mojo::Serialize*_() calls now propogate/return validation errors.

mojo/public/cpp/bindings/tests/serialization_warning_unittest.cc

Index: mojo/public/cpp/bindings/tests/serialization_warning_unittest.cc
diff --git a/mojo/public/cpp/bindings/tests/serialization_warning_unittest.cc b/mojo/public/cpp/bindings/tests/serialization_warning_unittest.cc
index b61d3bc57e41482d50bbcbef8eb50b4683d3f132..3d89d565b5ae0d28e832f5cf651edbbebb20d4a0 100644
--- a/mojo/public/cpp/bindings/tests/serialization_warning_unittest.cc
+++ b/mojo/public/cpp/bindings/tests/serialization_warning_unittest.cc
@@ -61,7 +61,7 @@ class SerializationWarningTest : public testing::Test {
 
     mojo::internal::FixedBufferForTesting buf(GetSerializedSize_(*obj));
     typename T::Data_* data;
-    Serialize_(obj.get(), &buf, &data);
+    EXPECT_EQ(expected_warning, Serialize_(obj.get(), &buf, &data));
 
     EXPECT_EQ(expected_warning, warning_observer_.last_warning());

[viettrungluu, 2015/10/09 01:35:20]

I wonder if we should get rid of this whole warning observer nonsense. (But we
can worry about that separately.)

[vardhan, 2015/10/09 22:31:29]

I was thinking about that..

It's normally a FATAL log on debug builds if the reporting is "turned off" (i.e.
outside of unittests), so I'll make it a WARNING now.  I'll just do it here
since it's very relevant? :-)

   }
@@ -74,7 +74,8 @@ class SerializationWarningTest : public testing::Test {
 
     mojo::internal::FixedBufferForTesting buf(GetSerializedSize_(obj));
     typename T::Data_* data;
-    SerializeArray_(&obj, &buf, &data, validate_params);
+    EXPECT_EQ(expected_warning,
+              SerializeArray_(&obj, &buf, &data, validate_params));
 
     EXPECT_EQ(expected_warning, warning_observer_.last_warning());
   }