Changes:
As documented in bug 1621483, `marionette` experiences a non-trivial amount of issues when run on ubuntu1804 docker image with GTK/GNOME desktop environment enabled.
GTK/GNOME has a higher degree of asynciness when manipulating window size/position and this leads to `marionette` and the derived suite `web-platform-tests-wdspec` reporting intermittent oranges for a number of tests.
While attempts were made to incorporate a fix for the marionette driver itself, the best attempts have only been able to achieve a ~50% reliability in green runs.
This patch reintroduces the use of bare `compiz` window manager exclusively for these two problematic test suites so that at least the tests are running on non-legacy software.
Differential Revision: https://phabricator.services.mozilla.com/D66482
--HG--
extra : moz-landing-system : lando
- Add enum AllowImportRules to CSS parsing.
- `replaceSync()` will skip over @import rules and continue parsing.
- `replace()` will skip over @import rules and continue parsing.
- `insertRule()` will throw a syntax error on @import rules.
- Modify WPT test cases to reflect these changes.
Differential Revision: https://phabricator.services.mozilla.com/D61882
--HG--
extra : moz-landing-system : lando
Tons of new passes, some fuzzy bits, and one failure from a test that makes bad
assumptions about themed and unthemed <input type=color> being equally tall.
Differential Revision: https://phabricator.services.mozilla.com/D65724
--HG--
extra : moz-landing-system : lando
We should throw a DOMException SyntaxError when setting pseudoElement a
syntactically invalid string or an unsupported pseudo.
Differential Revision: https://phabricator.services.mozilla.com/D66321
--HG--
extra : moz-landing-system : lando
Convert the gtest option parser from optparse to argparse. mochitest, reftest,
and other suites use argparse. Using argparse will simplify the integration
of gtest with the custom retrigger action.
Differential Revision: https://phabricator.services.mozilla.com/D66429
--HG--
extra : moz-landing-system : lando
Add additional function signature for mozalloc_abort, as observed in Windows crash reports.
I am a little concerned that there is a more generic issue, but since I cannot find
any other examples, adding this signature seems like the most expedient and least risky
way forward.
Differential Revision: https://phabricator.services.mozilla.com/D66418
--HG--
extra : moz-landing-system : lando
Automatic update from web-platform-tests
Fix WPT test redirect-mode.any.html (#22137)
Fix WPT test redirect-mode.any.html by also specifying the server port, this
is needed for WebKit local test runs.
--
wpt-commits: ba2e4f922451ce8cc99c3bede003b6306ffc9a98
wpt-pr: 22137
Automatic update from web-platform-tests
[TrustedTypes] Update working IDL files for StringContext extended attribute.
Previous CL(https://chromium-review.googlesource.com/c/chromium/src/+/2012284)
replaced an union types with WebIDL annotation and this CL updated IDL
files which use the union types to use WebIDL annotations instead.
Bug: 1043136
Change-Id: Ic17eb8f058cdc6ee047e415abb988bfc7b6b2741
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2013125
Commit-Queue: Yifan Luo <lyf@chromium.org>
Reviewed-by: Mike West <mkwst@chromium.org>
Reviewed-by: Daniel Vogelheim <vogelheim@chromium.org>
Reviewed-by: Yuki Shiino <yukishiino@chromium.org>
Cr-Commit-Position: refs/heads/master@{#748179}
--
wpt-commits: 227d9a1de819a91cbe503a10abddf898dfade828
wpt-pr: 21956
Automatic update from web-platform-tests
Fix bounding box calculation for <tspan> (and other LayoutSVGInlines)
Make the various bounding box calculations for LayoutSVGInline return
values derived from the associated flow boxes rather than delegating
to the ancestor LayoutSVGText. Refactor and reuse the code from the
similar functionality LayoutSVGText.
Since the reference box for these element should still remain the same
as the reference box for the ancestor <text> add a new helper to
compute (delegate) that and use that when resolving clips/filter/et.c.
Bug: 951706, 349835, 267481
Change-Id: Idf4a4c586ef9902cac2e795e9fbda0009b2ea0ad
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2088094
Commit-Queue: Fredrik Söderquist <fs@opera.com>
Reviewed-by: Stephen Chenney <schenney@chromium.org>
Reviewed-by: Philip Rogers <pdr@chromium.org>
Cr-Commit-Position: refs/heads/master@{#748175}
--
wpt-commits: 26f9e1b7c81da4aa1c7249d6ee2268e9dedfa96c
wpt-pr: 22104
Automatic update from web-platform-tests
COEP takes effect on a new service worker except for importScripts
Currently Cross-Origin-Embedder-Policy header is parsed in
ServiceWorkerNewScriptLoader when launching a new service worker.
However, the COEP in the response header for the main script isn't
applied for the network loader created on launching a service worker
because the network loader is created before starting to load the main
script.
This CL is to pause the service worker's script evaluation until the
main script is loaded. Once the main script loaded, new subresource
loaders are created by using the COEP value, and they are sent to the
service worker.
Imported scripts also need to refer to the new value, but the loaders
are in different place (ServiceWorkerScriptLoaderFactory), so it'll be
addressed in a later patch.
Bug: 1039613
Change-Id: Ice82eef86beec2f2f5c5e91f3aa136229f14c3ab
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2029403
Commit-Queue: Makoto Shimazu <shimazu@chromium.org>
Reviewed-by: Kent Tamura <tkent@chromium.org>
Reviewed-by: Kinuko Yasuda <kinuko@chromium.org>
Reviewed-by: Kenichi Ishibashi <bashi@chromium.org>
Auto-Submit: Makoto Shimazu <shimazu@chromium.org>
Cr-Commit-Position: refs/heads/master@{#748112}
--
wpt-commits: d4391fedfec67d69f00396873a11f501a97ee3f1
wpt-pr: 22080
Automatic update from web-platform-tests
Revert "Origin policy: implement window.originPolicyIds"
This reverts commit 10bfb977d51a0e7885bf9f3dfd4270f139fd3eeb.
Reason for revert: This CL seems cause build failure since https://ci.chromium.org/p/chromium/builders/ci/WebKit%20Linux%20MSAN/4651
Original change's description:
> Origin policy: implement window.originPolicyIds
>
> This plumbs the origin policy IDs from the net-side OriginPolicyContents
> through to the renderer, where they get exposed on Window. This does not
> yet tackle WorkerGlobalScope, but it does add idlharness tests for it,
> which fail for now.
>
> Bug: 1057123
> Change-Id: Ie611f03bab99ccdaa6221733d9305cf2323a129e
> Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2089994
> Commit-Queue: Domenic Denicola <domenic@chromium.org>
> Reviewed-by: Daniel Cheng <dcheng@chromium.org>
> Reviewed-by: James MacLean <wjmaclean@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#747893}
TBR=dcheng@chromium.org,domenic@chromium.org,wjmaclean@chromium.org
# Not skipping CQ checks because original CL landed > 1 day ago.
Bug: 1057123
Change-Id: I341fd06c5587bc05ea110307251ad9eb19365433
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2093453
Reviewed-by: Maggie Cai <mxcai@chromium.org>
Commit-Queue: Maggie Cai <mxcai@chromium.org>
Cr-Commit-Position: refs/heads/master@{#748110}
--
wpt-commits: b10ddff4a774ac4400a3d9b6f283caf5e5cf7af6
wpt-pr: 22132
Automatic update from web-platform-tests
Adds WPT tests for Imperative Shadow DOM Slot API.
Previous to this CL, Imperative shadow DOM Slot API was put on hold.
A previous implementation is still in place but disabled. Since then,
in 2019 TPAC F2F, the Imperative Slotting API behavior has been updated.
This CL adds new test cases which reflects this new behavior for
Imperative Shadow DOM API. They are turned off for now, but will be
enabled one by one as its functionality gets implemented.
This is the post for I2P:
https://groups.google.com/a/chromium.org/forum/#!topic/blink-dev/TNYIAu5E_M4
Bug: 869308
Change-Id: Ie875a3eb2b349942b6aa51af387460c470572491
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2083742
Commit-Queue: Yu Han <yuzhehan@chromium.org>
Reviewed-by: Mason Freed <masonfreed@chromium.org>
Cr-Commit-Position: refs/heads/master@{#748099}
--
wpt-commits: cab39d8d9a00eabf738a9897d71d5c0a69322fb2
wpt-pr: 22053
Automatic update from web-platform-tests
Remove wpt/feature-policy/parameters tests
This CL removes tests under directory
third_party/blink/web_tests/external/wpt/feature-policy/parameters/
because params are no longer to be supported for feature policy.
All parameterized policy are going to be migrated to document policy.
Change-Id: I758f3834459bfb0404b4edf09d31af190b619c7e
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2089618
Commit-Queue: Charlie Hu <chenleihu@google.com>
Reviewed-by: Ian Clelland <iclelland@chromium.org>
Cr-Commit-Position: refs/heads/master@{#748060}
--
wpt-commits: e9a8307cee32a6a5ee136841443b7080fdc8661c
wpt-pr: 22122
Automatic update from web-platform-tests
Cleanup min max (#22127)
* Use animation-play-state:paused rather than enormous durations. Add a red square to make failure obvious.
* Add a red square for obvious failure.
* Add a red square for obvious failure.
* Don't depend on the serialization of computed units being a particular thing.
* Math functions are aggressively reduced away if possible.
* Simplify with test_math_used().
* Add the ability, which other test functions have, to tack on an extra message to the built-in message.
* Simplify with test_math_used().
* Use 'margin-left' instead of 'left', since the latter's resolved value is the computed value if the element isn't positioned!
* Simplify with test_math_used().
* Restructure test_math_used() to accept a type specifier, and automatically choose a good property/base value/etc for that type. Also add test_math_computed, with different props where appropriate.
* Switch the inf/zero/nan functions over to just using the 'number' type so they're more obvious.
* Improve the documentation for numeric-testcommon.js.
* Make minmax-length-percent-serialize match the spec for serialization of computed values.
* Make minmax-length-serialize match the spec for serialization of computed values.
* Switch minmax-number-computed over to test_math_used().
* Fix errors in test_math_computed(), and add test_math_specified().
* Extract the base-selecting code, since it's shared by all the functions.
* Fix minmax-number-serialize to match spec for serialization.
* Fix minmax-percentage-computed to not rely on serialization.
* Well, the generic functions dont' let me test exact serialization, so start writing some that do.
* Add new serializ-testcommon helper, and rewrite all the serialization tests to use it.
* Remove the 'base' value from numeric-testcommon, as empty string works consistently across everything.
* [css-values] Final fixes/conversion of min()/max() tests to the newer templates and spec.
--
wpt-commits: 53876e32d827db82f4b7af38053529302c243d40
wpt-pr: 22127
Automatic update from web-platform-tests
[wakelock] Upstream remaining tests to WPT
Now that test_runner.set_permission() is available these Wake Lock API
tests can be upstreamed to Web Platform Tests.
Bug: 1015327
Change-Id: I7bb1c7cbc458c8b7b05c0c2361fbe5ea50444e8b
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2089968
Reviewed-by: Raphael Kubo da Costa <raphael.kubo.da.costa@intel.com>
Commit-Queue: Reilly Grant <reillyg@chromium.org>
Auto-Submit: Reilly Grant <reillyg@chromium.org>
Cr-Commit-Position: refs/heads/master@{#747928}
--
wpt-commits: 7c13aadc25f45b5075c9c3ef39241a27c86a25eb
wpt-pr: 22109
Automatic update from web-platform-tests
Origin policy: implement window.originPolicyIds
This plumbs the origin policy IDs from the net-side OriginPolicyContents
through to the renderer, where they get exposed on Window. This does not
yet tackle WorkerGlobalScope, but it does add idlharness tests for it,
which fail for now.
Bug: 1057123
Change-Id: Ie611f03bab99ccdaa6221733d9305cf2323a129e
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2089994
Commit-Queue: Domenic Denicola <domenic@chromium.org>
Reviewed-by: Daniel Cheng <dcheng@chromium.org>
Reviewed-by: James MacLean <wjmaclean@chromium.org>
Cr-Commit-Position: refs/heads/master@{#747893}
--
wpt-commits: ad9423bf2edd246111d00e6b7bd7f35359bf501f
wpt-pr: 22106
Automatic update from web-platform-tests
[wasm] Stricter MIME type test on WebAssembly streaming APIs
The WebAssembly Web API spec[1] prescribes that the responses for
`WebAssembly.{compile,instantiate}Streaming()` must have the
`Content-Type` header `application/wasm` without any extra parameters.
So far we ignored extra parameters, causing interop
issues with Firefox.
[1]: https://www.w3.org/TR/wasm-web-api-1/#streaming-modules
Bug: 1057973
Change-Id: I2273e2b5577dc6a11343536d55fa72b9e518c0f2
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2085257
Commit-Queue: Surma <surma@chromium.org>
Reviewed-by: Andreas Haas <ahaas@chromium.org>
Reviewed-by: Kenichi Ishibashi <bashi@chromium.org>
Cr-Commit-Position: refs/heads/master@{#747693}
--
wpt-commits: 8c6d6497707871cf1c073bae65ae06102b569167
wpt-pr: 22098
Automatic update from web-platform-tests
Make more robust the timeout check WebDriverProtocol.is_alive()
In 58988ab31a (#21889) the WebDriverProtocol.is_alive() check was modified
to use a 2 second timeout, but this only works when the connection with
the webdriver has been lost and a new one will be made for the is_alive()
check because calling socket.getdefaulttimeout() doesn't affect the ongoing
httplib connection with the WebDriver.
To fix this we introduce the timeout value as an optional parameter of the
WebDriver send_command() function. When timeout its specified, then the
socket timeout of the current httplib connection its modified just only
for this request (and restored after). This works both when the connection
with the WebDriver is lost (and a new one its made) as when the connection
it is working as expected.
--
wpt-commits: 5b7d10d8b299774f420ffb452d70ad9c7ead92c7
wpt-pr: 22044
Automatic update from web-platform-tests
[css-grid] Fix referencing implicit grid line after auto repeat()
The indices of both implicit grid lines defined with grid-template-areas
and explicit ones defined with grid-template-rows/columns used to be
stored together in NamedGridColumnLines and NamedGridRowLines.
However, this was problematic, because the former indices already refer
to the final explicit grid so they don't have to be increased when
expanding an auto repeat(), but the latter ones should.
Therefore, this patch stores the indices in separate fields and uses the
correct logic for each one.
BUG=966090
TEST=external/wpt/css/css-grid/placement/grid-placement-using-named-grid-lines-005.html
TEST=external/wpt/css/css-grid/placement/grid-placement-using-named-grid-lines-006.html
Change-Id: I6d423148af0e4dd865f130742f7a927a325cef90
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2081823
Reviewed-by: Javier Fernandez <jfernandez@igalia.com>
Commit-Queue: Oriol Brufau <obrufau@igalia.com>
Cr-Commit-Position: refs/heads/master@{#747669}
--
wpt-commits: fa8785321b8e3a1baa7bf08423a2c7df3eea5291
wpt-pr: 22062
Automatic update from web-platform-tests
[css-grid] Fix grid placement with '<integer> && <custom-ident>'
The logic for placing a grid item with both an integer and an identifier
in a placement property was interacting wrong with auto repeat() tracks.
This patch uses the same proper logic that was already implemented in
OrderedNamedLinesCollectorInGridLayout::CollectLineNamesForIndex
BUG=966090
TEST=external/wpt/css/css-grid/placement/grid-placement-using-named-grid-lines-004.html
Change-Id: I042ce1fd77a79959240d272d96eac21f06e014ac
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2082917
Reviewed-by: Javier Fernandez <jfernandez@igalia.com>
Commit-Queue: Oriol Brufau <obrufau@igalia.com>
Cr-Commit-Position: refs/heads/master@{#747631}
--
wpt-commits: fef449a2cfb2ae75cc8cf189ed932a30061a35a5
wpt-pr: 22060
Automatic update from web-platform-tests
Fix glyph assembly data for axisheight5000-verticalarrow14000.woff (#22084)
* Make full advance of the "bot" match the glyph height (19em).
instead of the incorrect 1em value.
* Force min overlap to be 0 and max part overlap to be 1em, so
that GlyphAssembly is valid per [1] and the engine has
flexibility to adjust the overlap in order to match the target
size.
[1] https://mathml-refresh.github.io/mathml-core/#the-glyphassembly-table
"For each GlyphPartRecord in GlyphAssembly.partRecords, the values of GlyphPartRecord.startConnectorLength and GlyphPartRecord.endConnectorLength must be at least omin. Otherwise, it is not possible to satisfy the condition of MathVariant.minConnectorOverlap."
--
wpt-commits: 73c32258a4380eb42452e3f2da7f6bc01d57ef0e
wpt-pr: 22084
Automatic update from web-platform-tests
[WPT/common/security-features] Add default spec.src.json
There were multiple `spec.src.json` for each project directories
that have many things in common, e.g. lists of subresources,
source contexts, etc.
This CL introduces `common/security-features/tools/spec.src.json`
that works as the default `spec.src.json`, and moves
common things into the default `spec.src.json`.
By doing this, we can ensure uniform test coverage across projects
(e.g. the same set of subresource requests are covered in
`referrer-policy` and `mixed-content` tests) more easily.
To keep the generated tests the same, this CL adds
exclusion rules named "... not yet tested".
This CL doesn't affect test behavior nor generated results.
Bug: 906850
Change-Id: Id06bc6e311a482f50531cd36e6da57b2e6df22ec
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2049953
Commit-Queue: Hiroshige Hayashizaki <hiroshige@chromium.org>
Reviewed-by: Hiroki Nakagawa <nhiroki@chromium.org>
Reviewed-by: Kenichi Ishibashi <bashi@chromium.org>
Reviewed-by: Mike West <mkwst@chromium.org>
Cr-Commit-Position: refs/heads/master@{#747577}
--
wpt-commits: 6d7f862e4afa43d45bb3f800c457e9f746cd8730
wpt-pr: 22079
Automatic update from web-platform-tests
Migrate font-display to document policy
This CL migrates font-display from feature policy to document policy.
Note: font-display-feature-polcy-01.tentative.html was migrated to document policy in a previous CL(https://chromium-review.googlesource.com/c/chromium/src/+/2017927)
Bug: 993790
Change-Id: Id4c822e0a42594ab75919819e4f1b78ca6a263f3
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2036518
Commit-Queue: Charlie Hu <chenleihu@google.com>
Reviewed-by: Daniel Cheng <dcheng@chromium.org>
Reviewed-by: Ian Clelland <iclelland@chromium.org>
Cr-Commit-Position: refs/heads/master@{#747448}
--
wpt-commits: bf43d143408caf60fc56d7a27470185798572221
wpt-pr: 21807
Edwin Takahashi
Ricky Stewart
Erik Nordin
Geoff Brown
Emilio Cobos Álvarez
Valentin Gosu
Boris Chiou
Nicholas Nethercote
Florin Strugariu
Andrew Halberstadt
moz-wptsync-bot
Rob Buis
Yifan Luo
Fredrik Söderquist
Makoto Shimazu
Maggie Cai
Yu Han
Mike West
Domenic Denicola
Charlie Hu
Tab Atkins Jr
Reilly Grant
Yoav Weiss
Surma
Philip Jägenstedt
Oriol Brufau
Frédéric Wang
Hiroshige Hayashizaki
autofoolip