c6e8b4e4c98df5b2b2dd5be92e1f2cbb8ae38345: Bug 1435296 Update the CSS Animations tests to handle our new Timer Precision decision draft
Tom Ritter <tom@mozilla.com> - Wed, 07 Feb 2018 20:35:38 -0600 - rev 752370
Push 98243 by bmo:tom@mozilla.com at Thu, 08 Feb 2018 02:38:28 +0000
Bug 1435296 Update the CSS Animations tests to handle our new Timer Precision decision This commit does several subtle things. 1: It changes ok() to opener.ok() ok is not defined, we have to use opener.ok. This was not caught before because this call is used to provide additional debugging information when a test fails. Test didn't fail, didn't hit that line. 2: It disables the call to opener.ok() we just fixed. As the comment there describes, we expect that function to fail, so we don't want to assert(false). 3: It inverts failures to successes if only the reduceTimerPrecision pref is set MozReview-Commit-ID: lpKKhJoDs6
6d13dd7b62c3c55db721b2366ced1e733a1c9d20: Bug 1435296 Address test failures caused by bumping timer precision to 2 ms draft
Tom Ritter <tom@mozilla.com> - Tue, 06 Feb 2018 15:11:56 -0600 - rev 752369
Push 98243 by bmo:tom@mozilla.com at Thu, 08 Feb 2018 02:38:28 +0000
Bug 1435296 Address test failures caused by bumping timer precision to 2 ms This affects several tests, and in all but one case merely disables timer precision reduction on those tests. In the other singular test, it no longer requires the end time of an operation to be strictly greater than the start time, and allows it to be equal to it. MozReview-Commit-ID: J59c7xQtZZJ
a8e1ff20f09d0a719cf2f71cbcb1ab8e9520464a: Bug 1435296 Do not apply timer clamping to CSS animations. draft
Tom Ritter <tom@mozilla.com> - Wed, 07 Feb 2018 16:50:57 -0600 - rev 752368
Push 98243 by bmo:tom@mozilla.com at Thu, 08 Feb 2018 02:38:28 +0000
Bug 1435296 Do not apply timer clamping to CSS animations. This patch creates the capability to have callsites specify if timestamps should be clamped only in Resist Fingerprinting Mode, or in the more expansive Timer PRecision Reduction Mode. Then it changes the CSS Animation callsite to only apply in RFP Mode. This avoids regressing RFP. MozReview-Commit-ID: B1pSri0kRk6
4e1dc5d9dbc1c6a7877fd156d871d2b430815efd: Bug 1435296 Bump the default timer precision resolution to 2ms r?overholt draft
Tom Ritter <tom@mozilla.com> - Fri, 02 Feb 2018 11:05:40 -0600 - rev 752367
Push 98243 by bmo:tom@mozilla.com at Thu, 08 Feb 2018 02:38:28 +0000
Bug 1435296 Bump the default timer precision resolution to 2ms r?overholt MozReview-Commit-ID: G33jNNJVRSU
8adb0a4cccf81a982808ca8881d329881019a8c9: Bug 1435803: Loosen an assertion that may not hold on the old style system. r?dbaron draft
Emilio Cobos Álvarez <emilio@crisal.io> - Thu, 08 Feb 2018 02:30:13 +0100 - rev 752366
Push 98242 by bmo:emilio@crisal.io at Thu, 08 Feb 2018 01:31:41 +0000
Bug 1435803: Loosen an assertion that may not hold on the old style system. r?dbaron MozReview-Commit-ID: HThEZ5k0Nxf
665235f92711f327fcb63604e500bae0b35b7133: Bug 1434802: Tweak specified font family serialization in Gecko so that it is simpler. r?xidorn draft
Emilio Cobos Álvarez <emilio@crisal.io> - Wed, 07 Feb 2018 18:20:23 +0100 - rev 752365
Push 98242 by bmo:emilio@crisal.io at Thu, 08 Feb 2018 01:31:41 +0000
Bug 1434802: Tweak specified font family serialization in Gecko so that it is simpler. r?xidorn In particular, every time that there's at least more than one identifier, switch to quoted family name, since the reconstruction of the serialization will be lossy anyway. This allows us to avoid copies and all that. What Chrome implements doesn't make much sense in the sense that they always serialize: font-family: "foo"; -> font-family: foo; font-family: foo bar; -> font-family: "foo bar"; font-family: foo\ bar; -> font-family: "foo bar"; This patch makes us match on the second case, but not on the rest, because I think Gecko's behavior is preferable in those cases. MozReview-Commit-ID: JwBECA93lfi
abc273052ee737c9d7cc00cc2521fb73295d2e1d: style: Derive ToCss for FontFamilyList. draft
Emilio Cobos Álvarez <emilio@crisal.io> - Wed, 07 Feb 2018 17:30:28 +0100 - rev 752364
Push 98242 by bmo:emilio@crisal.io at Thu, 08 Feb 2018 01:31:41 +0000
style: Derive ToCss for FontFamilyList. The extra reference in to_css is needed because the family list iterator returns by value in Gecko. Signed-off-by: Emilio Cobos Álvarez <emilio@crisal.io>
b33b60f26c69bc47f37ad2cefad34aba62e1f18e: Bug 1436612 - Package tooltool.py in mozharness.zip; r?glandium draft
Gregory Szorc <gps@mozilla.com> - Wed, 07 Feb 2018 16:53:59 -0800 - rev 752363
Push 98241 by bmo:gps@mozilla.com at Thu, 08 Feb 2018 00:54:27 +0000
Bug 1436612 - Package tooltool.py in mozharness.zip; r?glandium d09848a31f54 (bug 1436612) added a copy of tooltool.py in the repo. I did this because I thought the alternative and proper way of doing it would be hard. I have since realized that we produce mozharness.zip with test_archive.py. This means we can pull in files from anywhere in the build into the archive. So we pull in tooltool.py from its original location and delete the copy that was just added to testing/mozharness. MozReview-Commit-ID: JOnBnIothRW
e948a12825a62f40a4138fadee7573e687fc4252: Bug 1430952 - Add 'tl', 'km' and 'meh' to Fennec all-locales for single-locale builds, r?flod draft
Delphine Lebédel <dlebedel@mozilla.com> - Wed, 07 Feb 2018 16:21:37 -0800 - rev 752362
Push 98240 by bmo:lebedel.delphine@gmail.com at Thu, 08 Feb 2018 00:22:12 +0000
Bug 1430952 - Add 'tl', 'km' and 'meh' to Fennec all-locales for single-locale builds, r?flod MozReview-Commit-ID: FApINlDzv3d
e0d4f273e33e3a77558d954efe514b35399b099e: Bug 1436517 - Limit access to the server-timing header to HTTPS contexts. r?mcmanus draft
Nicholas Hurley <hurley@mozilla.com> - Wed, 07 Feb 2018 16:05:38 -0800 - rev 752361
Push 98239 by bmo:hurley@mozilla.com at Thu, 08 Feb 2018 00:13:19 +0000
Bug 1436517 - Limit access to the server-timing header to HTTPS contexts. r?mcmanus This also introduces a hidden pref to allow server-timing access from HTTP contexts for the purposes of our xpcshell tests. We'll remove that once we get h2 (and therefore tls test) support for server-timing trailers (https://bugzilla.mozilla.org/show_bug.cgi?id=1436601). This does not reject or otherwise error when receiving server-timing headers or trailers on non-HTTPS contexts, it just makes it unavailable outside the channel. MozReview-Commit-ID: qi4h0VQknE
7eddfd228c30746bf15b13b7140d5b0472ae5e48: Bug 1346983 - Add a new nsColumnSetWrapperFrame class to the build to be used as a wrapper for columnset and column span frames. r?dbaron draft
Neerja Pancholi <npancholi@mozilla.com> - Wed, 07 Feb 2018 15:56:00 -0800 - rev 752360
Push 98238 by bmo:npancholi@mozilla.com at Thu, 08 Feb 2018 00:04:44 +0000
Bug 1346983 - Add a new nsColumnSetWrapperFrame class to the build to be used as a wrapper for columnset and column span frames. r?dbaron MozReview-Commit-ID: J8cpsBKjWFX
82da371289b4ca98884a32f40289f5ebfd80fd15: Bug 1434741 - Only check final download URL against the application reputation whitelist. r?gcp draft
Francois Marier <francois@mozilla.com> - Mon, 05 Feb 2018 18:11:56 -0800 - rev 752359
Push 98237 by fmarier@mozilla.com at Wed, 07 Feb 2018 23:56:12 +0000
Bug 1434741 - Only check final download URL against the application reputation whitelist. r?gcp MozReview-Commit-ID: QCaStgteko
f5066287278bacedc97d0d07f4eda80ed2ffdcec: bug 1432229, land compare-locales 2.7 and python-fluent 0.6.1, r?gandalf draft
Axel Hecht <axel@pike.org> - Wed, 07 Feb 2018 20:22:37 +0100 - rev 752358
Push 98236 by axel@mozilla.com at Wed, 07 Feb 2018 23:38:02 +0000
bug 1432229, land compare-locales 2.7 and python-fluent 0.6.1, r?gandalf MozReview-Commit-ID: 2Oep88PhVXF
4221e203fb369d8cf8eebbe90b4cba40a38c52d0: Bug 1346983 - Add a new nsColumnSetWrapperFrame class to the build to be used as a wrapper for columnset and column span frames. r?dbaron draft
Neerja Pancholi <npancholi@mozilla.com> - Wed, 07 Feb 2018 15:32:48 -0800 - rev 752357
Push 98235 by bmo:npancholi@mozilla.com at Wed, 07 Feb 2018 23:37:26 +0000
Bug 1346983 - Add a new nsColumnSetWrapperFrame class to the build to be used as a wrapper for columnset and column span frames. r?dbaron MozReview-Commit-ID: J8cpsBKjWFX
076348220019fa145d5eeb8ed845b1b2f5bdc331: Bug 1404222 Part 6 - Turn off a 'todo' in a mochitest. draft
Brad Werth <bwerth@mozilla.com> - Wed, 07 Feb 2018 14:59:43 -0800 - rev 752356
Push 98234 by bwerth@mozilla.com at Wed, 07 Feb 2018 23:34:44 +0000
Bug 1404222 Part 6 - Turn off a 'todo' in a mochitest. MozReview-Commit-ID: 18bkqA6rxoZ
3408e7ce15b79eaa3413a41e56499ff6a5725934: Bug 1404222 Part 5 - Add a crashtest. draft
Ting-Yu Lin <aethanyc@gmail.com> - Thu, 25 Jan 2018 15:01:08 +0800 - rev 752355
Push 98234 by bwerth@mozilla.com at Wed, 07 Feb 2018 23:34:44 +0000
Bug 1404222 Part 5 - Add a crashtest. The browser should not crash if there's no width and height set on the div with linear-gradient. MozReview-Commit-ID: 2DbfpV7JT2m
9c4ab525120059b82093823e7f79e4480aed8dea: Bug 1404222 Part 4 - Add web-platform-tests for linear-gradient with writing-modes. draft
Ting-Yu Lin <aethanyc@gmail.com> - Thu, 25 Jan 2018 15:00:10 +0800 - rev 752354
Push 98234 by bwerth@mozilla.com at Wed, 07 Feb 2018 23:34:44 +0000
Bug 1404222 Part 4 - Add web-platform-tests for linear-gradient with writing-modes. There's no shape-outside: <image> test with writing-modes on web-platform-tests, so I added some. MozReview-Commit-ID: FekYjzweKRG
c9867bf48f59104b437730d2fc077f30288a46ee: Bug 1404222 Part 3 - Replace hash '#' with %23 in SVG data URIs, to fix XML parse errors. draft
Ting-Yu Lin <aethanyc@gmail.com> - Thu, 25 Jan 2018 14:58:07 +0800 - rev 752353
Push 98234 by bwerth@mozilla.com at Wed, 07 Feb 2018 23:34:44 +0000
Bug 1404222 Part 3 - Replace hash '#' with %23 in SVG data URIs, to fix XML parse errors. The hash symbol '#' is reserved as fragment identifier per bug 1430526 comment 1, so fix the test cases. MozReview-Commit-ID: 5o0qLg81iVi
5ca5b8e008e5bc1048cd782889178542954f3c2d: Bug 1404222 Part 2 - Trigger reflow after "shape-outside" image has finished decoding a frame. draft
Ting-Yu Lin <aethanyc@gmail.com> - Thu, 25 Jan 2018 14:56:43 +0800 - rev 752352
Push 98234 by bwerth@mozilla.com at Wed, 07 Feb 2018 23:34:44 +0000
Bug 1404222 Part 2 - Trigger reflow after "shape-outside" image has finished decoding a frame. When we finish decoding an image frame, we need to trigger reflow for the frame containing a float with shape-outside: <image>. MozReview-Commit-ID: 2eNsXsw4kxM
25d7e626b61214821156bd0c0580ace478aba160: Bug 1404222 Part 1 - Implement shape-outside: <image>. draft
Ting-Yu Lin <aethanyc@gmail.com> - Thu, 25 Jan 2018 14:55:18 +0800 - rev 752351
Push 98234 by bwerth@mozilla.com at Wed, 07 Feb 2018 23:34:44 +0000
Bug 1404222 Part 1 - Implement shape-outside: <image>. When creating ImageShapeInfo, it's likely that the image is still decoding. Part 2 will add mechanism to trigger reflow after the image is ready. MozReview-Commit-ID: FpWE7g7NK70
(0) -300000 -100000 -30000 -10000 -3000 -1000 -300 -100 -50 -20 +20 +50 +100 +300 +1000 +3000 +10000 +30000 tip