Chromium Code Reviews| Index: Source/web/WebPluginContainerImpl.cpp |
| diff --git a/Source/web/WebPluginContainerImpl.cpp b/Source/web/WebPluginContainerImpl.cpp |
| index 55e2802ecbbcb085f8b1adc129d7654ae5a40937..f310e0ec752bc4915331a0e3940a1cea5abd068d 100644 |
| --- a/Source/web/WebPluginContainerImpl.cpp |
| +++ b/Source/web/WebPluginContainerImpl.cpp |
| @@ -463,7 +463,7 @@ v8::Local<v8::Object> WebPluginContainerImpl::v8ObjectForElement() |
| return v8::Local<v8::Object>(); |
| v8::Handle<v8::Value> v8value = toV8(m_element.get(), scriptState->context()->Global(), scriptState->isolate()); |
| - ASSERT(v8value->IsObject()); |
| + ASSERT(!v8value.IsEmpty() && v8value->IsObject()); |
|
haraken
2015/04/28 04:24:59
Who guarantees it is not empty?
Also we can remov
bashi
2015/04/28 05:41:53
The existing ASSERT() implies this should not be e
haraken
2015/04/28 05:50:34
I think that is exactly what we're fixing with int
bashi
2015/04/28 06:41:14
Yeah, let's use !IsEmpty() assertion to see this a
Yuki
2015/04/28 06:43:46
The problem is, ASSERT will be noop in release bui
haraken
2015/04/28 06:46:08
If you argue that we can go with ASSERT here, that
bashi
2015/04/28 06:55:49
Ok, changed to return empty handle.
That's said,
|
| return v8::Handle<v8::Object>::Cast(v8value); |
| } |