Commit graph

2956 commits

Author SHA1 Message Date
Catrope 4303e9f30c Show annotations on nodes in demo debugger
Change-Id: I41948e8549345c713f1bd0762b08807b8f7617da
2012-11-27 14:41:39 -08:00
Catrope 49963c75fd Store the data model element in the DM tree
This is cleaner than passing around the attributes separately, and it
allows us to access the annotations in dm.LeafNode as well.

Change-Id: Ie5b90988114835831cbe5cdccf63c7cd45719e31
2012-11-27 14:36:29 -08:00
Inez Korczyński a9082e6dde Only apply HTML attributes to DOM nodes that are "safe"
* Added whitelist argument to setDomAttributes which allows filtering of attributes being set
* Added prefix argument to ve.dm.Node.getAttributes to allow extracting a subset of attributes by name prefix
* Added a whitelist to ve.ce.Node which was extracted from MediaWiki's Sanitizer class
* Replaced attribute copying code with a call to setDomAttributes using the whitelist argument, passing in attributes from a call to ve.dm.Node.getAttributes using the prefix argument

Also…

* Removed comment in constructor of ve.ce.Node, documentation for properties is usually in the getters/setters, and already was in this case
* Renamed ve.setDOMAttributes to ve.setDomAttributes
* Renamed ve.getDOMAttributes to ve.getDomAttributes
* Renamed ve.getDOMText to ve.getDomText
* Renamed ve.getDOMHash to ve.getDomHash
* Updated all callers of renamed methods

Change-Id: Id556172d5d18ea431044b9d402400e1f0e67a293
2012-11-27 14:34:29 -08:00
Trevor Parscal 7404d6438a Merge "Fix tests for fda2e6c1b54a0139a3123de413f672e895ff3352" 2012-11-27 21:38:24 +00:00
Trevor Parscal 1e41f63450 Merge "(bug 42172) Tag edits made by VE" 2012-11-27 21:37:59 +00:00
Trevor Parscal d4b3910013 Merge "(bug 42277) First character in empty document behaves strangely in Firefox" 2012-11-27 21:36:27 +00:00
Translation updater bot 4195e00631 Localisation updates from http://translatewiki.net.
Change-Id: I9d20942a1f9095ac9aeb4aa9868120f03e33dde5
2012-11-27 21:12:41 +00:00
Trevor Parscal ee80af5f96 (bug 42404) Errors passing "incorrect" offsets to ve.ce.Surface.getNodeAndOffset
* Adjust offsets to correct positions before showing selection with them
* Throw an error inside of getNodeAndOffset if we can't find a match rather than mysteriously not retuning anything

Change-Id: Ia7347527e6466262e819f456404b32926fd95e34
2012-11-26 15:01:35 -08:00
Catrope 1bc74b0f6d Fix tests for fda2e6c1b5
Change-Id: I2725f3f775e092bafe7aa9dca71d2a9022f16db8
2012-11-26 21:39:14 +00:00
Catrope 43684f88fc (bug 42172) Tag edits made by VE
Uses MW's built-in change tagging. Made this configurable so it can be
turned off if it looks like we're blowing up the tags system with the
volume of VE edits.

Change-Id: I63052d90772906e34021035de496a65fea7945f7
2012-11-26 21:39:11 +00:00
Catrope 568e0e5701 (bug 42277) First character in empty document behaves strangely in Firefox
This was because Firefox intially puts the selection before the first
paragraph, which translates to a model selection of (0,0). Typing with
the cursor at that position causes bad things to happen.

CE normally fixes up the selection when this happens, but it doesn't do
this automatically on initialization. So I added some code to ve.Surface
that causes this fixup to happen.

Change-Id: I0dcee3a29c1242c49ec30c743f1b69686fbb8436
2012-11-26 21:39:01 +00:00
Timo Tijhof 1ba75b7ea9 The last ever mw.ext.ve jshint fixup
Change-Id: I262673214dd59e5bcaf4e0855e68728365b041fe
2012-11-26 22:36:07 +01:00
Translation updater bot 2847e5b8cc Localisation updates from http://translatewiki.net.
Change-Id: Ie008e85ee25d436c63e43dc68b79ad25a35f64e7
2012-11-26 20:37:51 +00:00
James D. Forrester c33de35a98 (bug 42402) Split out two cases so as not to throw an error
Logic called for A || B but then used both; this splits that into two distinct
code paths.

Change-Id: I959e0649fbd5b4f404b5804e26806ba51c31c836
2012-11-23 18:25:23 -08:00
Translation updater bot 3017d9acff Localisation updates from http://translatewiki.net.
Change-Id: If4f899438fdf8990bd35401fbd71fa6618d57754
2012-11-23 21:46:45 +00:00
Catrope 8e2414677f (bug 42279) Inserted characters get annotated with empty set
Turns out this was actually a bug in addAnnotationsToData()

Change-Id: I0c9da662c091811e4053ab0ea2771c4a20b26e09
2012-11-21 17:05:10 -08:00
Trevor Parscal 9c74c97808 Fix even more inspector issues, and some ce ones as well
ve.dm.SurfaceFragment
* Removed flawed implementation of word mode for expandRange method and made use of new getNearestWordBoundary method in the document model

ve.dm.Surface
* Got rid of useInsertionAnnotations, which allowed disabling and enabling of insertion annotations - this isn't needed anymore because it was just a dirty hack around the improper starting and stopping of surface observer that's now solved more elegantly by emitting lock and unlock before committing or rolling back transactions
* Get annotations from the first character of the selection if the selection is not collapsed
* Only emit annotationChange events if it really changed

ve.dm.Document
* Added getNearestWordBoundary method which performs the work behind the surface fragment expandRange word method

ve.ce.SurfaceObserver
* Allow using an initial selection to avoid the observer thinking the selection has changed just because it started out with null
* Only emit selectionChange event if there was a meaningful change

ve.ce.Surface
* (bug 42279) Only annotate characters if insertion annotations are not empty
* Remove manual locking and unlocking, this is now done inside the change method of surface model
* Provide an initial selection to surface observer when we clear it
* Remove enabling and disabling of insertionAnnotations, this isn't needed anymore
* Stop/start observer on key presses that execute actions as well as those that have no special handling

ve.ce.Document
* Make getNodeFromOffsetand getSlugAtOffset return null when given -1 as an offset

Change-Id: Ibf6b26de299e54ae8688a2653bf5d5538927f8c3
2012-11-21 15:26:12 -08:00
Trevor Parscal c9d6d35968 (bug 42219) Fix for document sync issue
When working with a document containing only a slug in an empty paragraph, tree synchronization would break because it was trying to rebuild a non-existent text node.

This change makes the rebuild always occur on the outer range, rather than the inner range, which prevents absent text nodes from being asked to be rebuilt.

Thank you to Roan for debugging this for like 20 min.

Change-Id: I8c3dad921ace395f0694f77cec44305a680657fe
2012-11-21 14:59:09 -08:00
Translation updater bot 8e905f5e60 Localisation updates from http://translatewiki.net.
Change-Id: I86a9e953f42073deeecc237c53533462820ec39c
2012-11-21 22:30:47 +00:00
Trevor Parscal b6139ba65e Merge "(bug 42124) Store comments in the meta-linmod" 2012-11-21 22:12:41 +00:00
Trevor Parscal 43b2c86102 Merge "(bug 42140) Don't force spaces to underscores in internal links" 2012-11-21 22:00:24 +00:00
Trevor Parscal f825e0093a Merge "(bug 42212) Fix JS error when inserting after alien at the end" 2012-11-21 21:56:19 +00:00
Catrope bf7b243627 (bug 42121) Change markers lost for first paragraph on new page
When editing a new page, or loading an empty page into the editor, the
converter generates a paragraph so the document isn't completely empty.
This paragraph is then unwrapped on the way out, potentially destroying
change markers and generally producing strange HTML output.

Mark this paragraph with generated=empty rather than generated=wrapper,
and only unwrap it on the way out if it's still empty. This means we
cleanly round-trip empty documents (and empty list items and the like),
but if the user enters text, we create a paragraph like we're supposed
to.

Change-Id: Id0241221a67b769445676b833b5741320d99ea5f
2012-11-21 13:54:52 -08:00
Catrope 662880605c (bug 42119) Handle alienation in wrapping mode properly
When alienating in wrapping mode, we need to look at the type of tag to
decide whether to create a wrapped alienInline, or to interrupt the
paragraph for an alienBlock.

This was being done just fine for the general alienation case
(unrecognized tag), but not for the special cases (mw:unrecognized,
about groups).

* Centralize the logic for ending a wrapper in stopWrapping()
* Move the wrapping-contingent block/inline detection logic into
  createAlien()
* Simplify the terrible if statement to decide whether a future decision
  requires us to stop wrapping. Instead, detect the cases in each code
  path separately and call stopWrapping() as appropriate
* Add tests

Change-Id: I4054584ae05e7d5daa71edead3e6a6588cf5d3bb
2012-11-21 13:42:13 -08:00
Trevor Parscal f71247608b Merge "Add moar demo pages" 2012-11-21 21:26:19 +00:00
Trevor Parscal e0e5e43303 Merge "Make entity nodes work in CE" 2012-11-21 21:25:03 +00:00
Catrope 8cf9193b39 Add moar demo pages
Change-Id: Id25803821ea0913f36a3c851c0542de12f7e408b
2012-11-21 13:24:38 -08:00
Catrope 0e1a6912b0 Display the lengths of model nodes in the demo debugging thingy
Also use .text() instead of .html() where appropriate

Change-Id: I9dcfd9c11c5c01f1459c48c8993c155203c94b13
2012-11-21 12:58:52 -08:00
Catrope 9f1eb9b991 Merge "(Bug 42335) Adding IE to the browser blacklist for the December release" 2012-11-21 20:33:17 +00:00
James D. Forrester 1780d4e610 (Bug 42335) Adding IE to the browser blacklist for the December release
Unfortunately support for Internet Explorer is currently insufficient in the CE module
which means we have to kill it for December; it will return once we've worked out a way
around various bugs in IE.

Change-Id: I0b44ae2c1d75ffe748a5139ca74dcda615e12a6a
2012-11-21 12:30:14 -08:00
Catrope 6230c322ec Merge "Fixed inspector behavior" 2012-11-21 20:23:35 +00:00
Catrope 0516b1d7f0 Make entity nodes work in CE
Check for all node classes in getOffsetFromTextNode(), not just
branches and aliens (an entity is neither)

Render entities with contenteditable=false. Without this, selection was
still broken, because:

Foo|€Bar was really <p>Foo<span>|€</span>Bar</p> which maps correctly.
Foo€|Bar was really <p>Foo<span>€|</span>Bar</p> which maps to the same,
which is incorrect.

With cE=false, the cursor can't be inside the span, so we get:
Foo|€Bar is really <p>Foo|<span>€</span>Bar</p> which maps correctly.
Foo€|Bar is really <p>Foo<span>€</span>|Bar</p> which maps correctly.

Change-Id: Iaf603346590a9ad553c152565eb203136be7a399
2012-11-21 12:04:31 -08:00
Trevor Parscal 8fc98868c9 Fixed inspector behavior
ve.ui.Inspector
* Removed disabled state and interfaces - this isn't needed
* Renamed prepareSelection to onInitialize
* Using event emitter to run onInitialize, onOpen and onClose methods
* Left removal up to the child class to handle in the onClose method
* Replaced calls on context to close inspector to calling close directly
* Renamed prepareSelection stub to onInitialize
* Emitting initialize event from within the open method
* Added recursion guarding to close method
* Changed the close method's argument to be remove instead of accept - the more common case is to save changes, and the only time you wouldn't save changes is if you were to remove the annotation
* Moved focus restore to close method

ve.ui.Context
* Moved the majority of the code in openInspector and closeInspector to event handlers for onInspectorOpen and onInspectorClose
* Updated calls to closeInspector re: accept->remove argument change

ve.ui.LinkInspector
* Renamed prepareSelection to onInitialize and rewrote logic and documentation
* Removed unused onLocationInputChange method
* Moved restore focus (now it's in the inspector base class)

ve.dm.SurfaceFragment
* Added word mode for expandRange

ve.dm.Surface
* Added locking/unlocking while processing transactions - this was not an issue before because this was effectively being done manually throughout ce (which needs to be cleaned up) but once we started using the content action to insert content dm and ce started playing off each other and inserting in a loop - we already do this for undo/redo so it makes sense to do it here as well

ve.InspectorAction
* Updated arguments re: close method's accept->remove argument change

Change-Id: I38995d4101fda71bfb2e6fe516603507ce820937
2012-11-21 12:01:14 -08:00
Trevor Parscal 53016c294b Merge "(bug 42218) Add MWEntityNode" 2012-11-21 00:20:26 +00:00
Catrope 1234a702c9 (bug 42218) Add MWEntityNode
<span typeof="mw:Entity"> tags are now correctly represented in the
model, and rendered in CE. There are still issues with cursor movement
etc. in CE.

Because the prioritization mechanism for annotations vs nodes is broken
in the current "node API", I had to hack two special cases for mw:Entity
into the converter. I also had to change the converter to ignore the
children of inline nodes (this was a legitimate bug, but had never come
up before).

Change-Id: Ib9f70437c58b4ca06aa09f7272bf51d9c41b18f2
2012-11-20 16:19:55 -08:00
raymond a54441611c Fix message key, probably c&p error in I1cc00844
Change-Id: I4400ceeda59d76e433cda61c546a196cb3400ba8
2012-11-20 22:09:41 +01:00
Translation updater bot a1a6121c9d Localisation updates from http://translatewiki.net.
Change-Id: Ie3c5ab02ac00f19f5f7b0bf2f65782eb748e990b
2012-11-20 20:55:23 +00:00
Catrope 3a047e0208 (bug 42124) Store comments in the meta-linmod
* Make converter generate meta nodes with 'style': 'comment'
* Handle style==='comment' in MetaBlockNode toDOM converter
* Add some comments to the meta test case
** Update other tests accordingly
* Change getDomElementSummary() to actually assert presence of comment
  nodes (specifically, all non-text child nodes)

Change-Id: Ieef9418f4c47df3541477d9420aa2ab8df6e3df1
2012-11-19 20:01:09 -08:00
Catrope fda2e6c1b5 (bug 42140) Don't force spaces to underscores in internal links
MWInternalLinkAnnotation was normalizing spaces to underscores. This is
bad. Instead, we now do the following:
* Normalize underscores to spaces for display purposes
* Store the original title without underscore/space mangling
* If the user didn't change the title (display title === original title
  with s/_/ /g), use the original title. Otherwise use the user's title
  verbatim, without normalizing either underscores or spaces.

Also, per a conversation with Gabriel, we now only restore hrefPrefix
when we're also restoring origTitle, otherwise Parsoid will barf.

Change-Id: Ia74a493b2bce96c9345b60ed692eeb2e43ebceff
2012-11-19 18:55:49 -08:00
Trevor Parscal 05e39c1733 Always apply inspected annotations to the right range
When you leave the inspector by changing the selection, we need to apply changes to the old selection.

ve.ui.Inspector
* Added initialSelection
* Change getMatchingAnnotations to use a given fragment rather than generating it's own
* Set initialSelection on open

ve.ui.Context
* Make hiding the context accept changes

ve.ui.LinkInspector
* Passing a fragment into getMatchingAnnotations now
* Using fragment API instead of actions API to control the range of the fragment

Change-Id: If6c8845285d87d0f144b15d50c38e192c797be59
2012-11-19 17:54:55 -08:00
Catrope a833691421 Merge "Fixing Pre-Annotations" 2012-11-20 01:11:04 +00:00
Catrope f856a10fa0 Merge "Added undo-before-apply for new link annotations" 2012-11-20 01:10:29 +00:00
Trevor Parscal 9c22ee346a Added undo-before-apply for new link annotations
When the link inspector is used to create a new annotation, the text is annotated with the default link target derived from the selected text. Then if the inspector is used to change that value, yet another transaction is processed when the inspector is closed.

To avoid having to press undo 2x, this change makes the inspector undo it's first change before applying the changed annotation.

This change also introduces insert, remove and select content actions.

Change-Id: I3e29189158fb01336d6b053bc2a8bda2a91a0a46
2012-11-19 17:10:05 -08:00
Christian Williams 9787166bab Fixing Pre-Annotations
AnnotationAction and SurfaceFragment now use insertAnnotations.

ve.dm.Surface.test
* Removed test for annotate method (not needed anymore)

ve.dm.SurfaceFragment
* Now using getInsertionAnnotations method
* Added support for modifying insertion annotations when annotating a zero-length selection

ve.dm.Surface
* Moved in insertion annotations state from document model
* Added insertion annotation interface (enable, disable, areEnabled, get, set, add, and remove)
* Simplified handling of annotations on change
* Removed annotate method (not used anymore)

ve.dm.Document
* Removed insertion annotations (moved it to surface model)

ve.ce.Surface
* Cleaned up handleInsertion and changed it to use the insertion annotations interface on the surface model

ve.AnnotationAction
* Moved insertion annotation handling out of here since it's now included in the surface fragment

Change-Id: I047d656acf7fa1c63f726ca2b0801e1476f84f96
2012-11-19 17:09:08 -08:00
Trevor Parscal 96fcbbd695 Cleanup var statements
Put non-assignment vars first

Change-Id: Id868976bc4d7c21032aef5020fe5fd0eb5df7a14
2012-11-19 17:05:34 -08:00
Catrope 22a01bfc4f Merge "The great inspector and context rewrite of 2012" 2012-11-19 23:23:48 +00:00
Trevor Parscal d2476a26d2 The great inspector and context rewrite of 2012
ve.AnnotationAction
* Added filter to the clearAll method to allow clearing all matching annotations only

ve.dm.Document
* Some variable renaming for consistency

ve.dm.SurfaceFragment
* Added truncateRange method
* Added annotation scope to expandRange method
* Added support for passing an annotation object into annotateContent method
* Switched to using name instead of type in annotateContent method to match the ve.dm.Annotation class
* Fixed logic in annotation mode of expandSelection so that expansion only takes place if the annotation is found

ve.ui.LinkInspector
* Moved most of the functionality elsewhere
* General reorganization
* Changed setOverlayPosition to accept 2 arguments instead of an object with 2 properties and renamed it to positionOverlayBelow
* Check for annotation object before extracting target information from it
* Initialize default target as empty string to avoid undefined being cast to a string and the default target becoming 'undefined'

icons.ai, inspector.png, inspector.svg
* Added generic inspector icon which will be used when a custom icon is not specified in future inspector subclasses

ve.ui.Inspector.Icons
* Added inspector icon
* Renamed clear icon to remove to match it's actual image

ve.ui.Context
* Greatly simplified the interface, reducing the number of methods by inlining a few things and combining others
* Now always listening to resize events on the window rather than only while document is focused
* Not listening to scroll events anymore, they used to affect the top/bottom positioning of the menu which we don't do anymore
* Lots of cleanup and reorganization
* No need to fallback to empty array since getInspectorsForAnnotations does so already
* Only consider fully-covered annotations for inspectors

ve.ui.Frame
* Simplified the constructor by introducing the createFrame method
* General cleanup
* Typo fixes

ve.ui.Inspector
* Generalized lots of functionality previously located in the link inspector class which will be useful to all inspectors (such as title, clear button, saving changes, etc.)
* Added setDisabled and isDisabled methods to manage CSS changes and avoid needing to check the CSS to determine the state of the inspector (storing state in the view is evil)
* Added getMatchingAnnotations method for convenience
* Added prepareSelection stub
* Lots of cleanup and documentation
* Type pattern is now defined in base class
* Added stubs for onOpen and onClose with documentation so that subclass authors know what these methods do
* Removed checks for onOpen or onClose methods since they are now noop stubs and are always there
* Added stub and removed checks for onRemove
* Made esc key close and accept - the illusion is supposed to be that the link changes are applied instantly, even though they are only updated when you close, so all closing except for when removing should apply changes - i.e. esc is now equal to back rather than being a special function that doesn't have an associated affordance
* Only consider fully-covered annotations when getting matching annotations

ve.ui.InspectorFactory
* Depending on type pattern now since it's always there
* Added getInspectorsForAnnotations method
* Return empty array if annotation set is empty

VisualEditor, VisualEditor.i18n
* Added default inspector message

Change-Id: I1cc008445bcbc8cba6754ca4b6ac0397575980d5
2012-11-19 15:21:27 -08:00
Catrope 79e4b139fb (bug 42212) Fix JS error when inserting after alien at the end
TransactionProcessor was using parentOuterRange without checking whether
it was present, so it was exploding for indexInNode results.

Now checking for parentOuterRange presence, and falling back to
nodeOuterRange when missing.

This fix causes inconsistencies with zero-length text nodes. We should
fix these eventually, but for now I've just made the unit tests
tolerant of zero-length-text-node deviations.

Change-Id: Id9eadd57a0d5fcbaf009c0781da0a03928aebb31
2012-11-19 14:35:30 -08:00
Translation updater bot 3f731b76e0 Localisation updates from http://translatewiki.net.
Change-Id: I0fd2a0761ab62d22a4c76a218a85e82eb8cd0696
2012-11-19 20:21:34 +00:00
Translation updater bot b06324bd2b Localisation updates from http://translatewiki.net.
Change-Id: I24b10d41bc7ffac229c3335724df8edaf169f797
2012-11-18 20:39:47 +00:00