Bug 113864

Summary: [Qt] Add getter for JSC TestRunner to DumpRenderTreeQt.
Product: WebKit Reporter: Zeno Albisser <zeno>
Component: New BugsAssignee: Zeno Albisser <zeno>
Status: RESOLVED FIXED    
Severity: Normal CC: hausmann
Priority: P2    
Version: 528+ (Nightly build)   
Hardware: Unspecified   
OS: Unspecified   
Bug Depends on:    
Bug Blocks: 113865, 113866    
Attachments:
Description Flags
Patch hausmann: review+, hausmann: commit-queue-

Description Zeno Albisser 2013-04-03 03:42:18 PDT
[Qt] Add getter for JSC TestRunner to DumpRenderTreeQt.
Comment 1 Zeno Albisser 2013-04-03 03:43:40 PDT
Created attachment 196318 [details]
Patch
Comment 2 Simon Hausmann 2013-04-03 04:20:50 PDT
Comment on attachment 196318 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=196318&action=review

r=me, but the getter needs to be fixed before landing IMHO

> Tools/ChangeLog:7
> +        As long as we instantiate both TestRunner and TestRunnerQt,
> +        we need a separate getter for the non-Qt TestRunner instance.

ChangeLog shoudl also mention that this moves setDefersLoading at the same time. I think given the size of this patch it's okay if both go in at the same time. Just maybe fix the ChangeLog text before landing :)

> Tools/DumpRenderTree/qt/DumpRenderTreeQt.h:92
> +    PassRefPtr<TestRunner> jscTestRunner() const { return m_jscController; }

I think the common pattern is to use PassRefPtr as return value for create functions, but use raw pointers for getters, i.e. TestRunner* jscTestRunner() { return m_jscController.get(); }

If the caller wants to keep a ref, then the RefPtr ctor that takes a raw pointer will do the ref. Otherwise I think PassRefPtr doesn't give us anything here.
Comment 3 Zeno Albisser 2013-04-03 09:26:27 PDT
Committed r147563: <http://trac.webkit.org/changeset/147563>