2022187 - [wpt-sync] Sync PR 58385 - [line-clamp] Allow associating line-clamp ellipsis with forced breaks

Status: RESOLVED FIXED

(Core :: Layout: Scrolling and Overflow, task, P4)

Description

[Web Platform Test Sync Bot [:wpt-sync] (Matrix: #interop:mozilla.org)]

Sync web-platform-tests PR 58385 into mozilla-central (this bug is closed when the sync is complete).

PR: https://github.com/web-platform-tests/wpt/pull/58385
Details from upstream follow.

Andreu Botella <[e-mail]> wrote:

[line-clamp] Allow associating line-clamp ellipsis with forced breaks

In https://crrev.com/c/6802120, line-clamp ellipses added as part of
line-breaking (with the CSSLineClampLineBreakingEllipsis runtime
flag enabled) were associated with a layout object, corresponding to
the closest inline item before the ellipsis, if any. This was
important so the ellipsis would correctly repaint.

However, not all inline items worked to associate an ellipsis with
(e.g. out-of-flow or ruby items). That CL added a
CanUseItemForNeedsPaint function to check this. And importantly,
this function returns false for items that don't have
TextItemType::kNormal. Ideally this should have resulted in the same
layout object as an ellipsis produced by LineTruncator.

As it turns out, when the line to ellipsize ends in a forced line
break (an inline item which has TextItemType::kForcedLineBreak),
this does not match LineTruncator, which will use the layout object
corresponding to the forced break.

In most cases, this difference doesn't matter. However, when the
start tag for a position: relative inline is immediately before the
forced break, it does. This is because the inline creates a new layer,
and detecting visual overflow on it (through
LayoutBoxModelObject::RecalcVisualOverflow) finds all inline items
which are associated with that layout object, including the ellipsis.
Since such inline items should correspond to inline boxes, this breaks
some of the assumptions of such code and triggers assertion failures.

This CL fixes this by removing the TextItemType check.

Bug: 40336192, 489494018
Change-Id: Iaabe25e9e65f8c1dddbe2b40a6b811e45602e3b2
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/7649894
Reviewed-by: Koji Ishii \<[e-mail]>
Commit-Queue: Andreu Botella \<[e-mail]>
Cr-Commit-Position: refs/heads/main@{#1596921}

Comment 1

[Web Platform Test Sync Bot [:wpt-sync] (Matrix: #interop:mozilla.org)]

Pushed to try https://treeherder.mozilla.org/#/jobs?repo=try&revision=4b70f9034dbb5e371bd4ff6f2927c8aa5cd96b5a

Comment 2

[Web Platform Test Sync Bot [:wpt-sync] (Matrix: #interop:mozilla.org)]

The PR was not expected to affect any tests, but the try push wasn't a success. Check the try results for infrastructure issues

Comment 3

[Pulsebot]

Pushed by [e-mail]:
https://github.com/mozilla-firefox/firefox/commit/9bc4dacb2308
https://hg.mozilla.org/integration/autoland/rev/fde824ead5b2
[wpt PR 58385] - [line-clamp] Allow associating line-clamp ellipsis with forced breaks, a=testonly

Comment 4

[amarc]

https://hg.mozilla.org/mozilla-central/rev/fde824ead5b2