Title: [136509] trunk
Revision
136509
Author
[email protected]
Date
2012-12-04 07:41:44 -0800 (Tue, 04 Dec 2012)

Log Message

IntSize::scale() should return rounded value
https://bugs.webkit.org/show_bug.cgi?id=103984

Patch by Mikhail Pozdnyakov <[email protected]> on 2012-12-04
Reviewed by Kenneth Rohde Christiansen.

Source/WebCore:

IntSize::scale() is now rounding float value to the nearest integer instead of blind type casting.

Test: css3/device-adapt/viewport-width-not-affecting-next-page.html

* platform/graphics/IntSize.h:
(WebCore::IntSize::scale):

LayoutTests:

Test css3/device-adapt/viewport-width-not-affecting-next-page.html is not flaky any more.

* platform/efl-wk2/TestExpectations:

Modified Paths

Diff

Modified: trunk/LayoutTests/ChangeLog (136508 => 136509)


--- trunk/LayoutTests/ChangeLog	2012-12-04 15:19:32 UTC (rev 136508)
+++ trunk/LayoutTests/ChangeLog	2012-12-04 15:41:44 UTC (rev 136509)
@@ -1,3 +1,14 @@
+2012-12-04  Mikhail Pozdnyakov  <[email protected]>
+
+        IntSize::scale() should return rounded value
+        https://bugs.webkit.org/show_bug.cgi?id=103984
+
+        Reviewed by Kenneth Rohde Christiansen.
+
+        Test css3/device-adapt/viewport-width-not-affecting-next-page.html is not flaky any more.
+
+        * platform/efl-wk2/TestExpectations:
+
 2012-12-04  Stephen Chenney  <[email protected]>
 
         [chromium] Turn on the new Skia mask blur algorithm.

Modified: trunk/LayoutTests/platform/efl-wk2/TestExpectations (136508 => 136509)


--- trunk/LayoutTests/platform/efl-wk2/TestExpectations	2012-12-04 15:19:32 UTC (rev 136508)
+++ trunk/LayoutTests/platform/efl-wk2/TestExpectations	2012-12-04 15:41:44 UTC (rev 136509)
@@ -197,9 +197,6 @@
 webkit.org/b/103400 [ Debug ] compositing/overflow/content-gains-scrollbars.html [ Crash Pass ]
 webkit.org/b/103400 [ Debug ] compositing/overflow/content-loses-scrollbars.html [ Crash Pass ]
 
-# Easily reproducable flakiness on EFL WK2 only
-webkit.org/b/103882 css3/device-adapt/viewport-width-not-affecting-next-page.html [ Failure Pass ]
-
 webkit.org/b/103883 css3/selectors3/html/css3-modsel-1.html [ Failure Pass ]
 webkit.org/b/103883 css3/selectors3/xml/css3-modsel-1.xml [ Failure Pass ]
 webkit.org/b/103883 css3/selectors3/xhtml/css3-modsel-1.xml [ Failure Pass ]

Modified: trunk/Source/WebCore/ChangeLog (136508 => 136509)


--- trunk/Source/WebCore/ChangeLog	2012-12-04 15:19:32 UTC (rev 136508)
+++ trunk/Source/WebCore/ChangeLog	2012-12-04 15:41:44 UTC (rev 136509)
@@ -1,3 +1,17 @@
+2012-12-04  Mikhail Pozdnyakov  <[email protected]>
+
+        IntSize::scale() should return rounded value
+        https://bugs.webkit.org/show_bug.cgi?id=103984
+
+        Reviewed by Kenneth Rohde Christiansen.
+
+        IntSize::scale() is now rounding float value to the nearest integer instead of blind type casting.
+
+        Test: css3/device-adapt/viewport-width-not-affecting-next-page.html
+
+        * platform/graphics/IntSize.h:
+        (WebCore::IntSize::scale):
+
 2012-12-04  Adam Bergkvist  <[email protected]>
 
         Add support for generic types in arrays and sequences to the code generators

Modified: trunk/Source/WebCore/platform/graphics/IntSize.h (136508 => 136509)


--- trunk/Source/WebCore/platform/graphics/IntSize.h	2012-12-04 15:19:32 UTC (rev 136508)
+++ trunk/Source/WebCore/platform/graphics/IntSize.h	2012-12-04 15:41:44 UTC (rev 136509)
@@ -26,6 +26,8 @@
 #ifndef IntSize_h
 #define IntSize_h
 
+#include <wtf/MathExtras.h>
+
 #if USE(CG) || USE(SKIA_ON_MAC_CHROMIUM)
 typedef struct CGSize CGSize;
 #endif
@@ -83,8 +85,8 @@
 
     void scale(float widthScale, float heightScale)
     {
-        m_width = static_cast<int>(static_cast<float>(m_width) * widthScale);
-        m_height = static_cast<int>(static_cast<float>(m_height) * heightScale);
+        m_width = lroundf(static_cast<float>(m_width) * widthScale);
+        m_height = lroundf(static_cast<float>(m_height) * heightScale);
     }
     
     void scale(float scale)
_______________________________________________
webkit-changes mailing list
[email protected]
http://lists.webkit.org/mailman/listinfo/webkit-changes

Reply via email to