Title: [124209] trunk
- Revision
- 124209
- Author
- r...@google.com
- Date
- 2012-07-31 08:55:48 -0700 (Tue, 31 Jul 2012)
Log Message
skia: switch to conical gradient to match css/svg spec for two-point gradients
https://bugs.webkit.org/show_bug.cgi?id=92754
Reviewed by Stephen White.
Source/WebCore:
Switching to skia's TwoPointConical gradient, which has been written explicitly to match the css/svg spec.
Leaving the existing TwoPointRadial gradient in skia, for other callers that may want the old behavior.
No new tests. existing layouttests exercise this change (9 pending rebaselines associated with this CL).
* platform/graphics/skia/GradientSkia.cpp:
(WebCore::Gradient::platformGradient):
LayoutTests:
Update expectations to rebaseline 9 images after this change lands
* platform/chromium/TestExpectations:
Modified Paths
Diff
Modified: trunk/LayoutTests/ChangeLog (124208 => 124209)
--- trunk/LayoutTests/ChangeLog 2012-07-31 15:53:15 UTC (rev 124208)
+++ trunk/LayoutTests/ChangeLog 2012-07-31 15:55:48 UTC (rev 124209)
@@ -1,3 +1,14 @@
+2012-07-31 Mike Reed <r...@google.com>
+
+ skia: switch to conical gradient to match css/svg spec for two-point gradients
+ https://bugs.webkit.org/show_bug.cgi?id=92754
+
+ Reviewed by Stephen White.
+
+ Update expectations to rebaseline 9 images after this change lands
+
+ * platform/chromium/TestExpectations:
+
2012-07-31 Alexander Pavlov <apav...@chromium.org>
[Chromium] Unreviewed, rebaseline fast/ruby/overhang-horizontal.html.
Modified: trunk/LayoutTests/platform/chromium/TestExpectations (124208 => 124209)
--- trunk/LayoutTests/platform/chromium/TestExpectations 2012-07-31 15:53:15 UTC (rev 124208)
+++ trunk/LayoutTests/platform/chromium/TestExpectations 2012-07-31 15:55:48 UTC (rev 124209)
@@ -1255,7 +1255,7 @@
BUGWK83303 WIN LINUX : svg/text/foreignObject-text-clipping-bug.xml = PASS IMAGE
BUGWK83303 LINUX : svg/transforms/text-with-mask-with-svg-transform.svg = PASS IMAGE
BUGWK83303 WIN : svg/as-background-image/svg-as-background-1.html = PASS IMAGE
-BUGWK83303 WIN : svg/as-background-image/svg-as-background-3.html = PASS IMAGE
+//BUGWK83303 WIN : svg/as-background-image/svg-as-background-3.html = PASS IMAGE
BUGWK83303 WIN : svg/as-background-image/svg-as-background-6.html = PASS IMAGE
BUGWK83303 WIN : svg/custom/mouse-move-on-svg-container-standalone.svg = IMAGE PASS
BUGWK83303 WIN : svg/custom/mouse-move-on-svg-container.xhtml = IMAGE PASS
@@ -1867,6 +1867,19 @@
BUGWK45991 LINUX DEBUG : canvas/philip/tests/2d.gradient.radial.cone.top.html = TIMEOUT TEXT
BUGWK45991 LINUX DEBUG : platform/chromium/virtual/gpu/canvas/philip/tests/2d.gradient.radial.cone.top.html = TIMEOUT TEXT
+// Rebaseline these now that we're using Skia's twopointconical gradient, which
+// correctly matches the spec.
+BUGCR139796 : fast/gradients/crash-on-zero-radius.html = IMAGE
+BUGCR139796 : fast/gradients/generated-gradients.html = IMAGE
+BUGCR139796 : fast/gradients/list-item-gradient.html = IMAGE
+BUGCR139796 : fast/gradients/simple-gradients.html = IMAGE
+BUGCR139796 : svg/W3C-SVG-1.1/pservers-grad-13-b.svg = IMAGE
+BUGCR139796 : svg/as-background-image/svg-as-background-3.html = IMAGE
+BUGCR139796 : svg/as-background-image/svg-as-background-with-relative-size.html = IMAGE
+BUGCR139796 : svg/custom/radial-gradient-with-outstanding-focalPoint.svg = IMAGE
+BUGCR139796 : svg/zoom/page/zoom-svg-as-background-with-relative-size.html = IMAGE
+
+
BUGWK39212 : canvas/philip/tests/2d.imageData.create1.type.html = TEXT
BUGWK39212 : canvas/philip/tests/2d.imageData.create2.type.html = TEXT
BUGWK39212 : canvas/philip/tests/2d.imageData.get.type.html = TEXT
Modified: trunk/Source/WebCore/ChangeLog (124208 => 124209)
--- trunk/Source/WebCore/ChangeLog 2012-07-31 15:53:15 UTC (rev 124208)
+++ trunk/Source/WebCore/ChangeLog 2012-07-31 15:55:48 UTC (rev 124209)
@@ -1,3 +1,18 @@
+2012-07-31 Mike Reed <r...@google.com>
+
+ skia: switch to conical gradient to match css/svg spec for two-point gradients
+ https://bugs.webkit.org/show_bug.cgi?id=92754
+
+ Reviewed by Stephen White.
+
+ Switching to skia's TwoPointConical gradient, which has been written explicitly to match the css/svg spec.
+ Leaving the existing TwoPointRadial gradient in skia, for other callers that may want the old behavior.
+
+ No new tests. existing layouttests exercise this change (9 pending rebaselines associated with this CL).
+
+ * platform/graphics/skia/GradientSkia.cpp:
+ (WebCore::Gradient::platformGradient):
+
2012-04-06 Ilya Tikhonovsky <loi...@chromium.org>
Web Inspector: extract MemoryInstrumentationImpl into separate files for future reusing in unit tests
Modified: trunk/Source/WebCore/platform/graphics/skia/GradientSkia.cpp (124208 => 124209)
--- trunk/Source/WebCore/platform/graphics/skia/GradientSkia.cpp 2012-07-31 15:53:15 UTC (rev 124208)
+++ trunk/Source/WebCore/platform/graphics/skia/GradientSkia.cpp 2012-07-31 15:55:48 UTC (rev 124209)
@@ -152,7 +152,7 @@
// negative radius, ask for zero instead.
SkScalar radius0 = m_r0 >= 0.0f ? WebCoreFloatToSkScalar(m_r0) : 0;
SkScalar radius1 = m_r1 >= 0.0f ? WebCoreFloatToSkScalar(m_r1) : 0;
- m_gradient = SkGradientShader::CreateTwoPointRadial(m_p0, radius0, m_p1, radius1, colors, pos, static_cast<int>(countUsed), tile);
+ m_gradient = SkGradientShader::CreateTwoPointConical(m_p0, radius0, m_p1, radius1, colors, pos, static_cast<int>(countUsed), tile);
}
if (aspectRatio() != 1) {
_______________________________________________
webkit-changes mailing list
webkit-changes@lists.webkit.org
http://lists.webkit.org/mailman/listinfo/webkit-changes
