wine/gecko
0
0
Fork 0
mirror of https://gitlab.winehq.org/wine/wine-gecko.git synced 2024-09-13 09:24:08 -07:00
Code Issues Packages Projects Releases Wiki Activity
73,884 Commits 1 Branch 166 Tags 1.2 GiB
f3d98f2440
Commit Graph

795 Commits

Author SHA1 Message Date
Jeff Muizelaar
1901dd355e Bug 673417. imglib: remove mLoadTime. r=joe 
mLoadTime was added by stuart back in 2001. It hasn't as far as I can tell ever been used.

--HG--
extra : rebase_source : fca0f2fcffcae14044da00c5366c4d08edcf3d92
 2011-08-05 17:04:36 -04:00
Jeff Muizelaar
3d23c055b4 Bug 673417. imglib: remove more PRBools. r=joe 
--HG--
extra : rebase_source : cb8f4dfc1048b5e5af72d5544294a70624cdf35e
 2011-08-05 09:57:16 -04:00
Jeff Muizelaar
fefd80c54c Bug 672207. imglib: remove unnecessary scoping. r=joe 
--HG--
extra : rebase_source : 2bcd630768ed6f1782bd46f5c4956e9b326e3deb
 2011-08-04 17:34:19 -04:00
Justin Lebar
51d9635f4a Bug 664659 - Fix about:memory image reporters. r=joe 
--HG--
extra : rebase_source : 489def10afa7f9a33e988d9fc78acdb50e450817
 2011-07-18 09:20:27 -04:00
Jeff Muizelaar
0f62a1219f Bug 672207. imglib-telemetry: record the full time it takes to decode images. r=jlebar,joe 
Do this by accumulating decode latency in mDecodeTime which is added to imgDecodeWorker.
 2011-08-04 16:25:12 -04:00
Joe Drew
3c75d08723 Bug 627015 - Hold on to a reference to the timer, so it doesn't get collected from under us. over-the-shoulder-r=jrmuizel,jdm 2011-08-03 16:39:53 -04:00
Joe Drew
38c27dbd95 Bug 627015 - Add a crashtest for an image that takes a very long time to download the next frame, giving our animation timer a chance to fire. r=jrmuizel 2011-08-03 15:22:47 -04:00
Joe Drew
0474675146 Bug 627015 - Sometimes, images pause mid-frame when downloading. We handle this just fine, but unfortunately we assert that it's not the case. Since this assertion is just bogus, delete it. r=jrmuizel 2011-08-03 15:19:44 -04:00
Joe Drew
99b136a24e Bug 609499 - Crashtest for GIFs that abruptly terminate their LZW data. r=jrmuizel 2011-08-03 15:19:33 -04:00
Joe Drew
96ce34cefc Bug 609499 - Some invalid GIF files don't terminate their LZW data with a 0-length block. Handle them better by handling them less specially. r=jrmuizel 2011-08-03 15:19:19 -04:00
Scott Johnson
c92ae5ac1a Bug 673535 - RasterImage::Anim::ensureAnimExists() should be cleaned up. 2011-07-23 14:49:00 -04:00
Marco Bonardo
99ffcc6359 Backout changeset 984e656becec (Bug 664659) due to regressions in Tp5 XRES and RSS. 2011-08-01 10:20:18 +02:00
Justin Lebar
816e51259b Bug 664659 - Fix about:memory image reporters. r=joe 2011-07-18 09:20:27 -04:00
Daniel Holbert
5798400e19 Bug 670542 patch 2: Rev iids of nsIURI & descendants. r=jst 2011-07-26 22:11:40 -07:00
Michael Wu
706be3a5e8 Bug 671185 - Incorrect return of NS_ERROR_* codes in functions returning PRBool, r=mak,ehsan,taras,biesi,pike,khuey,dholbert,josh,bjacob,bsmith 2011-07-25 21:57:58 -07:00
Justin Lebar
59d336e3b6 Bug 672578 - Part 2. Stop decoding previously-discarded images when the containing tab loses focus. r=joe 2011-07-21 14:15:17 -04:00
Justin Lebar
ea250e229a Bug 672578 - Part 1. Don't trigger a decode when FrameUpdated is called. r=joe 2011-07-21 14:14:41 -04:00
Jeff Muizelaar
05dca460b0 Bug 672225. imglib: remove some PRBools that we're hurting my eyes. r=joe 
--HG--
extra : rebase_source : b06e2b61b1104234481bf30a71cd3742a8c9a54f
 2011-07-20 17:52:30 -04:00
Jeff Muizelaar
0455058ca3 Bug 672207. imglib: Time how long we spend in the decode loop. r=joe 
For now this is a global counter, the eventual plan is to make this per image
format. This will be our proxy for how long we block the event loop.

--HG--
extra : rebase_source : b283fff38585825f5f46884e12508b6323e178b3
 2011-07-20 17:51:54 -04:00
Justin Lebar
6132ee5300 Backing out bug 664659 (rev 7b49a2857e18) due to Mac memory regressions. 2011-07-19 19:25:19 -04:00
Jeff Muizelaar
f7a22f3c27 Bug 672207. imglib: Switch away from nsTime in favour of TimeStamp/TimeDuration. r=joe 2011-07-19 11:18:43 -04:00
Justin Lebar
ecdd58c973 Bug 664659 - Fix about:memory image reporters. r=joe,njn 2011-07-18 09:20:27 -04:00
Joe Drew
2ee1d16443 Bug 671906 - Test to make sure we always reload images validated using CORS if their principal has changed. r=bz 2011-07-17 13:40:26 -04:00
Joe Drew
f2e56e6dc3 Bug 671906 - Fix the logic for matching principals in imagelib CORS code. r=bz 2011-07-17 13:40:24 -04:00
Joe Drew
171af95914 Bug 641748 - Create a simulated GIF webcam, and test to make sure we actually get the second frame. r=jrmuizel 
This also implicitly tests that we don't hit this bug, because it was an
NS_ABORT_IF_FALSE, and that will be hit on debug reftest runs.
 2011-07-14 14:47:45 -04:00
Joe Drew
bbd5b84790 Bug 641748 - Remove RasterImage::AppendFrame in favour of RasterImage::EnsureFrame, because decoders get re-initialized when using multipart/x-mixed-replace, and this can lead to the frame count for RasterImages being different from the frame count for decoders. r=jrmuizel 2011-07-14 14:47:43 -04:00
Joe Drew
50cdf9643d Bug 641748 - Change imgFrame::mPaletteDepth from signed to unsigned. r=jrmuizel 2011-07-14 14:47:41 -04:00
Joe Drew
73e2449ba9 Bug 664299 - Support CORS in Imagelib by adding load flags for the various CORS modes. r=jrmuizel,bz 2011-07-14 14:47:34 -04:00
Joe Drew
952b8c93a0 Bug 664299 - Pass the loading principal to imgILoader::loadImage so we can operate on it later. r=bz 2011-07-14 14:47:32 -04:00
Justin Lebar
60f61836d3 Bug 481962 - Force logging on in imagelib. r=joe 
--HG--
extra : rebase_source : 110c088d4221d56aede72808e05c202a7e2a3f91
 2011-07-08 13:55:03 -04:00
Nicholas Nethercote
7b61ea2b0e Bug 653627 - Don't use char** in nsIMemoryReporter. r=khuey, sr=bz. 2011-07-08 12:45:16 +10:00
Joe Drew
db615ad885 Bug 619048 - Some "zero" (really max) height and width .ico files as crashtests. r=jrmuizel 
The fact that these are interpreted as zero-size is bug 668068.
 2011-07-01 13:08:19 -04:00
Joe Drew
088e5533a2 Bug 619048 - Correctly reject zero-height and zero-width images. r=jrmuizel 2011-07-01 13:03:47 -04:00
Joe Drew
2490cf4cd9 Bug 552605 - Tests to make sure a URI always maps to the same image in the same document, even when loaded with cache-control: no-cache. r=bz 2011-07-01 13:03:45 -04:00
Joe Drew
bcb9088b38 Bug 552605 - Write tests to make sure we reload images that are redirected with a Cache-Control: no-cache header. r=bz 2011-07-01 13:03:43 -04:00
Joe Drew
ebe4d8577b Bug 552605 - Rely on the necko cache to validate all cache-control headers rather than asking for it to load only if modified. This makes it so we don't have to implement HTTP cache-control correctness on redirects ourselves. r=jrmuizel 2011-07-01 13:03:42 -04:00
Joe Drew
28b8e17396 Bug 552605 - imgRequest::mKeyURI and mURI are now the same because we don't change the cache on redirect. Let's keep track of the current URI instead. r=jrmuizel 2011-07-01 13:03:40 -04:00
Joe Drew
d31fbb1416 Bug 552605 - Keep track of all redirects when validating cache entries, so we make sure cache headers are always kept track of. r=jrmuizel 2011-07-01 13:03:38 -04:00
Joe Drew
8e1674f49a Bug 552605 - check for file:// scheme in SetCacheValidation rather than on cache entry construction, so we don't have to have the URI when constructing cache entries. r=jrmuizel 2011-07-01 13:03:37 -04:00
Joe Drew
2fd2525747 Bug 552605 - Change "must validate if expired" to "must always validate" in the image cache, to fit with the spirit of the HTTP spec. r=jrmuizel f=bz 2011-07-01 13:03:35 -04:00
Joe Drew
fadc80934d Bug 552605 - Set cache expiry and validation on both OnStartRequest and OnRedirectVerifyCallback so that our image cache entries have the union of all cache headers. r=jrmuizel 2011-07-01 13:03:33 -04:00
Joe Drew
7c5a983ab5 Bug 552605 - Keep images in the image cache based on their original URI, not the URI they're redirected to. r=jrmuizel 2011-07-01 13:03:32 -04:00
Joe Drew
183549614c Bug 552605 - Fix some whitespace issues that crept in. r=jrmuizel 2011-07-01 13:03:30 -04:00
Marco Bonardo
6013ac2db8 Backout bug 552605 and bug 619048 due to increase in random failures in reftests on a CLOSED TREE 2011-07-01 18:08:24 +02:00
Joe Drew
85e1a7141b Bug 552605 - Some "zero" (really max) height and width .ico files as crashtests. r=jrmuizel 
The fact that these are interpreted as zero-size is bug 668068.
 2011-06-30 21:58:41 -04:00
Joe Drew
367ea5d97d Bug 619048 - Correctly reject zero-height and zero-width images. r=jrmuizel 2011-06-30 21:58:40 -04:00
Joe Drew
6ee9d7ab14 Bug 552605 - Tests to make sure a URI always maps to the same image in the same document, even when loaded with cache-control: no-cache. r=bz 2011-06-30 21:58:38 -04:00
Joe Drew
301a37eb71 Bug 552605 - Write tests to make sure we reload images that are redirected with a Cache-Control: no-cache header. r=bz 2011-06-30 21:58:36 -04:00
Joe Drew
f9164fc9c5 Bug 552605 - Rely on the necko cache to validate all cache-control headers rather than asking for it to load only if modified. This makes it so we don't have to implement HTTP cache-control correctness on redirects ourselves. r=jrmuizel 2011-06-30 21:58:34 -04:00
Joe Drew
fe121634b0 Bug 552605 - imgRequest::mKeyURI and mURI are now the same because we don't change the cache on redirect. Let's keep track of the current URI instead. r=jrmuizel 2011-06-30 21:58:33 -04:00