Title: [138176] trunk/Tools
Revision
138176
Author
jchaffr...@webkit.org
Date
2012-12-19 09:45:46 -0800 (Wed, 19 Dec 2012)

Log Message

2 clampTo unit tests are failing
https://bugs.webkit.org/show_bug.cgi?id=105253

Reviewed by Emil A Eklund.

Removed clampToUnsigned as it was actually testing exactly what
clampToUnsignedLong was doing. Also made clampToIntFloat more robust
to precision loss (fix suggested by Emil).

* TestWebKitAPI/Tests/WTF/MathExtras.cpp:

Modified Paths

Diff

Modified: trunk/Tools/ChangeLog (138175 => 138176)


--- trunk/Tools/ChangeLog	2012-12-19 17:42:35 UTC (rev 138175)
+++ trunk/Tools/ChangeLog	2012-12-19 17:45:46 UTC (rev 138176)
@@ -1,3 +1,16 @@
+2012-12-19  Julien Chaffraix  <jchaffr...@webkit.org>
+
+        2 clampTo unit tests are failing
+        https://bugs.webkit.org/show_bug.cgi?id=105253
+
+        Reviewed by Emil A Eklund.
+
+        Removed clampToUnsigned as it was actually testing exactly what
+        clampToUnsignedLong was doing. Also made clampToIntFloat more robust
+        to precision loss (fix suggested by Emil).
+
+        * TestWebKitAPI/Tests/WTF/MathExtras.cpp:
+
 2012-12-19  Thiago Marcos P. Santos  <thiago.san...@intel.com>
 
         [EFL] Move ImageDiff to the Tools directory root

Modified: trunk/Tools/TestWebKitAPI/Tests/WTF/MathExtras.cpp (138175 => 138176)


--- trunk/Tools/TestWebKitAPI/Tests/WTF/MathExtras.cpp	2012-12-19 17:42:35 UTC (rev 138175)
+++ trunk/Tools/TestWebKitAPI/Tests/WTF/MathExtras.cpp	2012-12-19 17:45:46 UTC (rev 138176)
@@ -87,8 +87,7 @@
     EXPECT_EQ(clampTo<int>(underflowInt), minInt);
 }
 
-// https://bugs.webkit.org/show_bug.cgi?id=105253
-TEST(WTF, DISABLED_clampToIntegerFloat)
+TEST(WTF, clampToIntegerFloat)
 {
     // This test is inaccurate as floats will round the min / max integer
     // due to the narrow mantissa. However it will properly checks within
@@ -101,10 +100,11 @@
     EXPECT_GT(overflowInt, maxInt);
     EXPECT_LT(underflowInt, minInt);
 
-    EXPECT_EQ(clampToInteger(maxInt), maxInt);
+    // If maxInt == 2^31 - 1 (ie on I32 architecture), the closest float used to represent it is 2^31.
+    EXPECT_NEAR(clampToInteger(maxInt), maxInt, 1);
     EXPECT_EQ(clampToInteger(minInt), minInt);
 
-    EXPECT_EQ(clampToInteger(overflowInt), maxInt);
+    EXPECT_NEAR(clampToInteger(overflowInt), maxInt, 1);
     EXPECT_EQ(clampToInteger(underflowInt), minInt);
 }
 
@@ -145,17 +145,6 @@
     EXPECT_EQ(clampToFloat(-std::numeric_limits<float>::infinity()), minFloat);
 }
 
-// https://bugs.webkit.org/show_bug.cgi?id=105253
-TEST(WTF, DISABLED_clampToUnsigned)
-{
-    unsigned long maxUnsigned = std::numeric_limits<unsigned>::max();
-    unsigned long overflowUnsigned = maxUnsigned + 1;
-    EXPECT_EQ(clampTo<unsigned>(maxUnsigned), maxUnsigned);
-
-    EXPECT_EQ(clampTo<unsigned>(overflowUnsigned), maxUnsigned);
-    EXPECT_EQ(clampTo<unsigned>(-1), 0u);
-}
-
 TEST(WTF, clampToUnsignedLong)
 {
     if (sizeof(unsigned long) == sizeof(unsigned))
_______________________________________________
webkit-changes mailing list
webkit-changes@lists.webkit.org
http://lists.webkit.org/mailman/listinfo/webkit-changes

Reply via email to