[webkit-reviews] review denied: [Bug 68206] Ensure TestWebKitAPI works on mac, win, chromium-mac and chromium-linux : [Attachment 107629] patch

[bugzilla-daemon at webkit.org]

Tony Chang <tony at chromium.org> has denied Xianzhu Wang
<wangxianzhu at chromium.org>'s request for review:
Bug 68206: Ensure TestWebKitAPI works on mac, win, chromium-mac and
chromium-linux
https://bugs.webkit.org/show_bug.cgi?id=68206

Attachment 107629: patch
https://bugs.webkit.org/attachment.cgi?id=107629&action=review

------- Additional Comments from Tony Chang <tony at chromium.org>
View in context: https://bugs.webkit.org/attachment.cgi?id=107629&action=review


> Is it intended that component=shared_library doesn't work for 'build-wekit
--chromium'?

We don't have a bot for this, so I'm not surprised it doesn't work.

I don't really understand exactly how TestWebKitAPI works on win/mac and I
don't have access to a win/mac right now.  I'll look again tomorrow.

> Source/WebKit/chromium/WebKit.gyp:1265
> +		   'public',

Is this include only needed for Chromium{CurrentTime,Threading}.cpp?  If so,
maybe it should be in the condition below?

> Source/WebKit/chromium/WebKit.gyp:1270
> +		   ''

Nit: Remove this blank line

> Source/WebKit/chromium/WebKit.gyp:1276
> +			   'src/ChromiumCurrentTime.cpp',
> +			   'src/ChromiumThreading.cpp',

In the shared_library build on windows, do we call currentTime() or
ChromiumThreading::callOnMainThread()?	Should we include separate stub
implementations when linking TestWebKitAPI rather than using the definitions in
webkit?  For example, we could add stubs in
WebKit/Tools/TestWebKitAPI/chromium/.

> Tools/TestWebKitAPI/Tests/WTF/MetaAllocator.cpp:32
> -#include <JavaScriptCore/MetaAllocator.h>
> -#include <JavaScriptCore/Vector.h>
> +#include <stdarg.h>
> +#include <wtf/MetaAllocator.h>
> +#include <wtf/Vector.h>

I don't understand this include/ForwardingHeader change.  Can you describe why
this is needed in the ChangeLog?  Why does it work in the Apple Mac build?