Commit Graph

4282 Commits

Author SHA1 Message Date
David Rajchenbach-Teller
a50a3b6067 Bug 1224374 - Profiler labels for the 25 top chrome hangs;r=BenWa,MarcoZ
Experience shows that we do not have enough profiler labels to make
BHR hang reports meaningful. This patch adds enough labels to let us
exploit hang reports matching the 25 topmost chrome hangs.
2016-01-14 23:03:11 +01:00
Jonathan Kew
e24bc5e3c4 Bug 1239301 - Use 'ch' units rather than 'ex' units when specifying a width that is intended to fit a specific number of characters. r=tbsaunde 2016-01-18 13:27:59 +00:00
Eitan Isaacson
6259120162 Bug 1239051 - Labels should expose labeled controllers action. r=tbsaunde 2016-01-14 12:37:15 -08:00
Jonathan Kew
bf7dba2e00 Bug 1239007 - Replace nsIntRect by strongly-typed rects in a couple of nsCocoaUtils functions. r=mstange 2016-01-13 07:32:55 +00:00
Till Schneidereit
bace318903 Bug 1101817 - Part 1: Remove usages of WeakMap.prototype.clear from Gecko. r=yzen,mak,yoric,gijs,jlongster 2016-01-04 13:06:50 +01:00
Nathan Froyd
e61e5f85d7 Bug 1222531 - turn off -Wextra-tokens on clang-cl in accessible/ directories; r=tbsaunde
The Windows MIDL code generator creates things like:

    #endif !_MIDL_USE_GUIDDEF_

which clang-cl complains about.  MSVC doesn't, so turn this warning off.
2015-11-06 13:10:03 -05:00
Trevor Saunders
2571248bc6 bug 1235965 - null check mDoc in AccessibleWrap::Shutdown() r=davidb 2016-01-08 14:23:08 -05:00
Trevor Saunders
5c8b5951d4 bug 1203861 - use strncmp() instead of nsCString.equals() to compare names r=davidb
AtkObject stores the name as a C string, but nsCString can store strings with
'\0' in the middle.  That means that if we compute a name containing '\0'
nsCString.equals() will nevr consider that string to be equal to
AtkObject::name, however we only want to compare up to the first '\0' because
that is the only part Atk will ever see.  So we should use strncmp() instead of
nsCString.equals().
2016-01-08 14:23:08 -05:00
Trevor Saunders
23b3925fed bug 1228400 - null check tabChild before notifying the parent process about new child documents r=davidb 2016-01-07 14:18:40 -05:00
Nicholas Nethercote
18f37c43aa Bug 1232852 (part 3) - Remove unused parameters from some accessibility code. r=tbsaunde. 2015-12-15 21:24:06 -08:00
Bogdan Postelnicu
780de9b590 Bug 1235373 - Add an assert to check validity of pointers: mElement->GetPrimaryFrame() and frame. r=surkov 2015-12-28 06:25:00 -05:00
Alexander Surkov
41b5e92c36 Bug 1231919 - crash in mozilla::a11y::DocAccessible::ValidateARIAOwned, r=davidb 2015-12-14 12:02:23 -05:00
Jim Mathies
21003e17d6 Bug 1198459 - Prevent accessibility from initializing in content processes when e10s is running. r=tbsaunde 2015-12-10 14:28:33 -05:00
Carsten "Tomcat" Book
6af019a9b2 Backed out changeset bcb4ebf6ffac (bug 1198459) for bustage 2015-12-10 11:14:27 +01:00
Jim Mathies
cbfa845299 Bug 1198459 - Prevent accessibility from initializing in content processes when e10s is running. r=tbsaunde 2015-12-09 08:35:06 -05:00
Phil Ringnalda
1bcc4dfe4c Bug 845176 - Skip relations/test_embeds.xul on Linux debug and ASan, where it times out constantly 2015-12-09 18:48:26 -08:00
Mike Hommey
5cd413a92c Bug 1231148 - Avoid requiring IDL files installed in $DIST/idl to run AccEventGen.py. r=ted
While working on growing the fastermake backend to do compilation, I ran into
the problem that there's an implicit dependency from AccEventGen.py on the fact
that IDL files are installed in $DIST/idl. While we may need some way to
express such things in general, it turns out that for this specific use case,
we don't actually /need/ IDLs in $DIST/idl, and a couple paths are enough to
get things to work.
2015-12-09 19:23:53 +09:00
Ted Mielczarek
5ba367a270 bug 1160185 - support GENERATED_FILES in EXPORTS. r=glandium
This change allows specifying objdir-relative paths in EXPORTS to enable
exporting entries from GENERATED_FILES. Objdir paths in EXPORTS that are
not in GENERATED_FILES will raise an exception.

Example:
```
EXPORTS += ['!g.h', 'f.h']
GENERATED_FILES += ['g.h']
```

Given the implementation, this should also work for FINAL_TARGET_FILES,
FINAL_TARGET_PP_FILES, and TESTING_FILES, but those are not well-tested.

This patch also renames the install manifest for '_tests' to match the
directory name for convenience in some code I refactored.
2015-12-01 09:53:16 -05:00
Bogdan Postelnicu
1fb24586ed Bug 1230118 - added asserts on aOwner and aOwner->Elm() r=surkov 2015-12-03 14:07:43 -08:00
Wes Kocher
f2b9245dc2 Backed out 2 changesets (bug 1230118, bug 1230110) for windows debug assertions in browser_perf-categories-js-calltree.js
Backed out changeset a9ce85f77af9 (bug 1230118)
Backed out changeset b9d8d05aa4ee (bug 1230110)
2015-12-03 15:38:04 -08:00
Bogdan Postelnicu
181c843650 Bug 1230118 - added asserts on aOwner and aOwner->Elm() r=surkov 2015-12-03 14:07:43 -08:00
Robert O'Callahan
13afa55648 Bug 1221043. Revert to including trailing whitespace for accessibility APIs. r=marcoz,mats 2015-12-01 02:21:25 +13:00
Nicholas Nethercote
b8de9bd5b7 Bug 1187139 (part 3) - Replace nsBaseHashtable::Enumerate() calls in accessible/ with iterators. r=tbsaunde. 2015-11-29 14:02:07 -08:00
Nicholas Nethercote
994d40de16 Bug 1187139 (part 2) - Replace nsBaseHashtable::Enumerate() calls in accessible/ with iterators. r=tbsaunde. 2015-11-29 14:02:07 -08:00
Nicholas Nethercote
834769f331 Bug 1187139 (part 1) - Replace nsBaseHashtable::Enumerate() calls in accessible/ with iterators. r=tbsaunde. 2015-11-29 14:02:07 -08:00
Nicholas Nethercote
16965dfefb Bug 1225007 (part 2, attempt 3) - Use LayoutDevicePixel more in Cocoa widget code. r=kats. 2015-11-19 17:34:20 +11:00
Alexander Surkov
2f24aa0686 Bug 1213281 - crash in mozilla::a11y::DocAccessible::UpdateTreeOnInsertion, r=davidb 2015-11-26 20:46:12 -05:00
Alexander Surkov
2bc7f3842b Bug 1226000 - control may pick up its name from wrong label, r=yzen 2015-11-24 14:44:24 -05:00
Trevor Saunders
921ee26633 bug 1218762 - proxy ia2Accessible::scrollTo{,Point}() r=davidb 2015-11-20 09:45:23 -05:00
Phil Ringnalda
a8f7f7dcab Backed out 2 changesets (bug 1225007) for frequent OS X "Shouldn't return empty rect" assertion failures, a=backout
Backed out changeset c5b352c74b35 (bug 1225007)
Backed out changeset 8a513c70ce7c (bug 1225007)
2015-11-19 20:55:26 -08:00
Nicholas Nethercote
2d1e9b597b Bug 1225007 (part 2, attempt 2) - Use LayoutDevicePixel more in Cocoa widget code. r=kats. 2015-11-19 17:34:20 +11:00
Kyle Huey
4f7c071591 Bug 1222127: Use the inner window to compute style. r=baku 2015-11-19 14:47:20 +08:00
Nicholas Nethercote
38a8c624ec Backout 52ba24123046 and 66768d6a50fa (bug 1225007) for causing lots of test assertions on Mac. CLOSED TREE 2015-11-17 17:02:39 -08:00
Nicholas Nethercote
d9626aa742 Bug 1225007 (part 2) - Use LayoutDevicePixel more in Cocoa widget code. r=kats. 2015-11-16 19:40:53 -08:00
Chris Manchester
70e0305f5e Bug 1217015 - Convert AccEventGen.py to GENERATED_FILES and get rid of most of accessible/xpcom/Makefile.in r=glandium,tbsaunde 2015-11-16 16:51:51 -08:00
John Daggett
2b6e16d734 Bug 1173260 - stub out tests that fail when 'cursive' is looked up via fontconfig. r=m_kato 2015-11-16 08:48:45 +09:00
Nicholas Nethercote
a1d143774a Bug 1224403 (part 8) - Use LayoutDeviceIntMargin more in HyperTextAccessible. r=kats. 2015-11-12 05:34:27 -08:00
Nicholas Nethercote
b900f1cf11 Bug 1224403 (part 2) - Split GetClientOffset() into typed and untyped versions. r=kats. 2015-11-11 19:56:06 -08:00
Nicholas Nethercote
ffaa7a3203 Bug 1223310 (part 2) - Use LayoutDeviceIntRect for bounds-related functions in nsIWidget. r=kats.
The patch renames the existing functions (GetBounds(), GetClientBounds(), etc)
by adding an |Untyped| suffix. It then adds typed equivalents, and uses those
typed equivalents in all the call sites where it's easy to do so. The trickier
remaining call sites are converted to use the Untyped-suffix version.
2015-11-09 21:37:32 -08:00
Trevor Saunders
1788357882 bug 1213402 - add another , to fix silly string concatonation r=me landed on a CLOSED TREE 2015-11-10 21:45:12 -05:00
Trevor Saunders
b7e56809a3 bug 1213402 - fixup silly array , on windows r=me landing on a CLOSED TREE 2015-11-10 17:39:27 -05:00
Trevor Saunders
3829790593 bug 1213402 - Fire numeric value change events for proxies r=davidb 2015-11-10 15:55:09 -05:00
Trevor Saunders
d77a67806c bug 1213402 - separate value change events into text value changes and numeric value changes r=davidb
Arguably these are different things, and it will be easier to proxy events for
atk this way because atk only wants the numeric value changes.
2015-11-10 15:55:08 -05:00
Nicholas Nethercote
228a925ff0 Bug 1222943 (part 1) - Change Touch::mRadius from nsIntPoint to LayoutDeviceIntPoint. r=kats.
This adds a three missing unit conversions for touch radii.
2015-11-09 21:37:31 -08:00
Alexander Surkov
88114b1385 Bug 1221542, bug 1221543 - crash in mozilla::a11y::DocAccessible::SeizeChild/PutChildrenBack, r=davidb 2015-11-09 14:47:38 -05:00
Justin Obara
0c6a085e6c Bug 1169019 - Removed screenreader announcement from gecko. r=yzen 2015-11-08 21:24:41 -08:00
Alexander Surkov
00ba473e42 Bug 1220897 - fix IAccessible2::get_accessibleWithCaret, r=tbsaunde 2015-11-06 23:31:12 -05:00
Trevor Saunders
1def7d87c4 bug 1221551 - unbind child documents before rebinding them r=davidb
It seems possible that a child document could be bound to a parent document
without first being unbound.  Some evidence for this being possible is that
OuterDocAccessible::InsertChildAt() explicitly checks for this and does
something similar if it already has a child.
2015-11-06 17:29:17 -05:00
Nathan Froyd
ed4a1b85f3 Bug 1222174 - add override declarations in AccessibleWrap.h; r=tbsaunde
clang-cl complains about not having these otherwise.
2015-11-05 15:20:09 -05:00
Alexander Surkov
51e1c0fdcc Bug 582024 - ARIA active-descendant should work for ARIA owned elements, r=yzen 2015-11-03 11:03:34 -05:00