Bug 240416 – Template TextBoxPainter on line layout path

• Status: RESOLVED
• Resolution: FIXED
• Component: Layout and Rendering

[Antti Koivisto]

Painting currently goes via line layout path agnostic iterator. This is inefficient since we always know what sort of layout we are painting.

[Antti Koivisto]

Created attachment 459348 [details]
Patch

[Antti Koivisto]

Created attachment 459353 [details]
patch

[EWS]

Committed r294204 (250565@main): <https://commits.webkit.org/250565@main>

All reviewed patches have been landed. Closing bug and clearing flags on attachment 459353 [details].

[Radar WebKit Bug Importer]

<rdar://problem/93293276>

[Michael Catanzaro]

Re-opening for pull request https://github.com/WebKit/WebKit/pull/650

[EWS]

Committed r294262 (250610@main): <https://commits.webkit.org/250610@main>

Reviewed commits have been landed. Closing PR #650 and removing active labels.

[Simon Fraser (smfr)]

Caused text rendering bugs: bug 240491.

[Antti Koivisto]

Created attachment 459498 [details]
Patch

[EWS]

Unable to find any modified ChangeLog in Attachment 459498 [details]

[Antti Koivisto]

Pull request: https://github.com/WebKit/WebKit/pull/675

[EWS]

Committed r294383 (250680@main): <https://commits.webkit.org/250680@main>

All reviewed patches have been landed. Closing bug and clearing flags on attachment 459498 [details].

[Michael Catanzaro]

(In reply to Antti Koivisto from comment #10)
> Pull request: https://github.com/WebKit/WebKit/pull/675

You landed the same build failure again.

Out of curiosity, is there any plan to make this non-experimental? I wonder if having a build flag is really worth it?

[Michael Catanzaro]

(Remember that ENABLE(EXPERIMENTAL_FEATURES) is enabled on EWS, and ENABLE(LAYOUT_FORMATTING_CONTEXT) is tied to that. So all the EWS will pass, but the releases will still be broken. Got to be careful....)

[zalan]

(In reply to Michael Catanzaro from comment #12)
> (In reply to Antti Koivisto from comment #10)
> > Pull request: https://github.com/WebKit/WebKit/pull/675
> 
> You landed the same build failure again.
> 
> Out of curiosity, is there any plan to make this non-experimental? I wonder
> if having a build flag is really worth it?
Yeah, the build flag makes very little sense at this point. We should get rid of it.

[Antti Koivisto]

(In reply to Michael Catanzaro from comment #13)
> (Remember that ENABLE(EXPERIMENTAL_FEATURES) is enabled on EWS, and
> ENABLE(LAYOUT_FORMATTING_CONTEXT) is tied to that. So all the EWS will pass,
> but the releases will still be broken. Got to be careful....)

All platforms should have LAYOUT_FORMATTING_CONTEXT enabled unconditionally. It is on Cocoa platforms.

[Antti Koivisto]

(and the flag should be removed)

[Michael Catanzaro]

(In reply to Antti Koivisto from comment #15)
> All platforms should have LAYOUT_FORMATTING_CONTEXT enabled unconditionally.
> It is on Cocoa platforms.

I see it's off at runtime on all platforms (WTF/Scripts/Preferences/WebPreferencesInternal.yaml). Removing the build flag makes sense to me regardless.

[Michael Catanzaro]

Re-opening for pull request https://github.com/WebKit/WebKit/pull/741

[EWS]

Committed r294426 (250708@main): <https://commits.webkit.org/250708@main>

Reviewed commits have been landed. Closing PR #741 and removing active labels.

[Antti Koivisto]

> I see it's off at runtime on all platforms
> (WTF/Scripts/Preferences/WebPreferencesInternal.yaml).

InlineFormattingContextIntegrationEnabled is behind this conditional and is enabled by default.

[Basuke Suzuki]

*** Bug 240590 has been marked as a duplicate of this bug. ***