5849e67a8dbab1086a2b5e2b9a569f446cb15995: Bug 1208489 - use resizebefore attribute on devtools horizontal splitter to avoid resizing notificationboxes, r=bgrins a=lizzard
Gijs Kruitbosch <gijskruitbosch@gmail.com> - Thu, 07 Jul 2016 16:38:27 +0100 - rev 391774
Push 23905 by mozilla@buttercookie.de at Fri, 22 Jul 2016 19:26:20 +0000
Bug 1208489 - use resizebefore attribute on devtools horizontal splitter to avoid resizing notificationboxes, r=bgrins a=lizzard MozReview-Commit-ID: 9ijDB4fFhA3
392198ab803237ab8746c9c2e737c252f05e98a7: Bug 1241892 - Make View Source work in the sidebar browser. r=jryans a=gchang
Mike Conley <mconley@mozilla.com> - Mon, 04 Jul 2016 15:44:09 -0400 - rev 391773
Push 23905 by mozilla@buttercookie.de at Fri, 22 Jul 2016 19:26:20 +0000
Bug 1241892 - Make View Source work in the sidebar browser. r=jryans a=gchang The sidebar browser is always non-remote, so when we open a new tab to open view source for it, we should ensure that that tab is also non-remote. MozReview-Commit-ID: 3pH9okO3Boz
120008d520c0b92d3fb50e9908a362ac6efc866e: Bug 1241892 - Make Page Info work for sidebar browser. r=florian a=gchang
Mike Conley <mconley@mozilla.com> - Mon, 04 Jul 2016 15:42:45 -0400 - rev 391772
Push 23905 by mozilla@buttercookie.de at Fri, 22 Jul 2016 19:26:20 +0000
Bug 1241892 - Make Page Info work for sidebar browser. r=florian a=gchang The modifications that were made in bug 1238180 assumed that gBrowser.selectedBrowser from the opening window context would be defined. This is true in normal browser tabs, but not so much in the sidebar browser. This patch makes it so that callers can supply a browser to override the assumption. MozReview-Commit-ID: FKSyl7WUu1V
7f4d9e1ed33f09ad8c68856f539d9c65a9013a2c: Bug 1261869 - Fix leaks in devtools; r=ejpbruel a=gchang
Nick Fitzgerald <fitzgen@gmail.com> - Wed, 13 Jul 2016 15:19:23 -0700 - rev 391771
Push 23905 by mozilla@buttercookie.de at Fri, 22 Jul 2016 19:26:20 +0000
Bug 1261869 - Fix leaks in devtools; r=ejpbruel a=gchang There are two leaks addressed in this commit: 1. The thread actor's `_debuggerSourcesSeen` set was never cleared. This set exists only as a performance optimization to speed up `_addSource` in cases where we've already added the source. Unfortunately, this set wasn't getting cleared when we cleared debuggees out and it ended up keeping the `Debugger.Source`, its referent, and transitively its referent's global alive. I figured it was simpler to make it a `WeakSet` than to add it as a special case in `ThreadActor.prototype._clearDebuggees` and manage the lifetimes by hand. I think this fits well with its intended use as an ephemeral performance optimization. 2. Due to a logic error, we were not clearing debuggees in the memory actor's `Debugger` instance on navigations. This isn't really a "proper" leak, in that if you forced a GC, the old debuggees would go away as `Debugger` holds them weakly, however if there was no GC between navigations, then you could still see the old windows (and everything they "retained") as roots in the snapshot. This issue is straightforward to fix once identified: ensure that `_clearDebuggees` is actually called on navigation. Finally, this commit adds a test that we don't leak Window objects when devtools are open and we keep refreshing a tab. When it fails, it prints out the leaking window's retaining paths.
39f61988e8fb81841b557f8b94d8aa4a5c0bf27c: Bug 1264809 - (Part 3) Add reftest. r=dholbert a=lizzard
Kevin Chen <kechen@mozilla.com> - Fri, 01 Jul 2016 00:44:00 +0200 - rev 391770
Push 23905 by mozilla@buttercookie.de at Fri, 22 Jul 2016 19:26:20 +0000
Bug 1264809 - (Part 3) Add reftest. r=dholbert a=lizzard
9b3ee31983dcdcd9884f5ba9c971fd0008f3d8f5: Bug 1264809 - (Part 2) Preserve scale ratio if SVG dosen't have viewport size and viewBox. r=dholbert a=lizzard
Kevin Chen <kechen@mozilla.com> - Wed, 13 Jul 2016 15:18:30 -0700 - rev 391769
Push 23905 by mozilla@buttercookie.de at Fri, 22 Jul 2016 19:26:20 +0000
Bug 1264809 - (Part 2) Preserve scale ratio if SVG dosen't have viewport size and viewBox. r=dholbert a=lizzard
d183d71a7545d5f6a724d9ec27ca55fbd335e250: Bug 1264809 - (Part 1) Decouple SetImageOverridePreserveAspectRatio from mIsPaintingSVGImageElement setting. r=dholbert a=lizzard
Kevin Chen <kechen@mozilla.com> - Fri, 01 Jul 2016 00:41:00 +0200 - rev 391768
Push 23905 by mozilla@buttercookie.de at Fri, 22 Jul 2016 19:26:20 +0000
Bug 1264809 - (Part 1) Decouple SetImageOverridePreserveAspectRatio from mIsPaintingSVGImageElement setting. r=dholbert a=lizzard
9ff2ed0592d83e5e68ad32c27172291fc7eeb1c8: Bug 1278644 - Stable getItemID implementation for ClientsAdapter r=sebastian a=lizzard
Grigory Kruglov <gkruglov@mozilla.com> - Wed, 13 Jul 2016 15:14:50 -0700 - rev 391767
Push 23905 by mozilla@buttercookie.de at Fri, 22 Jul 2016 19:26:20 +0000
Bug 1278644 - Stable getItemID implementation for ClientsAdapter r=sebastian a=lizzard
bd3207c8ef3811b55e2073248c6ca39576fcf3a6: Bug 1282577 - Guard against multiple intr messages causes us to multi-delete GMP actors. r=jesup,a=lizzard
Chris Pearce <cpearce@mozilla.com> - Mon, 11 Jul 2016 21:52:31 +1200 - rev 391766
Push 23905 by mozilla@buttercookie.de at Fri, 22 Jul 2016 19:26:20 +0000
Bug 1282577 - Guard against multiple intr messages causes us to multi-delete GMP actors. r=jesup,a=lizzard This is a regression from bug 1162358. We must be hitting the #ifndef SHMEM_ALLOC_IN_CHILD block in GMPVideoDecoderChild::Alloc() with multiple allocs doing intr calls at once. If this happens when a DecodingComplete() comes in, we'll end up sending one task to re-call RecvDecodingComplete for every Alloc() blocked on an intr response. This would result in us ending up trying to Send__delete__() in RecvDecodingComplete() twice. Which causes the runtime abort we're seeing here. I think that could happen in the WidevineVideoDecoder if a Decode message comes in, and goes into a ReturnOutput(), tries to alloc a frame and has to spin on an intr message response, and another Decode message comes in and does the same, so GMPVideoDecoderChild::mNeedShmemIntrCount will be 2, and then a DecodingComplete comes in, and when two tasks on the stack in GMPVideoDecoderChild::Alloc() finish they both end up dispatching a task to re-call GMPVideoDecoderChild::RecvDecodingComplete(). So we end up trying to Send__delete__() in RecvDecodingComplete() twice. I expect the same problem exists in the GMPVideoEncoder too. intr, or spinning event loops in general for that matter, is evil. MozReview-Commit-ID: AKsvP62G3Cx
1609e0acfe66720db5ac91b5cb5748f3c27a9468: Bug 1285240: Update tests to prevent redefinitions r=ato a=test-only
David Burns <dburns@mozilla.com> - Thu, 07 Jul 2016 22:20:32 +0100 - rev 391765
Push 23905 by mozilla@buttercookie.de at Fri, 22 Jul 2016 19:26:20 +0000
Bug 1285240: Update tests to prevent redefinitions r=ato a=test-only The tests had the same name which caused a redefinition meaning only one of the tests was run. MozReview-Commit-ID: LuflhcD4wKl
f3111d767fd777339107a8f7f0bea5e5630cbf6c: Bug 1285240: Trim leading and trailing space when finding elements by link text r=ato a=test-only
David Burns <dburns@mozilla.com> - Thu, 07 Jul 2016 23:06:17 +0100 - rev 391764
Push 23905 by mozilla@buttercookie.de at Fri, 22 Jul 2016 19:26:20 +0000
Bug 1285240: Trim leading and trailing space when finding elements by link text r=ato a=test-only This gives up parity with the WebDriver project and brings us in line with the current specification http://w3c.github.io/webdriver/webdriver-spec.html#dfn-link-text MozReview-Commit-ID: 9OD6B2pKTfx
fa4681fe062f30712caef3b2724bd3ac71862468: Bug 1283999 - Publish XPath find element functions; r=automatedtester a=test-only
Andreas Tolfsen <ato@mozilla.com> - Sat, 02 Jul 2016 21:51:21 +0100 - rev 391763
Push 23905 by mozilla@buttercookie.de at Fri, 22 Jul 2016 19:26:20 +0000
Bug 1283999 - Publish XPath find element functions; r=automatedtester a=test-only Publish the findByXPath and findAllByXPath functions to match the style of the recently introduces findByLinkText and findByPartialLinkText functions. Do note that this patch rearranges the input arguments these functions take. Possibly we do not have to send the root element as input at all. MozReview-Commit-ID: 4H7kpZBUsTw
42fc9e1652199459ddb667b1fb0b541602550393: Bug 1283999 - Fix lookup of hyperlinks in XHTML documents; r=automatedtester a=test-only
Andreas Tolfsen <ato@mozilla.com> - Sat, 02 Jul 2016 21:38:48 +0100 - rev 391762
Push 23905 by mozilla@buttercookie.de at Fri, 22 Jul 2016 19:26:20 +0000
Bug 1283999 - Fix lookup of hyperlinks in XHTML documents; r=automatedtester a=test-only Lower-case "a" matches hyperlinks in XHTML documents as well as HTML documents. Upper-case "A" only matches HTML documents. The patch also refactors link text- and partial link text lookup into distinct functions, so that there is no more worry about variable scoping in match blocks. MozReview-Commit-ID: FB7MAmosBoR
86b1691ed273fa8de4373b35e374a8be087b8214: Bug 1283999 - Run element retrieval tests on XHTML documents; r=automatedtester a=test-only
Andreas Tolfsen <ato@mozilla.com> - Sat, 02 Jul 2016 21:38:01 +0100 - rev 391761
Push 23905 by mozilla@buttercookie.de at Fri, 22 Jul 2016 19:26:20 +0000
Bug 1283999 - Run element retrieval tests on XHTML documents; r=automatedtester a=test-only MozReview-Commit-ID: 5dhnqsZhysn
1c7797992d06ab99d1dc2c4c6d128d8dc4aedc65: Bug 1283999 - Rename element retrieval tests to match spec; r=automatedtester a=test-only
Andreas Tolfsen <ato@mozilla.com> - Sat, 02 Jul 2016 18:28:27 +0100 - rev 391760
Push 23905 by mozilla@buttercookie.de at Fri, 22 Jul 2016 19:26:20 +0000
Bug 1283999 - Rename element retrieval tests to match spec; r=automatedtester a=test-only The test_findelement.py name is misleading because it covers tests for commands Find Element, Find Elements, Find Element From Element, Find Elements From Element, and Get Active Element. MozReview-Commit-ID: J9UyDewDE7N
74e83213e48a6b8bf7c14448b0a02c1577224968: Bug 1283999 - Rewrite and complement element retrieval tests; r=automatedtester a=test-only
Andreas Tolfsen <ato@mozilla.com> - Sat, 02 Jul 2016 18:26:17 +0100 - rev 391759
Push 23905 by mozilla@buttercookie.de at Fri, 22 Jul 2016 19:26:20 +0000
Bug 1283999 - Rewrite and complement element retrieval tests; r=automatedtester a=test-only Many tests for the plural Find Element, Find Element From Element, and FInd Elements From Element commands were missing. MozReview-Commit-ID: DFGJYY8rkqT
50dfed915c24291922414771d3622c336d28b87e: Bug 1275917 - 1st transaction on dontresused h2 session. r=hurley, a=lizzard
Patrick McManus <mcmanus@ducksong.com> - Tue, 05 Jul 2016 12:33:34 -0400 - rev 391758
Push 23905 by mozilla@buttercookie.de at Fri, 22 Jul 2016 19:26:20 +0000
Bug 1275917 - 1st transaction on dontresused h2 session. r=hurley, a=lizzard
385340f31f4d6bbc00b1a0f8c59d583ce3f8d4e3: Bug 1255555 - When a frame reconstruction triggers caret elements to be cloned, ensure the dummy touch listeners are re-registered on the clones. r=tylin a=gchang
Kartikaya Gupta <kgupta@mozilla.com> - Mon, 11 Jul 2016 08:16:39 -0400 - rev 391757
Push 23905 by mozilla@buttercookie.de at Fri, 22 Jul 2016 19:26:20 +0000
Bug 1255555 - When a frame reconstruction triggers caret elements to be cloned, ensure the dummy touch listeners are re-registered on the clones. r=tylin a=gchang MozReview-Commit-ID: 8txWmd64LC9
49c2e05e529086767b4e27006630107193b8b09c: Backed out changeset de766fceaddc (bug 1255555) for bustage
Carsten "Tomcat" Book <cbook@mozilla.com> - Tue, 12 Jul 2016 15:11:54 +0200 - rev 391756
Push 23905 by mozilla@buttercookie.de at Fri, 22 Jul 2016 19:26:20 +0000
Backed out changeset de766fceaddc (bug 1255555) for bustage
16de5bb4010b55a8bb645922512d3d4d5ded5db6: Bug 1282246 - don't use gradients with non-finite endpoints in DrawTargetSkia. r=mchang, a=lizzard
Lee Salzman <lsalzman@mozilla.com> - Wed, 29 Jun 2016 20:01:48 -0400 - rev 391755
Push 23905 by mozilla@buttercookie.de at Fri, 22 Jul 2016 19:26:20 +0000
Bug 1282246 - don't use gradients with non-finite endpoints in DrawTargetSkia. r=mchang, a=lizzard
(0) -300000 -100000 -30000 -10000 -3000 -1000 -300 -100 -50 -20 +20 +50 +100 +300 +1000 +3000 +10000 +30000 +100000 +300000 tip