WebKit Bugzilla
New
Browse
Log In
×
Sign in with GitHub
or
Remember my login
Create Account
·
Forgot Password
Forgotten password account recovery
RESOLVED FIXED
Bug 84251
[shadow] needs style recalculation by changing applyAuthorStyles dynamically
https://bugs.webkit.org/show_bug.cgi?id=84251
Summary
[shadow] needs style recalculation by changing applyAuthorStyles dynamically
Takashi Sakamoto
Reported
2012-04-18 10:05:12 PDT
Currently applyAuthorStyles API doesn't cause style recalculation. Only when attaching elements (i.e. when creating renderer), applyAuthorStyles works. It should cause style recalculation.
Attachments
Patch
(6.01 KB, patch)
2012-05-13 23:34 PDT
,
Takashi Sakamoto
no flags
Details
Formatted Diff
Diff
Patch
(5.99 KB, patch)
2012-05-15 23:27 PDT
,
Takashi Sakamoto
no flags
Details
Formatted Diff
Diff
Patch
(7.08 KB, patch)
2012-05-18 04:13 PDT
,
Takashi Sakamoto
no flags
Details
Formatted Diff
Diff
Show Obsolete
(2)
View All
Add attachment
proposed patch, testcase, etc.
Takashi Sakamoto
Comment 1
2012-05-13 23:34:29 PDT
Created
attachment 141647
[details]
Patch
Hajime Morrita
Comment 2
2012-05-15 22:22:28 PDT
Comment on
attachment 141647
[details]
Patch View in context:
https://bugs.webkit.org/attachment.cgi?id=141647&action=review
In addition to reftests, It would be desired to check style value using widow.getComputedStyle() before and after the second change. That allows you to ensure that the value is actually changed dynamically.
> Source/WebCore/dom/ShadowRoot.cpp:201 > + owner()->needsStyleRecalc();
needsStyleRecalc() doesn't have any side effect. It looks what you want is setNeedsReattachHostChildrenAndShadow().
> LayoutTests/fast/dom/shadow/shadow-root-applyAuthorStyles.html:94 > + shadowRoot.applyAuthorStyles = true;
You need to force layout here (using offsetLeft or getComputedStyle like thing)
> LayoutTests/fast/dom/shadow/shadow-root-applyAuthorStyles.html:108 > + shadowRoot.applyAuthorStyles = false;
Ditto.
Takashi Sakamoto
Comment 3
2012-05-15 23:27:57 PDT
Created
attachment 142162
[details]
Patch
Takashi Sakamoto
Comment 4
2012-05-18 01:35:09 PDT
(In reply to
comment #2
)
> (From update of
attachment 141647
[details]
) > View in context:
https://bugs.webkit.org/attachment.cgi?id=141647&action=review
> > In addition to reftests, It would be desired to check style value using widow.getComputedStyle() > before and after the second change. That allows you to ensure that the value is actually changed dynamically. > > > Source/WebCore/dom/ShadowRoot.cpp:201 > > + owner()->needsStyleRecalc(); > > needsStyleRecalc() doesn't have any side effect. > It looks what you want is setNeedsReattachHostChildrenAndShadow().
Done.
> > LayoutTests/fast/dom/shadow/shadow-root-applyAuthorStyles.html:94 > > + shadowRoot.applyAuthorStyles = true; > > You need to force layout here (using offsetLeft or getComputedStyle like thing) > > > LayoutTests/fast/dom/shadow/shadow-root-applyAuthorStyles.html:108 > > + shadowRoot.applyAuthorStyles = false; > > Ditto.
I see. I added "div.offsetLeft".
Takashi Sakamoto
Comment 5
2012-05-18 04:13:06 PDT
Created
attachment 142682
[details]
Patch
Takashi Sakamoto
Comment 6
2012-05-18 05:36:05 PDT
(In reply to
comment #2
)
> (From update of
attachment 141647
[details]
) > View in context:
https://bugs.webkit.org/attachment.cgi?id=141647&action=review
> > In addition to reftests, It would be desired to check style value using widow.getComputedStyle() > before and after the second change. That allows you to ensure that the value is actually changed dynamically.
Done.
Hajime Morrita
Comment 7
2012-05-20 17:39:32 PDT
Comment on
attachment 142682
[details]
Patch Looks good.Thanks for the fix!
WebKit Review Bot
Comment 8
2012-05-20 17:49:17 PDT
Comment on
attachment 142682
[details]
Patch Clearing flags on attachment: 142682 Committed
r117716
: <
http://trac.webkit.org/changeset/117716
>
WebKit Review Bot
Comment 9
2012-05-20 17:49:22 PDT
All reviewed patches have been landed. Closing bug.
Note
You need to
log in
before you can comment on or make changes to this bug.
Top of Page
Format For Printing
XML
Clone This Bug