[Webkit-unassigned] [Bug 221294] conic-gradient with stops starting after 1turn is blank

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=221294

--- Comment #4 from Antti Koivisto <koivisto at iki.fi> ---
Comment on attachment 423887
  --> https://bugs.webkit.org/attachment.cgi?id=423887
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=423887&action=review

>> Source/WebCore/ChangeLog:9
>> +        Tests: web-platform-tests/css/css-images/normalization-conic-2.html
> 
> Please add a description here of the bug and what you changed to fix it.

The test is currently marked failing in LayoutTests/TestExpectations. The patch should remove that line so the test runs.

> Source/WebCore/css/CSSGradientValue.cpp:243
> +        while (firstZeroOrGreaterIndex < numStops && * stops[firstZeroOrGreaterIndex].offset < 0) {

WebKit coding style is '*stops' without a space. See https://webkit.org/code-style-guidelines/ for more.

> Source/WebCore/css/CSSGradientValue.cpp:245
> +            firstZeroOrGreaterIndex++;
>          }

"One-line control clauses should not use braces unless comments are included or a single statement spans multiple lines"

> Source/WebCore/css/CSSGradientValue.cpp:271
> +        while (lastOneOrLessIndex > 0 && *stops[lastOneOrLessIndex].offset > 1) {
> +            lastOneOrLessIndex--;
>          }

same here

> Source/WebCore/css/CSSGradientValue.cpp:290
> -                    stops[i].offset = 1;
> +                    *stops[i].offset = 1;

I suppose this saves a branch if we know stop Optionals are already initialized. Not sure if that is meaningful, at least it should be done consistently in other offset assignments in this function too.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.webkit.org/pipermail/webkit-unassigned/attachments/20210322/4d58d3bb/attachment.htm>