67e02399d88a85c7ef6932eea25ba5a3f32754d1: Bug 508908. Disable windowed plugins in popup windows, since they don't work. r=dbaron
Robert O'Callahan <robert@ocallahan.org> - Wed, 30 Sep 2009 08:18:20 +1000 - rev 33373
Bug 508908. Disable windowed plugins in popup windows, since they don't work. r=dbaron
8cccae76de91dcda353e05308b34bc07e13d10c5: Bug 519136. Clean up media-related logging. r=doublec
Robert O'Callahan <robert@ocallahan.org> - Wed, 30 Sep 2009 07:32:44 +1000 - rev 33372
Bug 519136. Clean up media-related logging. r=doublec
ac37ce54ddb0c3ec11d1510351a0e36e19910c54: Bug 508945. Additional fixes for plugin printing when the plugin has borders or padding. r=dbaron
Robert O'Callahan <robert@ocallahan.org> - Fri, 11 Sep 2009 13:44:20 +1200 - rev 33371
Bug 508945. Additional fixes for plugin printing when the plugin has borders or padding. r=dbaron
5cc6bebd0dd90096327aef75a018f1d3fdf078e8: Bug 508495. Remove CreateWidgetForView from nsIFrame and nsObjectFrame since it's no longer used there. r=dbaron
Robert O'Callahan <robert@ocallahan.org> - Wed, 30 Sep 2009 07:32:43 +1000 - rev 33370
Bug 508495. Remove CreateWidgetForView from nsIFrame and nsObjectFrame since it's no longer used there. r=dbaron
29e974a886024b696404a7a213e958ed70871731: Bug 508495. Let CSS borders and padding apply to plugin elements, and fix layout, painting and event handling to work with them. r=dbaron
Robert O'Callahan <robert@ocallahan.org> - Fri, 11 Sep 2009 13:44:20 +1200 - rev 33369
Bug 508495. Let CSS borders and padding apply to plugin elements, and fix layout, painting and event handling to work with them. r=dbaron
b540248e3163e1a73d31e3772dbfead13e3ee937: Bug 513082. Instead of passing aDirtyRect to nsDisplayItem::Paint, have nsDisplayList::OptimizeVisibility (renamed to ComputeVisibility) compute the intersection of the visible region with the bounds of each display item and store the result in the item's mVisibleRect. This lets us do less work to paint display items which have a small intersection with the visible region but a large intersection with the bounding-box of the visible region. r=dbaron
Robert O'Callahan <robert@ocallahan.org> - Mon, 07 Sep 2009 12:35:14 +1200 - rev 33368
Bug 513082. Instead of passing aDirtyRect to nsDisplayItem::Paint, have nsDisplayList::OptimizeVisibility (renamed to ComputeVisibility) compute the intersection of the visible region with the bounds of each display item and store the result in the item's mVisibleRect. This lets us do less work to paint display items which have a small intersection with the visible region but a large intersection with the bounding-box of the visible region. r=dbaron
6bfb2b05290f5d5b7aa9d2c4d53da75cb1a13e70: Bug 513082. Round out 'region of moving visible content' to device pixels to reduce incidence of stray subpixel rectangles causing unnecessary painting. r=dbaron
Robert O'Callahan <robert@ocallahan.org> - Mon, 07 Sep 2009 12:35:14 +1200 - rev 33367
Bug 513082. Round out 'region of moving visible content' to device pixels to reduce incidence of stray subpixel rectangles causing unnecessary painting. r=dbaron
5d82c6bfa38a6d147b0f0a81c0406da37c001f7b: Bug 510856. Additional scrolling tests.
Robert O'Callahan <robert@ocallahan.org> - Mon, 07 Sep 2009 12:35:13 +1200 - rev 33366
Bug 510856. Additional scrolling tests.
7ac2d53330dbe11759ff2465e5e4eb270aa47b9f: Bug 517818. Wave decoder should not look beyond the Wave data area to see if we have a 'next frame' to play. r=kinetik
Robert O'Callahan <robert@ocallahan.org> - Wed, 30 Sep 2009 07:32:41 +1000 - rev 33365
Bug 517818. Wave decoder should not look beyond the Wave data area to see if we have a 'next frame' to play. r=kinetik
a1e033c280f00fdd36348da68e59b318cfa02104: Bug 517818. Treat HTTP errors (other than 416 'Requested Range Not Available') as network errors. And do not allow an error page to be treated as media data. r=kinetik
Robert O'Callahan <robert@ocallahan.org> - Wed, 30 Sep 2009 07:32:30 +1000 - rev 33364
Bug 517818. Treat HTTP errors (other than 416 'Requested Range Not Available') as network errors. And do not allow an error page to be treated as media data. r=kinetik
6e4ad2b2048214a3d8cfe22e0d546005ad1eabdd: Bug 517818. When nsMediaCache::Update wants to read from the end of the stream, don't resume if we're already suspended. And if the channel is not already positioned at the end, suspend instead of seeking to the end, since there is no point in seeking to the end and it may cause harmful side effects. r=doublec
Robert O'Callahan <robert@ocallahan.org> - Wed, 30 Sep 2009 06:48:05 +1000 - rev 33363
Bug 517818. When nsMediaCache::Update wants to read from the end of the stream, don't resume if we're already suspended. And if the channel is not already positioned at the end, suspend instead of seeking to the end, since there is no point in seeking to the end and it may cause harmful side effects. r=doublec
62b992597da67dfaba6a46faf74cdb7f7a214a25: Bug 510856. ComputeRepaintRegionForCopy should not look outside the update rect for moving and non-moving visible content. r=dbaron
Robert O'Callahan <robert@ocallahan.org> - Mon, 07 Sep 2009 12:35:13 +1200 - rev 33362
Bug 510856. ComputeRepaintRegionForCopy should not look outside the update rect for moving and non-moving visible content. r=dbaron
bce848ea984d4af25cd051861bf1813d878697e2: Bug 396392. Fix test to not assume that the height of a block matches the height of the text inside it.
Robert O'Callahan <robert@ocallahan.org> - Fri, 25 Sep 2009 17:18:21 +1200 - rev 33361
Bug 396392. Fix test to not assume that the height of a block matches the height of the text inside it.
8d1d0e97877603fd1ec497d00f16fb968cf9e90f: Bug 442715. Invalidate the whole border-box when an element's size changes and its background is being clipped by a border-radius. r=dbaron
Robert O'Callahan <robert@ocallahan.org> - Fri, 25 Sep 2009 11:52:13 +1200 - rev 33360
Bug 442715. Invalidate the whole border-box when an element's size changes and its background is being clipped by a border-radius. r=dbaron
298772a42ecc8e3ab88ff9034430ef650ac2e188: Bug 509693. Don't use SW_SCROLLCHILDREN if some child has a descendant window that's outside the scrolled rectangle. r=jmathies
Robert O'Callahan <robert@ocallahan.org> - Thu, 24 Sep 2009 21:37:01 +1200 - rev 33359
Bug 509693. Don't use SW_SCROLLCHILDREN if some child has a descendant window that's outside the scrolled rectangle. r=jmathies
bfbd630be11a85141bc887af3326b4710ab9a2ee: Bug 509693. Make Windows test plugin in windowed mode have a child widget, and add API so we can check that that child widget is in the right place. r=josh
Robert O'Callahan <robert@ocallahan.org> - Thu, 24 Sep 2009 21:35:17 +1200 - rev 33358
Bug 509693. Make Windows test plugin in windowed mode have a child widget, and add API so we can check that that child widget is in the right place. r=josh
f3c07913015f8a79464adb52186e4188374b720a: Bug 509693. Remove eWindowType_java (unused) and actually use eWindowType_plugin. r=josh
Robert O'Callahan <robert@ocallahan.org> - Thu, 24 Sep 2009 21:32:20 +1200 - rev 33357
Bug 509693. Remove eWindowType_java (unused) and actually use eWindowType_plugin. r=josh
e5ff8bfef7847cb22b8077a98a1b1f904d8c0c2a: Bug 514660. Use classes instead of an explicit 'style' attribute to style the anonymous DIV of a text control, so that the styles are treated as user-agent styles and not removed when author styling is disabled. r=bzbarsky
Robert O'Callahan <robert@ocallahan.org> - Wed, 23 Sep 2009 23:02:27 +1200 - rev 33356
Bug 514660. Use classes instead of an explicit 'style' attribute to style the anonymous DIV of a text control, so that the styles are treated as user-agent styles and not removed when author styling is disabled. r=bzbarsky
8cbc47eee6593699d684026ddb0a670a0154cc81: bug 441324: implement infallible ::operator new(), malloc() and friends. make |new Foo()| infallible, but leave |malloc()| fallible for the time being. r=blassey sr=bsmedberg,vlad
Chris Jones <jones.chris.g@gmail.com> - Thu, 01 Oct 2009 21:50:09 -0500 - rev 33355
bug 441324: implement infallible ::operator new(), malloc() and friends. make |new Foo()| infallible, but leave |malloc()| fallible for the time being. r=blassey sr=bsmedberg,vlad
865f83caace107cb27d2b1ecb6f9fd162b4c89e7: Testing impact on Ts Shutdown
Dão Gottwald <dao@mozilla.com> - Fri, 02 Oct 2009 01:51:49 +0200 - rev 33354
Testing impact on Ts Shutdown
(0) -30000 -10000 -3000 -1000 -300 -100 -50 -20 +20 +50 +100 +300 +1000 +3000 +10000 +30000 tip