[Webkit-unassigned] [Bug 146997] [Cairo] SVG path not rendered with all-zero dasharray

bugzilla-daemon at webkit.org bugzilla-daemon at webkit.org
Tue Jul 21 20:16:59 PDT 2015


https://bugs.webkit.org/show_bug.cgi?id=146997

Martin Robinson <mrobinson at webkit.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
 Attachment #256903|review?                     |review-
              Flags|                            |

--- Comment #5 from Martin Robinson <mrobinson at webkit.org> ---
Comment on attachment 256903
  --> https://bugs.webkit.org/attachment.cgi?id=256903
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=256903&action=review

Nice. Fix. I have a few small nits.

> Source/WebCore/platform/graphics/cairo/GraphicsContextCairo.cpp:907
> +    // Avoid all-zero patterns that would trigger the CAIRO_STATUS_INVALID_DASH context error state.
> +    bool allZero = true;
> +    for (auto& dash : dashes) {
> +        if (dash) {
> +            allZero = false;
> +            break;
> +        }

I think you could pull this bit out into a helper function that returns a boolean. Something like isDashArrayAllZero.

> Source/WebCore/platform/graphics/cairo/GraphicsContextCairo.cpp:911
> +    if (allZero)
> +        return;
> +

What should the behavior of an all-zero dash array be? Should it disable dashes? If so, perhaps in that case you should call cairo_set_dash with num_dashes=0. Either way, I think you need to adjust the state appropriately.

> LayoutTests/fast/canvas/script-tests/canvas-lineDash.js:107
> +// Verify all-zero dash rendering

Nit: missing period.

> LayoutTests/fast/canvas/script-tests/canvas-lineDash.js:110
> +ctx.lineWidth = 4; // To make the test immune to plaform anti-aliasing discrepancies

Ditto.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.webkit.org/pipermail/webkit-unassigned/attachments/20150722/a901b704/attachment-0001.html>


More information about the webkit-unassigned mailing list