[Webkit-unassigned] [Bug 86296] Implement LayoutTest's eventSender.beginDragWithFiles interface in windows platform

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=86296





--- Comment #19 from huangxueqing <huangxueqing at baidu.com>  2012-05-27 23:00:36 PST ---
(In reply to comment #16)
> (From update of attachment 143991 [details])
> View in context: https://bugs.webkit.org/attachment.cgi?id=143991&action=review
> 
> Looking much better.  Can you file a bug about check-webkit-style not running on Chinese Windows?  Please include a stack trace.
> 
Done

> > Tools/DumpRenderTree/win/DRTDataObject.cpp:373
> > +            m_formats[m_formats.size() - 1] = 0;
> > +            m_formats.removeLast();
> 
> Do we need to set "m_formats[m_formats.size() - 1] = 0" if we're going to remove it?
> 
Yes, set m_formats[m_formats.size() - 1] = 0 void it was destructed since we had assign m_formats[m_formats.size() - 1] to m_formats[position].

> > Tools/DumpRenderTree/win/DRTDataObject.h:68
> > +    Vector<FORMATETC*> m_formats;
> > +    Vector<STGMEDIUM*> m_medium;
> 
> Can we use Vector<OwnPtr<FORMATETC> > and Vector<OwnPtr<STGMEDIUM> >?
> 
I don't think we should use OwenPtr<FORMATTEC> and OwnPtr<STGMEDIUM> than FORMATTEC* and STGMEDIUM*, the reasons as bellow:
1. OwnPtr did not provide a method like void deleteOwnedPtr(STGMEDIUM ptr), hence we need explicit call ReleaseStgMedium(medium) event though we use OwnPtr<STGMEDIUM>;
2. DRTDataObject was a COM object, i think we should avoid use smart pointer as a member of COM object.

> > Tools/DumpRenderTree/win/DRTDropSource.h:36
> > +    virtual ULONG STDMETHODCALLTYPE AddRef(void);
> > +    virtual ULONG STDMETHODCALLTYPE Release(void);
> 
> Can we remove void?
Done

> 
> > Tools/DumpRenderTree/win/EventSender.cpp:668
> > +    DROPFILES* dropFiles = reinterpret_cast<DROPFILES *>(GlobalLock(medium.hGlobal));
> 
> Nit: No space between DROPFILES and *.
Done

-- 
Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.