searching for reviewer(matspal)
74f132824fdaee3b9d3ffaa88a072d4594559c7b: Bug 1005588. Fix getBoxQuads test to work when the feature prefs are disabled by default. r=matspal
Robert O'Callahan <robert@ocallahan.org> - Tue, 03 Jun 2014 13:54:20 +1200 - rev 187044
Push 26909 by emorley@mozilla.com at Fri, 06 Jun 2014 15:55:30 +0000
Bug 1005588. Fix getBoxQuads test to work when the feature prefs are disabled by default. r=matspal
f7d67e4e0aa670c473182f551d599079e10020b5: Bug 1014252. Optimize clearing of textruns via RemoveInFlows. r=matspal
Robert O'Callahan <robert@ocallahan.org> - Wed, 28 May 2014 15:39:25 +1200 - rev 185234
Push 26846 by cbook@mozilla.com at Wed, 28 May 2014 12:22:26 +0000
Bug 1014252. Optimize clearing of textruns via RemoveInFlows. r=matspal We're keeping the core idea that, before we remove the frames-to-be-destroyed from the continuation chain, their textruns need to be disconnected/destroyed. However, nsContinuingTextFrame::DestroyFrom tries to optimize when the destroying frames that aren't mentioned in the userdata for the textrun, and certain other conditions are met; we need a similar optimization here. It's simpler here because the other conditions are definitely met, since all the text for the frames being deleted has already been consumed and reflowed by previous frames. We don't need the TEXT_STYLE_MATCHES_PREV_CONTINUATION state bit anymore because nsContinuingTextFrame::DestroyFrom will never see any textruns when called via RemoveEmptyInFlows.
9ae5146d1267816f241a9eeae589a57a3f53ab0f: Bug 917755. Part 6: Add tests for getBoxQuads. r=matspal
Robert O'Callahan <robert@ocallahan.org> - Fri, 20 Sep 2013 22:21:42 +1200 - rev 174674
Push 26458 by cbook@mozilla.com at Fri, 21 Mar 2014 14:15:06 +0000
Bug 917755. Part 6: Add tests for getBoxQuads. r=matspal
d5ffda84968f25999862bb2af49ced96ed719c0a: Bug 917755. Part 5: Implement GeometryUtils GetBoxQuads. r=matspal
Robert O'Callahan <robert@ocallahan.org> - Wed, 12 Mar 2014 09:11:38 +0800 - rev 174672
Push 26458 by cbook@mozilla.com at Fri, 21 Mar 2014 14:15:06 +0000
Bug 917755. Part 5: Implement GeometryUtils GetBoxQuads. r=matspal
4388c77d8dabb5ccfe35655e3bf994a0d3791eb6: Bug 917755. Part 1.5: Make nsImageLoadingContent more robust in unified builds. r=matspal
Robert O'Callahan <robert@ocallahan.org> - Wed, 12 Mar 2014 09:11:37 +0800 - rev 174668
Push 26458 by cbook@mozilla.com at Fri, 21 Mar 2014 14:15:06 +0000
Bug 917755. Part 1.5: Make nsImageLoadingContent more robust in unified builds. r=matspal
9b73b517cb535b8d6a9daf83d83430d06017d596: Bug 917755. Part 1: Add nsLayoutUtils::TransformCSSPoints and nsLayoutUtils::GetFirstNonAnonymousFrame. r=matspal
Robert O'Callahan <robert@ocallahan.org> - Fri, 20 Sep 2013 22:21:03 +1200 - rev 174667
Push 26458 by cbook@mozilla.com at Fri, 21 Mar 2014 14:15:06 +0000
Bug 917755. Part 1: Add nsLayoutUtils::TransformCSSPoints and nsLayoutUtils::GetFirstNonAnonymousFrame. r=matspal
98e31d225a5a6773672b94986762e33d66184e7e: Bug 917755. Part 6: Add tests for getBoxQuads. r=matspal
Robert O'Callahan <robert@ocallahan.org> - Fri, 20 Sep 2013 22:21:42 +1200 - rev 174631
Push 26458 by cbook@mozilla.com at Fri, 21 Mar 2014 14:15:06 +0000
Bug 917755. Part 6: Add tests for getBoxQuads. r=matspal
5117e3f594e7ddcc10138074d8fd713126d8caf4: Bug 917755. Part 5: Implement GeometryUtils GetBoxQuads. r=matspal
Robert O'Callahan <robert@ocallahan.org> - Wed, 12 Mar 2014 09:11:38 +0800 - rev 174629
Push 26458 by cbook@mozilla.com at Fri, 21 Mar 2014 14:15:06 +0000
Bug 917755. Part 5: Implement GeometryUtils GetBoxQuads. r=matspal
b049571a7cce602eabe119cdc24a97cfa36ac1c9: Bug 917755. Part 1.5: Make nsImageLoadingContent more robust in unified builds. r=matspal
Robert O'Callahan <robert@ocallahan.org> - Wed, 12 Mar 2014 09:11:37 +0800 - rev 174625
Push 26458 by cbook@mozilla.com at Fri, 21 Mar 2014 14:15:06 +0000
Bug 917755. Part 1.5: Make nsImageLoadingContent more robust in unified builds. r=matspal
472cb7738e14e4e04c698cface9e27c0ec2722a4: Bug 917755. Part 1: Add nsLayoutUtils::TransformCSSPoints and nsLayoutUtils::GetFirstNonAnonymousFrame. r=matspal
Robert O'Callahan <robert@ocallahan.org> - Fri, 20 Sep 2013 22:21:03 +1200 - rev 174624
Push 26458 by cbook@mozilla.com at Fri, 21 Mar 2014 14:15:06 +0000
Bug 917755. Part 1: Add nsLayoutUtils::TransformCSSPoints and nsLayoutUtils::GetFirstNonAnonymousFrame. r=matspal
1b7530ddb0afc45c5323ec8cd5cdef9b30a9b90a: Bug 917755. Part 6: Add tests for getBoxQuads. r=matspal
Robert O'Callahan <robert@ocallahan.org> - Fri, 20 Sep 2013 22:21:42 +1200 - rev 174246
Push 26446 by cbook@mozilla.com at Wed, 19 Mar 2014 15:08:44 +0000
Bug 917755. Part 6: Add tests for getBoxQuads. r=matspal
4771c2b309ad3389e96de6960e5eb315fb4e0a06: Bug 917755. Part 5: Implement GeometryUtils GetBoxQuads. r=matspal
Robert O'Callahan <robert@ocallahan.org> - Wed, 12 Mar 2014 09:11:38 +0800 - rev 174244
Push 26446 by cbook@mozilla.com at Wed, 19 Mar 2014 15:08:44 +0000
Bug 917755. Part 5: Implement GeometryUtils GetBoxQuads. r=matspal
002dbc057a74f3e37641388d4b5dd3a05e4696b6: Bug 917755. Part 1.5: Make nsImageLoadingContent more robust in unified builds. r=matspal
Robert O'Callahan <robert@ocallahan.org> - Wed, 12 Mar 2014 09:11:37 +0800 - rev 174240
Push 26446 by cbook@mozilla.com at Wed, 19 Mar 2014 15:08:44 +0000
Bug 917755. Part 1.5: Make nsImageLoadingContent more robust in unified builds. r=matspal
094decae8ff3c7be54f7302dfe29c8beb653f7a4: Bug 917755. Part 1: Add nsLayoutUtils::TransformCSSPoints and nsLayoutUtils::GetFirstNonAnonymousFrame. r=matspal
Robert O'Callahan <robert@ocallahan.org> - Fri, 20 Sep 2013 22:21:03 +1200 - rev 174239
Push 26446 by cbook@mozilla.com at Wed, 19 Mar 2014 15:08:44 +0000
Bug 917755. Part 1: Add nsLayoutUtils::TransformCSSPoints and nsLayoutUtils::GetFirstNonAnonymousFrame. r=matspal
6a469c5a27699870a136b84ca82e8be898749a16: Bug 975931. Part 10: Refactor nsDOMWindowUtils to use a shared GetDocument method. r=matspal
Robert O'Callahan <robert@ocallahan.org> - Sat, 08 Mar 2014 22:22:28 +1300 - rev 172767
Push 26380 by ryanvm@gmail.com at Mon, 10 Mar 2014 20:53:14 +0000
Bug 975931. Part 10: Refactor nsDOMWindowUtils to use a shared GetDocument method. r=matspal
5e992c8ea14f9146e31e6b684fc2f79431c552dd: Bug 944291. Part 2: Mark parent frames whose child frames' overflow have changed as always needing UpdateOverflow called. r=matspal
Robert O'Callahan <robert@ocallahan.org> - Sat, 14 Dec 2013 02:21:03 +1300 - rev 162160
Push 25936 by emorley@mozilla.com at Mon, 06 Jan 2014 12:52:01 +0000
Bug 944291. Part 2: Mark parent frames whose child frames' overflow have changed as always needing UpdateOverflow called. r=matspal
ca4c295c94542e09918bec9884f34c20b2f2f6a7: Bug 944198 - Only update the color if we're still a color input. r=bz, r=matspal
Mats Palmgren <matspal@gmail.com> - Mon, 02 Dec 2013 13:39:32 -0500 - rev 158355
Push 25741 by ryanvm@gmail.com at Mon, 02 Dec 2013 21:39:06 +0000
Bug 944198 - Only update the color if we're still a color input. r=bz, r=matspal
9bd43aa9a0186b56df71fa5c73736789f28d17a6: Bug 919129. Fix nsLayoutUtils::UpdateImageVisibilityForFrame so that it doesn't expand the image rect so much by replacing it with the scroll port rect. r=matspal
Timothy Nikkel <tnikkel@gmail.com> - Sat, 21 Sep 2013 16:20:22 -0500 - rev 148212
Push 25330 by philringnalda@gmail.com at Sun, 22 Sep 2013 16:14:09 +0000
Bug 919129. Fix nsLayoutUtils::UpdateImageVisibilityForFrame so that it doesn't expand the image rect so much by replacing it with the scroll port rect. r=matspal
329f3f03ee1c5a0c8f59b1a21c8da82539ca2e6e: Bug 722923. Part 4: Don't walk continuation chain in SyncViewsAndInvalidateDescendants since DoApplyRenderingChangeToTree already does that. r=matspal
Robert O'Callahan <robert@ocallahan.org> - Sat, 06 Jul 2013 23:53:42 +1200 - rev 137563
Push 24927 by ryanvm@gmail.com at Mon, 08 Jul 2013 01:19:31 +0000
Bug 722923. Part 4: Don't walk continuation chain in SyncViewsAndInvalidateDescendants since DoApplyRenderingChangeToTree already does that. r=matspal
d3a1e0e44245e55cd146b6a6d5043e64b9822a16: Bug 722923. Part 3: Rename UpdateViewsForFrame and document it properly. r=matspal
Robert O'Callahan <robert@ocallahan.org> - Sat, 06 Jul 2013 23:53:40 +1200 - rev 137562
Push 24927 by ryanvm@gmail.com at Mon, 08 Jul 2013 01:19:31 +0000
Bug 722923. Part 3: Rename UpdateViewsForFrame and document it properly. r=matspal
929fcbfd5618dd5f5ef3355f9e86190d75344da1: Bug 841363. Ensure padding-rect edges are included in scrollable area used to compute scrollWidth/scrollHeight for overflow:visible elements. r=matspal
Robert O'Callahan <robert@ocallahan.org> - Tue, 09 Apr 2013 12:05:12 +1200 - rev 129052
Push 24553 by ryanvm@gmail.com at Wed, 17 Apr 2013 16:44:13 +0000
Bug 841363. Ensure padding-rect edges are included in scrollable area used to compute scrollWidth/scrollHeight for overflow:visible elements. r=matspal
64abee9b00f1b03f6b5d735f44628e93e63a73c4: Bug 841363. Ensure padding-rect edges are included in scrollable area used to compute scrollWidth/scrollHeight for overflow:visible elements. r=matspal
Robert O'Callahan <robert@ocallahan.org> - Tue, 09 Apr 2013 12:05:12 +1200 - rev 128893
Push 24545 by ryanvm@gmail.com at Tue, 16 Apr 2013 13:13:04 +0000
Bug 841363. Ensure padding-rect edges are included in scrollable area used to compute scrollWidth/scrollHeight for overflow:visible elements. r=matspal
b3db97a3cbb7251994c9d87c9a692f5d78f05e39: Bug 846772 - Fuzz/un-fuzz tests for the dynamic toolbar on Android. r=jwatt,matspal
Chris Lord <chrislord.net@gmail.com> - Tue, 12 Mar 2013 18:32:26 +0000 - rev 124585
Push 24427 by emorley@mozilla.com at Wed, 13 Mar 2013 12:28:55 +0000
Bug 846772 - Fuzz/un-fuzz tests for the dynamic toolbar on Android. r=jwatt,matspal The dynamic toolbar enlarges the browser slightly, which causes extra fuzzing to be required on dynamic-clipPath-01.svg and scroll-rounding layout tests, but also makes translate-pattern-1.sg pass.
0661df360cc2fabc889b42e63fc48bdf870886af: Bug 846772 - Fuzz dynamic-clipPath-01 and scroll-rounding tests for Android. r=jwatt,matspal
Chris Lord <chrislord.net@gmail.com> - Tue, 12 Mar 2013 18:32:26 +0000 - rev 124532
Push 24427 by emorley@mozilla.com at Wed, 13 Mar 2013 12:28:55 +0000
Bug 846772 - Fuzz dynamic-clipPath-01 and scroll-rounding tests for Android. r=jwatt,matspal These tests need extra fuzzing when the dynamic toolbar is enabled.
ffe1610a971a61fe299dd38dde35ce91c30a8433: Bug 684622 - Print scaling and shrink to fit page width does not work from page 2. r=matspal
Steven Michaud <smichaud@pobox.com> - Fri, 08 Feb 2013 19:14:30 -0600 - rev 121414
Push 24284 by ryanvm@gmail.com at Sat, 09 Feb 2013 15:43:50 +0000
Bug 684622 - Print scaling and shrink to fit page width does not work from page 2. r=matspal
7cd51e4e7d09a2a46a4051ca78997da643b921b4: Bug 818276. Update margins and scrollbar settings for a child document whenever the corresponding IFRAME state changes, even if we do not recreate the presentation. r=matspal
Robert O'Callahan <robert@ocallahan.org> - Fri, 07 Dec 2012 12:58:15 +1300 - rev 115343
Push 24003 by eakhgari@mozilla.com at Sun, 09 Dec 2012 18:17:18 +0000
Bug 818276. Update margins and scrollbar settings for a child document whenever the corresponding IFRAME state changes, even if we do not recreate the presentation. r=matspal
9d5f464dc1c011439ef075789c6b3df2c7ceb257: Bug 703241. Part 3: nsLayoutUtils::GetTransformToAncestor should handle the case where aAncestor == aFrame. r=matspal
Robert O'Callahan <robert@ocallahan.org> - Wed, 17 Oct 2012 16:13:24 +1300 - rev 110648
Push 23704 by emorley@mozilla.com at Thu, 18 Oct 2012 17:12:58 +0000
Bug 703241. Part 3: nsLayoutUtils::GetTransformToAncestor should handle the case where aAncestor == aFrame. r=matspal
4033a060e7af1e2ae99194e0c489d3788ba2f944: Bug 703241. Part 4: Don't let 'frame' be treated as its own container in PresShell::DoScrollContentIntoView. r=matspal
Robert O'Callahan <robert@ocallahan.org> - Wed, 17 Oct 2012 17:44:06 +1300 - rev 110647
Push 23704 by emorley@mozilla.com at Thu, 18 Oct 2012 17:12:58 +0000
Bug 703241. Part 4: Don't let 'frame' be treated as its own container in PresShell::DoScrollContentIntoView. r=matspal
fbf63fe2c1b0dc22f30c337ab3c78ee4e71435a7: Bug 703241. PresShell::DoScrollContentIntoView and PresShell::ScrollFrameRectIntoView should take transforms into account. r=matspal
Robert O'Callahan <robert@ocallahan.org> - Fri, 22 Jun 2012 17:06:49 +1200 - rev 110645
Push 23704 by emorley@mozilla.com at Thu, 18 Oct 2012 17:12:58 +0000
Bug 703241. PresShell::DoScrollContentIntoView and PresShell::ScrollFrameRectIntoView should take transforms into account. r=matspal
0c8ac346e41bac80534be22964967550785a0607: Bug 799078. Part 1: Expose nsContentUtils::FindPresShellForDocument. r=matspal
Robert O'Callahan <robert@ocallahan.org> - Tue, 09 Oct 2012 00:23:11 +1300 - rev 110253
Push 23680 by emorley@mozilla.com at Tue, 16 Oct 2012 08:09:24 +0000
Bug 799078. Part 1: Expose nsContentUtils::FindPresShellForDocument. r=matspal
b78108ed9da38e40bb935fe37f79e348e9b40da1: Bug 626245. Part 5: Move plugin widgets before compositing the window, not after. r=matspal
Robert O'Callahan <robert@ocallahan.org> - Wed, 10 Oct 2012 23:25:57 +1300 - rev 110192
Push 23677 by ryanvm@gmail.com at Sun, 14 Oct 2012 20:39:34 +0000
Bug 626245. Part 5: Move plugin widgets before compositing the window, not after. r=matspal
eed1630b856f7857a8bc91d4979d0c852125fc90: Bug 626245. Part 4.5: Forward WillPaintWindow/DidPaintWindow notifications to the presshell and do plugin geometry updates from there. r=matspal
Robert O'Callahan <robert@ocallahan.org> - Wed, 10 Oct 2012 23:25:57 +1300 - rev 110191
Push 23677 by ryanvm@gmail.com at Sun, 14 Oct 2012 20:39:34 +0000
Bug 626245. Part 4.5: Forward WillPaintWindow/DidPaintWindow notifications to the presshell and do plugin geometry updates from there. r=matspal
9a5b9cc9b42c30a6b4e9e48a5a03c7800b364678: Bug 626245. Part 3: Make aAncestor const in TransformFrameRectToAncestor. r=matspal
Robert O'Callahan <robert@ocallahan.org> - Mon, 08 Oct 2012 17:52:52 +1300 - rev 110188
Push 23677 by ryanvm@gmail.com at Sun, 14 Oct 2012 20:39:34 +0000
Bug 626245. Part 3: Make aAncestor const in TransformFrameRectToAncestor. r=matspal
c4da85db655437a37af15da1d5fa498c40efc27b: Bug 626245. Part 2: Fix test_convertpoint to poll until the plugin is in the right place. r=matspal
Robert O'Callahan <robert@ocallahan.org> - Mon, 08 Oct 2012 17:46:52 +1300 - rev 110187
Push 23677 by ryanvm@gmail.com at Sun, 14 Oct 2012 20:39:34 +0000
Bug 626245. Part 2: Fix test_convertpoint to poll until the plugin is in the right place. r=matspal
70c50c331112e6b2f9d03cb933d4d6c2c15b2545: Bug 626245. Part 1: Refactor test_plugin_scroll_invalidation to definitely wait for a paint between each call to waitForPaint. r=matspal
Robert O'Callahan <robert@ocallahan.org> - Mon, 08 Oct 2012 17:45:16 +1300 - rev 110186
Push 23677 by ryanvm@gmail.com at Sun, 14 Oct 2012 20:39:34 +0000
Bug 626245. Part 1: Refactor test_plugin_scroll_invalidation to definitely wait for a paint between each call to waitForPaint. r=matspal The current code sets the plugin size to 0,0 and then does some stuff after a setTimeout(0), then sets the size back to 1,1 and expects to be sure to get a new paint event. This is wrong since the timing of plugin geometry changes isn't guaranteed, so we might simply not resize the plugin between setting the size to 0,0 and 1,1.
df1a1a1cfaf49840942f554500873c4befe6d071: Bug 787089. Rip out mUpdatePluginGeometryForFrame optimization. It's not known to be a significant win, and creates problems when prescontext ancestor chains are broken; the complexity probably isn't worthwhile. Also, soon (or now) we can rework plugin bounds computation to be use the display list built for rendering. r=matspal
Robert O'Callahan <robert@ocallahan.org> - Wed, 05 Sep 2012 22:30:10 +1200 - rev 104284
Push 23417 by ryanvm@gmail.com at Thu, 06 Sep 2012 02:27:31 +0000
Bug 787089. Rip out mUpdatePluginGeometryForFrame optimization. It's not known to be a significant win, and creates problems when prescontext ancestor chains are broken; the complexity probably isn't worthwhile. Also, soon (or now) we can rework plugin bounds computation to be use the display list built for rendering. r=matspal
243d341c4edd1b873009348ccfab99dbf3a7b55b: Bug 703241. PresShell::DoScrollContentIntoView and PresShell::ScrollFrameRectIntoView should take transforms into account. r=matspal
Robert O'Callahan <robert@ocallahan.org> - Fri, 22 Jun 2012 17:06:49 +1200 - rev 97366
Push 22970 by emorley@mozilla.com at Fri, 22 Jun 2012 16:12:29 +0000
Bug 703241. PresShell::DoScrollContentIntoView and PresShell::ScrollFrameRectIntoView should take transforms into account. r=matspal
80fa9af29d8ae67620e8bfd3306c8192c0e2d5b0: Bug 754196. If ScrollToShowRect has no need to scroll, don't call ScrollTo, so pending smooth scrolls won't be interrupted. r=matspal
Robert O'Callahan <robert@ocallahan.org> - Wed, 16 May 2012 17:40:57 +1200 - rev 94109
Push 22700 by ryanvm@gmail.com at Thu, 17 May 2012 00:48:56 +0000
Bug 754196. If ScrollToShowRect has no need to scroll, don't call ScrollTo, so pending smooth scrolls won't be interrupted. r=matspal
75840c5beb37e873322a0b564c6f4c72cdce85dc: Bug 754556. Use ScrollToCSSPixels in nsScrollBoxObject::ScrollTo/ScrollBy. r=matspal
Robert O'Callahan <robert@ocallahan.org> - Wed, 16 May 2012 17:40:39 +1200 - rev 94108
Push 22700 by ryanvm@gmail.com at Thu, 17 May 2012 00:48:56 +0000
Bug 754556. Use ScrollToCSSPixels in nsScrollBoxObject::ScrollTo/ScrollBy. r=matspal
3bd87d01b52eede7e2b364fa0904f58cdaabe25e: Bug 754556. Ensure that setting scroll positions in CSS pixels doesn't unexpectedly move the scroll position, especially not in the wrong direction. r=matspal
Robert O'Callahan <robert@ocallahan.org> - Tue, 15 May 2012 17:58:09 +1200 - rev 93967
Push 22688 by emorley@mozilla.com at Tue, 15 May 2012 11:33:31 +0000
Bug 754556. Ensure that setting scroll positions in CSS pixels doesn't unexpectedly move the scroll position, especially not in the wrong direction. r=matspal If the current scroll position is fractional, e.g. y=N.4 CSS pixels, and something tries to read the position in CSS pixels and scroll to that position (e.g. calling window.scrollTo(0, rootElem.scrollTop), or equivalently window.scrollBy(0, 0)), it can actually end up scrolling backwards. So create a new method nsIScrollableFrame::ScrollToCSSPixels which ensures that scrolling to a CSS pixel offset tries to preserve the current fractional scroll position if that's possible, and if that's not possible at least does not allow the scroll position to move in the wrong direction.
67091352b7d2fd86a872d94342a5d70a00bc017d: Bug 681192. Part 17: Disable test_transformed_scrolling_repaints_2.html on Mac. r=matspal
Robert O'Callahan <robert@ocallahan.org> - Thu, 10 May 2012 17:24:21 +1200 - rev 93701
Push 22658 by emorley@mozilla.com at Thu, 10 May 2012 12:06:32 +0000
Bug 681192. Part 17: Disable test_transformed_scrolling_repaints_2.html on Mac. r=matspal
ebb8c64d97fe902a1e4becbce2ceae2804134648: Bug 681192. Part 16: Reduce assertion counts some more. r=matspal
Robert O'Callahan <robert@ocallahan.org> - Thu, 10 May 2012 17:24:21 +1200 - rev 93700
Push 22658 by emorley@mozilla.com at Thu, 10 May 2012 12:06:32 +0000
Bug 681192. Part 16: Reduce assertion counts some more. r=matspal
63898c3a1984f111a8d7a33e4bbdb78bc74dabeb: Bug 681192. Part 15: Fix test now that scrollbox scrollWidth/Heights isn't special. r=matspal
Robert O'Callahan <robert@ocallahan.org> - Thu, 10 May 2012 17:24:20 +1200 - rev 93699
Push 22658 by emorley@mozilla.com at Thu, 10 May 2012 12:06:32 +0000
Bug 681192. Part 15: Fix test now that scrollbox scrollWidth/Heights isn't special. r=matspal
3db8899c075cffd6a805bd7e42c9e0e7bc2a51fe: Bug 681192. Part 14: Fix test to avoid failure due to event coordinate rounding. r=matspal
Robert O'Callahan <robert@ocallahan.org> - Thu, 10 May 2012 17:24:20 +1200 - rev 93698
Push 22658 by emorley@mozilla.com at Thu, 10 May 2012 12:06:32 +0000
Bug 681192. Part 14: Fix test to avoid failure due to event coordinate rounding. r=matspal
583790efc2f536fb3ff5b26298a83919c7804877: Bug 681192. Part 13: Make nsDOMWindowUtils event coordinate calculations more accurate. r=matspal
Robert O'Callahan <robert@ocallahan.org> - Thu, 10 May 2012 17:24:20 +1200 - rev 93697
Push 22658 by emorley@mozilla.com at Thu, 10 May 2012 12:06:32 +0000
Bug 681192. Part 13: Make nsDOMWindowUtils event coordinate calculations more accurate. r=matspal
23b728dcf8b9c977979b5c59f6f1e3700a1db743: Bug 681192. Part 12: more fixes to remove unnecessary rounding to pixels. r=matspal
Robert O'Callahan <robert@ocallahan.org> - Thu, 10 May 2012 17:24:20 +1200 - rev 93696
Push 22658 by emorley@mozilla.com at Thu, 10 May 2012 12:06:32 +0000
Bug 681192. Part 12: more fixes to remove unnecessary rounding to pixels. r=matspal
9d9a3edaa0b985acaafe7894bfe2bafa2efe62f4: Bug 681192. Part 11: Don't snap scrollrange endpoints to device pixels anymore. r=matspal
Robert O'Callahan <robert@ocallahan.org> - Thu, 10 May 2012 17:24:20 +1200 - rev 93695
Push 22658 by emorley@mozilla.com at Thu, 10 May 2012 12:06:32 +0000
Bug 681192. Part 11: Don't snap scrollrange endpoints to device pixels anymore. r=matspal
9343f212a3e38dc8be63d5f36a7d755215842329: Bug 681192. Part 8: Use a generous allowed range when scrolling in CurPosAttributeChanged. r=matspal
Robert O'Callahan <robert@ocallahan.org> - Thu, 10 May 2012 17:24:19 +1200 - rev 93692
Push 22658 by emorley@mozilla.com at Thu, 10 May 2012 12:06:32 +0000
Bug 681192. Part 8: Use a generous allowed range when scrolling in CurPosAttributeChanged. r=matspal There was a subtle reftest failure on Mac in RTL text-overflow tests. I tracked it down to RTL overflow:auto areas being automatically scrolled by a subpixel amount. This was because we try to set scrollbar "curpos" attributes to an integer number of CSS pixels representing the scroll amount, relative to the top-left of GetScrolledRect, which is a noninteger number of CSS pixels in this case. Then in ReflowFinished() we scroll to the saved curpos values plus the top-left of GetScrolledRect. This patch fixes the problem by making CurPosAttributeChanged pass an allowed scroll range to ScrollToWithOrigin. We allow any scroll destination that, when rounded to CSS pixels, would give the same value as "curpos". This fixes the bug, ensuring that ReflowFinished's call to CurPosAttributeChanged will not normally need to scroll because the current position will be in the acceptable range. Also, it means that code that scrolls by setting the "curpos" attribute will be optimized to try to hit a layer pixel boundary.
e54039233ebe094c4d037b771399fef22b92a461: Bug 681192. Part 7: nsTypedSelection should be scrolling a 0,0 size into view to get a particular coordinate into view, not 1,1. r=matspal
Robert O'Callahan <robert@ocallahan.org> - Thu, 10 May 2012 17:24:19 +1200 - rev 93691
Push 22658 by emorley@mozilla.com at Thu, 10 May 2012 12:06:32 +0000
Bug 681192. Part 7: nsTypedSelection should be scrolling a 0,0 size into view to get a particular coordinate into view, not 1,1. r=matspal When asked to scroll a 1,1 appunit rectangle into view, ScrollFrameRectIntoView will now actually succeed! For example if the window is 6000 appunits high and we ask to scroll a 1x1 rect at 0,6000 into view, before bug 681192 was fixed we'd actually fail to do so. We'd compute a desired scroll destination of 0,1, and ScrollTo would round that to 0,0 and we wouldn't scroll at all. Now, we compute a desired scroll destination of 0,1 but also an allowed scroll range for y of "1 to <someting large>", so ScrollFrameRectIntoView will scroll down by a full pixel to 0,60. This is correct --- it gets the subpixel area into view, which the previous code didn't --- but it's not really what DoAutoScroll wants, at least the way test_mousecapture.xul is written. test_mousecapture.xul expects DoAutoScroll to scroll windowheight+N into view by scrolling down by exactly N pixels, so the desired point is exactly at the bottom edge of the window rect. Using a zero-sized rect achieves this.
2453c819127bae3d7553494f8d30fe66fc82fcb9: Bug 681192. Part 6: Test that various scrolling operations inside scaled transforms don't trigger unnecessary repainting. r=matspal
Robert O'Callahan <robert@ocallahan.org> - Thu, 10 May 2012 17:24:19 +1200 - rev 93690
Push 22658 by emorley@mozilla.com at Thu, 10 May 2012 12:06:32 +0000
Bug 681192. Part 6: Test that various scrolling operations inside scaled transforms don't trigger unnecessary repainting. r=matspal