Bug 285754

Summary:

Element with `overflow: scroll` does not correctly account for the right margin of its child to determine the scrollable area

Product:

WebKit

Reporter:

Antoine Quint <graouts>

Component:

Layout and Rendering

Assignee:

alan <zalan>

Status:

RESOLVED FIXED

Severity:

Normal

CC:

bfulgham, fujii.hironori, karlcow, kiet.ho, koivisto, ntim, simon.fraser, webkit-bug-importer, zalan

Priority:

Keywords:

InRadar

Version:

WebKit Nightly Build

Hardware:

Unspecified

OS:

Unspecified

See Also:

https://bugs.webkit.org/show_bug.cgi?id=286149
https://bugs.webkit.org/show_bug.cgi?id=287638

Bug Depends on:

286149, 287350

Bug Blocks:

Attachments:

Description	Flags
Test	none
Patch	none
Patch	none
Patch	none
Patch	none
[fast-cq]Patch	none

Antoine Quint

Reported 2025-01-10 09:50:17 PST

Created attachment 473859 [details] Test Look at the attached test case which is this simple HTML and CSS: ``` <!DOCTYPE html> <style> #scroller { overflow: scroll; width: 200px; height: 200px; background-color: gray; } #target { margin: 200px; width: 100px; height: 100px; background-color: green; } </style> <div id="scroller"> <div id="target"></div> </div> ``` In Chrome, `scroller.scrollWidth` returns `500` while in Safari Technology Preview 210 it returns `300`, as if `margin-right` for the `#target` was `0` and not `200px`.

Attachments
Test (351 bytes, text/html) 2025-01-10 09:50 PST, Antoine Quint	no flags	Details
Patch (6.18 KB, patch) 2025-02-09 06:56 PST, alan	no flags	Details Formatted Diff Diff
Patch (6.18 KB, patch) 2025-02-10 07:00 PST, alan	no flags	Details Formatted Diff Diff
Patch (6.22 KB, patch) 2025-02-10 07:48 PST, alan	no flags	Details Formatted Diff Diff
Patch (14.78 KB, patch) 2025-02-10 14:59 PST, alan	no flags	Details Formatted Diff Diff
[fast-cq]Patch (16.51 KB, patch) 2025-02-10 17:23 PST, alan	no flags	Details Formatted Diff Diff
Show Obsolete (4) View All Add attachment proposed patch, testcase, etc.

Antoine Quint

Comment 1 2025-01-10 09:51:07 PST

This affects the WPT test `scroll-animations/css/view-timeline-inset-animation.html` which fails to scroll deep enough in the scroller to reach the expected animation values.

Simon Fraser (smfr)

Comment 2 2025-01-10 10:13:12 PST

This is about overflow computation.

Radar WebKit Bug Importer

Comment 3 2025-01-10 10:13:34 PST

<rdar://problem/142695395>

alan

Comment 4 2025-02-09 06:56:24 PST

Created attachment 474173 [details] Patch

alan

Comment 5 2025-02-10 07:00:58 PST

Created attachment 474176 [details] Patch

alan

Comment 6 2025-02-10 07:48:54 PST

Created attachment 474177 [details] Patch

alan

Comment 7 2025-02-10 14:59:06 PST

Created attachment 474194 [details] Patch

alan

Comment 8 2025-02-10 17:23:31 PST

Created attachment 474195 [details] [fast-cq]Patch

EWS

Comment 9 2025-02-11 05:19:21 PST

Committed 290207@main (dd84bf8fcfcc): <https://commits.webkit.org/290207@main> All reviewed patches have been landed. Closing bug and clearing flags on attachment 474195 [details].

Fujii Hironori

Comment 10 2025-02-11 21:55:25 PST

imported/w3c/web-platform-tests/css/css-anchor-position/anchor-scroll-nested.html is unexpectedly passing after 290207@main. History: https://results.webkit.org/?suite=layout-tests&test=imported%2Fw3c%2Fweb-platform-tests%2Fcss%2Fcss-anchor-position%2Fanchor-scroll-nested.html I'm going to unmark it.

Fujii Hironori

Comment 11 2025-02-11 21:58:04 PST

Re-opening for pull request https://github.com/WebKit/WebKit/pull/40469

alan

Comment 12 2025-02-12 05:15:20 PST

(In reply to Fujii Hironori from comment #10) > imported/w3c/web-platform-tests/css/css-anchor-position/anchor-scroll-nested. > html is unexpectedly passing after 290207@main. > History: > https://results.webkit.org/?suite=layout-tests&test=imported%2Fw3c%2Fweb- > platform-tests%2Fcss%2Fcss-anchor-position%2Fanchor-scroll-nested.html > I'm going to unmark it. Thank you.

EWS

Comment 13 2025-02-12 11:34:35 PST

Test gardening commit 290284@main (0385ef866f07): <https://commits.webkit.org/290284@main> Reviewed commits have been landed. Closing PR #40469 and removing active labels.

Kiet Ho

Comment 14 2025-02-13 11:39:39 PST

This test has been flaky for a very long time I believe, so it'd be better to mark it as flaky instead of passing.

Kiet Ho

Comment 15 2025-02-13 11:45:00 PST

(please disregard my comment, I got the test wrong)

Note You need to log in before you can comment on or make changes to this bug.