[Chromium] Move fileSystem to Platform.h

[Mark Pilgrim (Google)]

[Chromium] Move fileSystem to Platform.h

[Mark Pilgrim (Google)]

Created attachment 140451 [details]
Patch

[WebKit Review Bot]

Please wait for approval from [e-mail], [e-mail], [e-mail], [e-mail] or [e-mail] before submitting, as this patch contains changes to the Chromium public API. See also https://trac.webkit.org/wiki/ChromiumWebKitAPI.

[Adam Barth]

Comment on attachment 140451 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=140451&action=review

> Source/Platform/chromium/public/Platform.h:43
> +class WebFileSystem;

We should move WebFileSystem and its dependencies into Source/Platform/chromium/public as well.  I've already moved a bunch of them into Source/WebKit/chromium/public/platform, which is something of a staging area.  I think it would be ok to just move the WebFileSystem dependencies that are already in Source/WebKit/chromium/public/platform

[Mark Pilgrim (Google)]

Created attachment 140528 [details]
WIP patch (compile fails in WebFrame.h)

[Mark Pilgrim (Google)]

Work in progress, fails in WebFrame.h (can not find file WebFileSystem.h). Help?

[WebKit Review Bot]

Comment on attachment 140528 [details]
WIP patch (compile fails in WebFrame.h)

Attachment 140528 [details] did not pass chromium-ews (chromium-xvfb):
Output: http://queues.webkit.org/results/12655003

[Adam Barth]

Comment on attachment 140528 [details]
WIP patch (compile fails in WebFrame.h)

View in context: https://bugs.webkit.org/attachment.cgi?id=140528&action=review

> Source/WebKit/chromium/public/WebFrame.h:40
>  #include "platform/WebCanvas.h"
> -#include "platform/WebFileSystem.h"
>  #include "platform/WebReferrerPolicy.h"
>  #include "platform/WebURL.h"
> +#include <public/WebFileSystem.h>

The trick is to leave these includes the way they are and to add a "forwarding" header like <http://trac.webkit.org/browser/trunk/Source/WebKit/chromium/public/platform/WebURL.h>.  Once we've got everything in the right place, we'll remove the forwarding headers, but for now they help prevent these sorts of compile failures.

[Mark Pilgrim (Google)]

Created attachment 140558 [details]
Patch with forwarding header

[Adam Barth]

Comment on attachment 140558 [details]
Patch with forwarding header

View in context: https://bugs.webkit.org/attachment.cgi?id=140558&action=review

> Source/WebKit/chromium/public/WebCommonWorkerClient.h:35
> -#include "platform/WebFileSystem.h"
> +#include <public/WebFileSystem.h>

I'd leave the includes in the API using the forwarding header in case not all consumers of the API have their include paths set up to use the <public/...> version.

> Source/WebKit/chromium/src/AssertMatchingEnums.cpp:107
> +#include <public/WebFileSystem.h>

Changing the includes in the cpp files (and the non-public h files) is good though.

[Mark Pilgrim (Google)]

Created attachment 140563 [details]
Patch

[WebKit Review Bot]

Comment on attachment 140563 [details]
Patch

Clearing flags on attachment: 140563

Committed r116336: <http://trac.webkit.org/changeset/116336>

[WebKit Review Bot]

All reviewed patches have been landed.  Closing bug.