Bug 75330 – REGRESSION (r92823): Background color not preserved when copying and pasting a table row

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

Created attachment 120708 [details]
Email message for reproducing the bug

Steps to reproduce:
0. In Mail preferences > Compositing, select “Include selected text, if any; otherwise include all text”
1. Open the attached patch.eml in Mail
2. Tripe-click the green line to select it in its entirety
3. Choose Message > Reply

Expected results:
In the new compose window, the quoted row maintains its green background color.

Actual result:
In the new compose window, the quoted row is missing the green background color.

Regression:
Caused by <http://trac.webkit.org/r92823>

Notes:
Alternatively, you can reproduce this by opening the message in Mail, choosing Message > Send again, selecting the row, cutting it, and pasting it, or even copying and pasting in-place.

[Radar WebKit Bug Importer]

<rdar://problem/10630181>

[Ryosuke Niwa]

Created attachment 121020 [details]
reduction

Reproduction steps:
1. Open the attached file
2. Copy
3. Paste inside iframe

[Ryosuke Niwa]

Created attachment 121029 [details]
fixes the bug

[Tony Chang]

Comment on attachment 121029 [details]
fixes the bug

View in context: https://bugs.webkit.org/attachment.cgi?id=121029&action=review

> Source/WebCore/editing/EditingStyle.cpp:99
> +static PassRefPtr<CSSMutableStyleDeclaration> editingStyleFromComputedStyle(PassRefPtr<CSSComputedStyleDeclaration> style, bool includeNonInheritableProperties = false)

Nit: Should we convert includeNonInheritableProperties to an enum?

> LayoutTests/editing/pasteboard/copy-element-with-conflicting-background-color-from-rule.html:5
> +The pated text should have lightgreen background color.</p>

pated -> pasted

[Ryosuke Niwa]

Committed r104068: <http://trac.webkit.org/changeset/104068>

[Csaba Osztrogonác]

(In reply to comment #5)
> Committed r104068: <http://trac.webkit.org/changeset/104068>

It broke 12 tests on GTK and on Qt: 
http://build.webkit.org/results/GTK%20Linux%2064-bit%20Release/r104068%20(15677)/results.html

[Ryosuke Niwa]

(In reply to comment #6)
> (In reply to comment #5)
> > Committed r104068: <http://trac.webkit.org/changeset/104068>
> 
> It broke 12 tests on GTK and on Qt: 
> http://build.webkit.org/results/GTK%20Linux%2064-bit%20Release/r104068%20(15677)/results.html

I'm confused. Qt bot hasn't even caught up yet. Also, my patch can't affect non-editing tests as EditingStyle isn't used anywhere but in editing.

[Csaba Osztrogonác]

(In reply to comment #7)
> I'm confused. Qt bot hasn't even caught up yet. Also, my patch can't affect non-editing tests as EditingStyle isn't used anywhere but in editing.

Our internal bots caught your patch: http://build.webkit.sed.hu/waterfall?category=core

[Ryosuke Niwa]

(In reply to comment #8)
> (In reply to comment #7)
> > I'm confused. Qt bot hasn't even caught up yet. Also, my patch can't affect non-editing tests as EditingStyle isn't used anywhere but in editing.
> 
> Our internal bots caught your patch: http://build.webkit.sed.hu/waterfall?category=core

Oops, that's because I have a busted Element.cpp change :( Didn't see that. Will revert it in a minute.

[Ryosuke Niwa]

(In reply to comment #9)
> Oops, that's because I have a busted Element.cpp change :( Didn't see that. Will revert it in a minute.

Reverted it in http://trac.webkit.org/changeset/104081.

[Ryosuke Niwa]

Fixed in http://trac.webkit.org/changeset/104068.