c45b3867a2aafd40e6941754b1b415b4
2012-06-13 16:16:07 -07:00
Trevor Parscal 429225fb57 Whitespace fixes (jshint was tweaking out)
Change-Id: Id75a097ae4958029caf0a0315343c2fab9c99b89
2012-06-13 16:06:21 -07:00
Trevor Parscal 169b7e754d Cleaned up toolbar annotation updates
Change-Id: I8cefd6c871ed6d1ad71f64eb66236de1b8b719b5
2012-06-13 14:31:46 -07:00
Rob Moen bf08347922 Fix js error
Change-Id: I3424f205649345a97a7f4d28c4c040c3c3bb6aa7
2012-06-12 11:17:46 -07:00
Rob Moen 2ea1030ab4 Rewrite document annotation getters to return collection objects
vs arrays

Change-Id: Ib48a39fff501cdb769f5b9b9fc7717805b74236e
2012-06-12 11:10:29 -07:00
Catrope ab0c37def9 Rip out the offset map
The offset map was broken from the start because it wasn't updated when
adjusting the length of a text node, and if we fix that bug it's
doubtful whether the costs of updating the offset map outweigh the
benefits, especially considering that adjusting the length of a text
node is something we do for almost every keypress. If it turns out
having an offset map does make sense, we can always reintroduce it
later.

Change-Id: I59e8bc154f7d07aa1bab2f473c13ff466d0e463f
2012-06-11 18:10:08 -07:00
Inez Korczynski a609afe52f Port static method addAnnotationsToData from ve to ve2 2012-06-11 15:31:55 -07:00
Christian Williams 80194260f3 Undo and Redo
Change-Id: I2d7c1bfd164a42e5d52001054e52e47ca245ad18
2012-06-08 16:21:47 -07:00
Trevor Parscal 5aa6f4b342 Fix bug in fixupInsertion()
When a text node was closed (either by encountering a non-text node or by reaching the end of the document), the fixupStack would not be processed. This led to attributes being dropped when nodes were split because of text being inserted into them

Change-Id: I41f6d20e0c1bfc8d8689b7e6325e724dd8156ab1
2012-06-08 12:23:58 -07:00
Rob Moen e1b8f9422d Fix bug in getAnnotationsFromRange where if the range length was 0
it would return no annotations.

Change-Id: Iec33366b2a1249122a21617d6681ae039557c2fe
2012-06-07 14:53:30 -07:00
Catrope 3c1de3dc4f Add ve.dm.Document.getBalancedData()
Change-Id: I2420ef5369b9eae053962898b64dd7324e714bbd
2012-06-07 13:17:57 -07:00
Catrope feac2f6069 Factor out element cloning into a getClonedElement() function
Change-Id: Id6024c15d4d9c9141bc4e1afa39fce29992b8647
2012-06-07 13:17:57 -07:00
Christian Williams 51aa317df3 Undo and Redo support
Change-Id: Icdbebccefe3cc162b9ff6c0d978d5e4fc1f2c561
2012-06-06 20:28:06 -07:00
Rob Moen 232f2c58f7 Fix linkInspector tool
Change-Id: I89fa17b1a72c3f45bce5a11b49aa5445c5990f68
2012-06-06 17:50:01 -07:00
Rob Moen 8db70ed263 Annotation buttons now working
Created new document method to determine if a specific annotation
object is inside an annotation array.

Change-Id: Id645929cbf31030b8b0fcacb8dfb36e61aaad129
2012-06-06 17:06:37 -07:00
Catrope 02b19f337f Add zero-length text nodes to empty content nodes
This is needed to make the results of certain transactions' tree sync
round-trip cleanly through the ve.dm.Document constructor

Change-Id: I2ab0758ec6bd7afba5b6645c7330f9fa2d45205d
2012-06-06 15:33:42 -07:00
Rob Moen 2439c0149a Created secondary method for getMatchingAnnotations
new static method looks for annotation in annotation object.
ve2 Cleanup on annotate method and surface model.
Partially revive UI tools by exchaning old method usage
for ve2 methods.

Change-Id: Id0ac58330292d76801bbcf1d71a919b493f8ab9e
2012-06-06 11:27:23 -07:00
Catrope db793009be fixupInsertion() variable documentation and cleanup
Functional changes (fixes):
* Make writeElement() also update parentNode and parentType for openings
* Also add to fixupStack when opening a wrapper for a text node

Non-functional changes (cleanup&docs):
* Document all variables at the beginning of the function
* Group variables according to where/how they're used
* Move expectedType into writeElement()
* Kill node, duplicates parentNode unnecessarily
* Kill paragraphOpened, was misnamed and unnecessary
* Rename closedElements to reopenElements

Change-Id: Ie5b4e4f30b267943048fdc170accb29139039192
2012-06-04 16:07:16 -07:00
Catrope f7445b37b8 Retain attributes when reopening closed nodes
* Push entire elements onto openingStack rather than type strings
* When closing an element, build a clone of the opening and push it onto
  closedElements, then insert that clone when reopening the element

Change-Id: I8b0fb44394aed6c471dc6dacaab03e44c2333733
2012-06-04 16:07:16 -07:00
Catrope 109624e8b3 fixupInsertion fixes, wrapping content works now
Change-Id: I1eee6afcffbf09955578b7f0534aa5b7234802df
2012-06-04 16:01:13 -07:00
Trevor Parscal f6864b0c04 Changed use of "factory" to "nodeFactory" to make way for other kinds of factories
Change-Id: I96db0f28bb220f1c3b23990824e9523278cb8f9b
2012-05-31 15:20:58 -07:00
Catrope f6ca37926d Add ve.dm.Document.isContentData()
Some of the replacement code was assuming that "does not contain
elements" and "is content" were the same. They're not any more, because
we have content nodes (like image) now, so I need a separate function
to distinguish between these cases.

Change-Id: I206ccdf082b7baddf99d382eb3cdd77ea34fb479
2012-05-31 06:19:34 -07:00
Catrope e92db86046 Fix bug in ve.dm.Document constructor for input ending with text
If the last element of the input data array was text, the resulting text
node would have length=0 rather than the expected length value.

Change-Id: I3d089a80b8a447a12ba411b2e11c1b84f14f2959
2012-05-31 06:17:23 -07:00
Trevor Parscal 14862a1212 Fixed bug in isStructuralOffset which didn't consider neighboring a non-content leaf to be a structural offset
This was caused by the fact that a non-structural leaf can not have children, which makes it appear incompatible as a sibling to an arbitrary structural element (like a paragraph) but since it can not contain content we can check that instead.

Change-Id: Ie3c58b4b43f2aa6921f8f82aa82511e231207854
2012-05-25 13:19:56 -07:00
Inez Korczynski afe18a685a Fix methods getNearestStructuralOffset and getNearestContentOffset for case of passing offsets that are already structural or content
Change-Id: I23cd2daab9cef181b01131b3b9c6ead5ec47f9f4
2012-05-24 20:53:50 -07:00
Trevor Parscal c9dddf60da Made getRelativeOffset return -1 if no valid offset is found in document
Change-Id: If4cef5fe800b241c3efe1d27cdf01cd1ceef2248
2012-05-24 11:43:47 -07:00
Trevor Parscal 1ddaec8acb Added getRelativeOffset and several wrappers around it
Change-Id: Ifd10922366650165bc4c21ea8cfdd2f9b7163432
2012-05-23 17:43:24 -07:00
Trevor Parscal 541d786ced Added optional "unrestricted" argument to isStructuralOffset
Using this argument will only return true if the offset is a place you can add any element to (hence the unrestricted part of it). This is good for testing if a paragraph could potentially be inserted there.

Change-Id: I6cc91da437c52493de03eb687b28966198270fea
2012-05-23 14:56:33 -07:00
Catrope 3be3f55c91 Rough first stab at ve.dm.fixupInsertion()
This code still needs a lot of work, but it seems to work for most
cases. Things that still need to be done:
* Documentation and comments
* Handling of content and text nodes
** Use Trevor's isContent/canContainContent code which I don't have yet
* Preserve attributes when reopening closed elements
* Tests :)

Change-Id: I3bc16c964ef158693490a61ce12beb21e6fe2a9d
2012-05-23 12:45:28 -07:00
Trevor Parscal 86591e5325 Restructured static node rules
Change-Id: I7023e1f334694d59767491f1010e7c4c59c0ce54
2012-05-21 17:39:12 -07:00
Rob Moen 272f502693 Modify getAnnotationsFromOffset to account for annotated leaf nodes.
Refactor remaining annotation methods
Refactor tests to use correct annotation types

Change-Id: Ia3ce42b3d11296b4d63277e0e2a1997d23236613
2012-05-21 12:02:04 -07:00
Trevor Parscal 445006d4a8 Made tests a bit stricter before iterating over annotations
Change-Id: I7c7f11395cfb4ad392bad9380119b647c0c1b629
2012-05-17 14:58:01 -07:00
Trevor Parscal 43d62ee8a7 Refactored getMatchingAnnotations and it's tests, and added offsetContainsMatchingAnnotations
Change-Id: I9098ef8d9c7f3a4f5db112cfdcb9edb7ffd17b80
2012-05-17 14:37:29 -07:00
Rob Moen a937ad22f7 Fix getAnnotatedRange to return propper contained range.
Fixed tests

Change-Id: I305eb81644fb3c804455b4fc13bdabcce97ea471
2012-05-17 13:17:18 -07:00
Trevor Parscal 24c35b853f Whitespace fixes
Change-Id: I691c4e16cf22de8dce0858cdcad522d1f6cfef72
2012-05-17 11:27:25 -07:00
Trevor Parscal 08765feab3 Merged ve.dm.DocumentFragment and ve.dm.Document
* Also fixed lots of whitespace issues and some global variable introductions

Change-Id: Ia6c82ef0c3b5c4938cf72afa86da6d79e4cb385a
2012-05-17 11:23:56 -07:00
Trevor Parscal 58aa0e8137 Whitespace fixes and cleanup
Change-Id: Id2220009ed7af7e895990dd4bb373b43d6089b7b
2012-05-16 20:26:05 -07:00
Catrope bcd7431818 Remove getScope(), no longer used
Change-Id: I79e067a8e6b3eb151d3ebe1dfc3aaebe242aed39
2012-05-16 20:16:47 -07:00
Catrope 7e1aa2336e Support splitting nodes in replace tree sync
This means inserting things like </p><p> are now synced correctly and
split the paragraph in the model tree. Merges (removing e.g. </p><p>)
aren't supported yet.

Also, this needs tests, Trevor tells me he's working on porting replace
tests from the old ve/ directory

Change-Id: Ic5050849d7d007a1696dc36548654979aedb53a8
2012-05-14 21:46:39 -07:00
Trevor Parscal e0de881a8a Removed trailing whitespace in multi-line comment blocks
JSHint and various git tools complain about this, so let's just do away with them altogether

Change-Id: I731866bd21f1ee0088fb0a71df3abf92692aca23
2012-05-14 15:05:09 -07:00
Trevor Parscal 03be3c0668 Removed debugger statement from 5b9a2645a6
Change-Id: I047b0c89a7420a86b9350a8d0c0fc8aec695227f
2012-05-14 14:34:50 -07:00
Catrope 5b9a2645a6 Port getScope() from the old VE code and add tests for it
Change-Id: I8c34fed919e60fa1b8cb6a13cb9ca98f3e695421
2012-05-11 16:03:46 -07:00
Trevor Parscal 9cdfb84f9b Added tests to getRelativeContentOffset
Also fixed a typo in the implementation

Change-Id: I86855795009e34d42a3a5857addc5ccb6366130f
2012-05-11 12:14:29 -07:00
Trevor Parscal 1475100a22 Ported isContentOffset, isStructuralOffset, isElementData, containsElementData and getRelativeContentOffset
Tests for all but getRelativeContentOffset are done too

Change-Id: Iee953647ec89e13259ae2a402c50c809f93bc655
2012-05-10 17:20:57 -07:00
Trevor Parscal 1798a478e7 Massive refactor of node trees, generalizing lots of previously dm only stuff
Change-Id: I9dbb7094f86338ea2e725ad29f66e9e4ca2b52ed
2012-05-09 21:11:09 -07:00
Catrope 17f4dc220a Don't use data.length in selectNodes()
I was using data.length to check if the range was out of bounds, but
this is a problem when using selectNodes() inside of tree sync code
(which happens when performing rebuilds). While tree sync is in
progress, the model tree and the linear model don't match, so we
shouldn't be looking at the linear model for information about the model
tree. Instead, get the length of the DocumentNode and use that.
`

Change-Id: I11a378544ce1281a89cdcd4363c5cb1bf56f3434
2012-05-08 18:31:51 -07:00
Catrope 22c88f77c3 Add a 'siblings' mode to selectNodes(), and add tests for it
Change-Id: Ib3fd3ee5aba7aab7c26193a2a9c48612f624862a
2012-05-08 18:31:50 -07:00
Trevor Parscal 123c64cbd5 Added more tests for selectNodes
Change-Id: Ibd9672fb9b6ce682905d20f9346fada6fb47b380
2012-05-07 19:03:59 -07:00