Bug 94014 – Deleting a button element leaves the style inside the button element

• Status: RESOLVED
• Resolution: FIXED
• Component: HTML Editing

[Ryosuke Niwa]

Insert a button element in a contenteditable element, and delete it by a backspace, then insert some text. The style inside the button bleeds outside the button so that the text will be centered for example.

[Ryosuke Niwa]

Reduction: http://www.drlongghost.com/dlg/chrome_bug

http://crbug.com/86938

[Sukolsak Sakshuwong]

Created attachment 158488 [details]
Patch

[WebKit Review Bot]

Comment on attachment 158488 [details]
Patch

Attachment 158488 [details] did not pass chromium-ews (chromium-xvfb):
Output: http://queues.webkit.org/results/13492821

New failing tests:
editing/deleting/delete-button.html

[WebKit Review Bot]

Created attachment 158491 [details]
Archive of layout-test-results from gce-cr-linux-05

The attached test failures were seen while running run-webkit-tests on the chromium-ews.
Bot: gce-cr-linux-05  Port: <class 'webkitpy.common.config.ports.ChromiumXVFBPort'>  Platform: Linux-2.6.39-gcg-201203291735-x86_64-with-Ubuntu-10.04-lucid

[Sukolsak Sakshuwong]

Created attachment 158492 [details]
Patch

[Ryosuke Niwa]

Comment on attachment 158492 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=158492&action=review

> Source/WebCore/ChangeLog:9
> +        Make the DeleteSelectionCommand save typing style only for formattable nodes
> +        according to the HTML Editing specification.

I'm not sure what you're referring here but given that our implementation doesn't match the specification at all, we should probably follow our convention for now.
Could you explain what you're trying to do here in more detail?

[Ojan Vafai]

Comment on attachment 158492 [details]
Patch

sukolsak no longer works at Google. Removing the r? since he clearly isn't around to complete this patch.

[Neil Ashford]

Hi,
I've discovered what is essentially the same bug using a regular <input> element, thought I'd try to bump this issue instead of making a duplicate.

Reproduction case: https://fiddle.tiny.cloud/r8haab (ignore that it's a TinyMCE fiddle, this just uses regular content editable elements).

Safari Version: Version 14.0.3 (16610.4.3.1.7)

[Ahmad Saleem]

[e-mail] - We tried something similar in this pull request by trying to merge "Blink patch".

https://github.com/WebKit/WebKit/pull/3675

Do you think the test case from this patch would suffice? Thanks!

BY THE WAY they are are duplicates of each other, so I can mark this later on. Thanks!

[Ryosuke Niwa]

They're related but not a duplicate. Button element isn't atomic.

[Ahmad Saleem]

(In reply to Ryosuke Niwa from comment #10)
> They're related but not a duplicate. Button element isn't atomic.

but the Chromium bug in Comment 1 takes me to this commit:

https://src.chromium.org/viewvc/blink?view=revision&revision=154067

which is same I was trying to merge in PR

https://chromium.googlesource.com/chromium/blink/+/27371d2980715475f3443f7025141ebff8537179

______

I think the patch attached to this is of different bug and Comment 1 is referring to something else.

[Ahmad Saleem]

I took the test case from Patch and changed it into JSFiddle:

Link - https://jsfiddle.net/zvk1juo4/

Only Chrome Canary 107 is showing it as PASS while both Firefox Nightly 107 and STP 152 is showing it as "FAILED".

[Ahmad Saleem]

Trying my luck through landing the again - https://github.com/WebKit/WebKit/pull/4009

Fingers crossed!

[EWS]

Committed 254285@main (c75ccbd4e966): <https://commits.webkit.org/254285@main>

Reviewed commits have been landed. Closing PR #4009 and removing active labels.

[Radar WebKit Bug Importer]

<rdar://problem/99715641>