Even with the fix above, we get the caret baseline wrong because the
line is non-empty, but the baseline is all the way to the top.
Maybe an alternative to this would be to move the line-height hack to
the beginning of layout but that sounds a bit more fishy.
Otherwise we should do this more often, I suspect. I added one test that
fails because of nsInlineFrame... Anyhow probably follow-up material?
Depends on D73362
Differential Revision: https://phabricator.services.mozilla.com/D73363
The internal logic accesses adapters during command recording.
This PR makes up to keep that reference for while the device lives.
Differential Revision: https://phabricator.services.mozilla.com/D73278
This was an issue around "six" and "click", this thread sheds some light:
https://github.com/pallets/click/issues/564.
This issue was reproducible by having a local python 2 flask app with Flask==1.0.2, Werkzeug==0.14.1 and Click==7.0. Upon self-shutdown, the TypeError was raised.
Upgrading Click to 7.1.2 resolves this issue both in that separate "reproduce app" and here in-tree.
Differential Revision: https://phabricator.services.mozilla.com/D73351
This check was done in GetNextNonEmptyTextFragmentInSameBlock which I
removed in bug 1627643.
Hopefully the fuzzer finds an easy reduced test-case to land as a
crashtest, otherwise I can try to construct one.
Differential Revision: https://phabricator.services.mozilla.com/D72926
Automatic update from web-platform-tests
LazyLoad: Fix feature-policy tests
This updates the lazyload attribute as 'loading' with values of auto,
eager and lazy for the feature policy tests.
TBR=hiroshige@chromium.org
Fixed: 1074608
Change-Id: If8ba1d09bce518ac36b70bd2a7b01687ffec87c1
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2166638
Commit-Queue: rajendrant <rajendrant@chromium.org>
Reviewed-by: Scott Little <sclittle@chromium.org>
Cr-Commit-Position: refs/heads/master@{#763010}
--
wpt-commits: ddad50b245ec811f4cbce8c84639122e9a7732a9
wpt-pr: 23250
Automatic update from web-platform-tests
Don't serialize closed shadow roots with getInnerHTML()
This CL corrects the behavior of getInnerHTML() so that closed
shadow roots are not serialized. It also adds a test of the
getInnerHTML() function.
Bug: 1042130
Change-Id: Iac214b16799cee45bcd90be8f618cc5560eb1db4
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2166383
Commit-Queue: Kouhei Ueno <kouhei@chromium.org>
Reviewed-by: Kouhei Ueno <kouhei@chromium.org>
Auto-Submit: Mason Freed <masonfreed@chromium.org>
Cr-Commit-Position: refs/heads/master@{#763236}
--
wpt-commits: c88ed219c1a65af9f83844181b84111e9462f8b6
wpt-pr: 23243
Automatic update from web-platform-tests
[css-speech] rename Integer.html to voice-family-integer.html (#23060)
Renamed because it's one of few tests with capitalized filenames.
--
wpt-commits: b5e6c40a2bd4c4ae8bd64dad3f3732c023f150b4
wpt-pr: 23060
Automatic update from web-platform-tests
[FragmentItem] Prevent dangling LayoutObject ptrs in block fragmentation.
NGInlineCursor doesn't work with block fragmentation, so we'd end up
with NGFragmentItem objects with dangling pointers to LayoutObject. To
fix this, perform a full walk through all fragment items in all
fragments, if it's likely that we're in block fragmentation.
CL:2152824 caused two tests to crash because of this, and it therefore
got reverted. Add another test now, which would also crash without the
aforementioned CL.
Bug: 1075517
Change-Id: I97e2fae63d89031ea258ba146898e7fcfe56bd45
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2165800
Reviewed-by: Koji Ishii <kojii@chromium.org>
Commit-Queue: Morten Stenshorne <mstensho@chromium.org>
Cr-Commit-Position: refs/heads/master@{#763233}
--
wpt-commits: e9ed833d0c9483012493d839434137de7e7ba00d
wpt-pr: 23279
Automatic update from web-platform-tests
Treat floorOrigin as mojoFromFloor
Per the webxr-test-api all origins are based on the "local" space when
the "session" started; which means that all of our origins passed in are
inherently "mojoFromFoo." The standingTransform (in need of renaming)
is, in blink and by devices, floorFromMojo. Since the tests were
directly setting floorOrigin to the standingTransform, the tests were
incorrectly treating floor origin as floorFromMojo. This change inverts
the floorOrigin before assigning it to the standingTransform, so that
Chrome begins treating the floorOrigin as mojoFromFloor. Given that
Chrome was the only ones running these particular tests, the output
values were *also* written as if the passed in floorOrigins were
floorFromMojo. In order to stick with the expectations, this inverts
the input transforms as well.
In fixing this, a bug was discovered in XRMathHelper.inverse wherein the
comatrix was not created properly. This change also addresses that bug.
Fixed:1071630
Change-Id: I1626d80099169c4ceba0df68ad84bbf181f76647
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2165881
Auto-Submit: Alexander Cooper <alcooper@chromium.org>
Commit-Queue: Piotr Bialecki <bialpio@chromium.org>
Reviewed-by: Piotr Bialecki <bialpio@chromium.org>
Cr-Commit-Position: refs/heads/master@{#763104}
--
wpt-commits: 6a2f0e6a0110fd0830fc3196680f671bb83c2747
wpt-pr: 23272
Automatic update from web-platform-tests
Adds slotchange event for Imperative Slot API.
Prior to this CL slotchange event is not fired when calling the
Imperative Slot API.
This CL adds the ability to fire slotchange event for Imperative Slot
API. It does so by detecting candidates nodes changed during the
call and fires slotchange event if changed. This CL only implements the
assign() function's portion of the slotchange event. In the next CL,
I will implement slotchange event for removing nodes and slots.
Bug: 869308
Change-Id: I13851d1788fc6cf8b0ea109fb136a2f029e61e70
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2161369
Reviewed-by: Mason Freed <masonfreed@chromium.org>
Reviewed-by: Hayato Ito <hayato@chromium.org>
Commit-Queue: Yu Han <yuzhehan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#763111}
--
wpt-commits: ff47fb409e6e7ef72893ec6123f6abde845372db
wpt-pr: 23181
Automatic update from web-platform-tests
[wpt] enhanceents to check-layout-th.js (#23184)
There are 3 enhancements that I've found useful.
1. Warn if there are any unexpected data-* attributes
A common error is to have a typo in data-* attribute name.
When this happens, test will pass, because attribute is never read.
This enhancement prints a warning if unexpected attribute is found.
Running this over the entire test suite, I've found several errors:
css/css-grid/grid-definition/grid-auto-repeat-intrinsic-001.html
css/css-grid/grid-items/grid-items-percentage-margins-vertical-rl-001.html
css/css-grid/grid-items/grid-items-percentage-margins-vertical-rl-002.html
css/css-grid/grid-items/grid-items-relative-offsets-001.html
css/css-grid/grid-items/grid-items-relative-offsets-002.html
have data-expected-x, data-expected-y
Tried replacing some of these with data-offset-x, and tests fail.
css/css-grid/grid-definition/grid-percentage-rows-indefinite-height-002
has data-offset-top
2. highlight error on failure flag. Defaults to false.
If set, this flag will draw an outline around failed Element.
Useful when debugging pages with many elements.
3. toggle printing dom on error.
Sometimes, it is useful not to print dom on error for clarity.
Change-Id: I8424aa2dc002f63fde18d0fc54bdfe77666ce951
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2161379
Reviewed-by: Philip Jägenstedt <foolip@chromium.org>
Reviewed-by: Manuel Rego <rego@igalia.com>
Commit-Queue: Aleks Totic <atotic@chromium.org>
Cr-Commit-Position: refs/heads/master@{#763069}
Co-authored-by: Aleks Totic <atotic@chromium.org>
--
wpt-commits: a5a3b7353c884ebcca0230818632a42ad2fbd9b2
wpt-pr: 23184
Automatic update from web-platform-tests
Effective Touch Action when <body> is the viewport defining element
When <body> is deemed to be the viewport defining element <body>'s
overflow needs to be set to visible. Because of this, during effective
touch action calculation we consider <body> to be a scroller when it is
not.
The fix attempts to detect when <body> is deemed the viewport defining
element and correct the effective touch action calculation.
Bug: 1031745
Change-Id: Ie9a3c966af78fdbe0dc3e57f3f192aca9110fd0e
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2152950
Commit-Queue: Liviu Tinta <liviutinta@chromium.org>
Reviewed-by: David Bokan <bokan@chromium.org>
Reviewed-by: Rune Lillesveen <futhark@chromium.org>
Cr-Commit-Position: refs/heads/master@{#763147}
--
wpt-commits: 1650a7c1569ad9e3e10bc1da400f9a44230a5a78
wpt-pr: 23216
Automatic update from web-platform-tests
Improve web engine detection in webxr_util.js (#23200)
The check (window.XRTest === undefined) is not valid to discriminate between Chromium and other browsers. Improve engine detection so that we don't end up accessing entities that do not exist in other engines (like navigator.xr.test.Debug).
--
wpt-commits: d9312410d445b129727af4fbd4aa8733e99b0e64
wpt-pr: 23200
Automatic update from web-platform-tests
*.any.js: explicitly list defaults, part 1 (#23117)
* *.any.js: explicitly list defaults, part 1
Helps with #23111.
* add a couple more
--
wpt-commits: e1e713c842e54ea0a9410ddc988b63d0e1d31973
wpt-pr: 23117
Automatic update from web-platform-tests
[wptserve] Fix an exception handler in stash
internal_key is a tuple whose items are guaranteed to be strings.
It should be used as the key directly.
--
wpt-commits: d2025b24b0b1f4d2e69e83cf83e60c021d9f7b42
wpt-pr: 23257
Automatic update from web-platform-tests
UpgradeInsecureRequest: Add WPTs for static-import in dedicated/shared workers
This CL adds upgrade-insecure-request wpts to test static-import from
dedicated workers and shared workers.
The test scheme is implemented in the previous CL:
https://chromium-review.googlesource.com/c/chromium/src/+/2102036.
This CL only removes 'worker-import' and 'sharedworker-import' from
the excluded subresource list in spec.src.json and auto-generates tests.
Bug: 989399
Change-Id: I3b799d7743306a8b5fca0d7292360a88d9b36ded
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2117591
Reviewed-by: Hiroki Nakagawa <nhiroki@chromium.org>
Reviewed-by: Hiroshige Hayashizaki <hiroshige@chromium.org>
Commit-Queue: Eriko Kurimoto <elkurin@google.com>
Cr-Commit-Position: refs/heads/master@{#753164}
--
wpt-commits: 19a0095b86b988a1afa5833c691a102ccde8cd68
wpt-pr: 22420
Automatic update from web-platform-tests
Fix report-to CSP directive to only allow one endpoint
According to https://w3c.github.io/webappsec-csp/#directive-report-to,
the Content-Security-Policy directive "report-to" should only accept one
token (endpint). However, our previous implementation allowed several
endpoints to be specified.
Bug: 916265
Change-Id: Ie11ee736f577d015921a5291824dcedcbc790177
Fixed: 916265
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2162826
Reviewed-by: Arthur Sonzogni <arthursonzogni@chromium.org>
Reviewed-by: Mike West <mkwst@chromium.org>
Commit-Queue: Antonio Sartori <antoniosartori@chromium.org>
Cr-Commit-Position: refs/heads/master@{#762813}
--
wpt-commits: fa3d5b0da213daee43f5de9621a26984ef70d149
wpt-pr: 23230
