[Webkit-unassigned] [Bug 202739] [GTK] Implement conic gradient

bugzilla-daemon at webkit.org bugzilla-daemon at webkit.org
Mon Oct 14 06:28:46 PDT 2019


https://bugs.webkit.org/show_bug.cgi?id=202739

--- Comment #8 from Carlos Alberto Lopez Perez <clopez at igalia.com> ---
Comment on attachment 380822
  --> https://bugs.webkit.org/attachment.cgi?id=380822
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=380822&action=review

> Source/WebCore/css/CSSGradientValue.cpp:1364
> +    float startRadius = sqrt(powf(size.width() - centerPoint.x(), 2) + powf(size.width() - centerPoint.y(), 2));

Not sure if you missed my previous comment, but you can use std::hypot here (see bug 198483)

> Source/WebCore/platform/graphics/cairo/GradientCairo.cpp:137
> +static void
> +addConicSector(cairo_pattern_t *gradient, float cx, float cy, float r, float angleRadians,
> +    Gradient::ColorStop from, Gradient::ColorStop to, float globalAlpha)

Please fix the style here (return value in same line than function declaration)

> Source/cmake/OptionsGTK.cmake:165
> +WEBKIT_OPTION_DEFAULT_PORT_VALUE(ENABLE_CSS_CONIC_GRADIENTS PRIVATE ON)

The implementation looks generic enough to work also on the WPE port. Is that the case? Then why not enable it there as well?

> LayoutTests/ChangeLog:6
> +        [GTK] Implement conic gradient
> +        https://bugs.webkit.org/show_bug.cgi?id=202739
> +
> +        * platform/gtk/fast/gradients/conic-gradient-alpha-expected.html: Added.

Other than adding the new expectation you have to unskip this test for platform GTK. For that add the line below to the file LayoutTests/platform/gtk/TestExpectations

fast/gradients/conic-gradient-alpha.html [ Pass ]

-- 
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/20191014/a33186ad/attachment.html>


More information about the webkit-unassigned mailing list