Bug 13212 – Submit initial version of WebKit Apollo bitmap interface for feedback

• Status: RESOLVED
• Resolution: WONTFIX
• Component: WebKit API

[Chris Brichford]

Creating a patch with this one file that defines the interfaces between WebKit Apollo and the bitmap surfaces it draws into.  The point of this patch is to get feed back on the coding style and the directory layout in WebKit/apollo.

[Chris Brichford]

Created attachment 13838 [details]
Initial patch of a few web kit apollo interface files.

Initial patch of a few web kit apollo interface files.  Please just review.  Do not commit.

[Chris Brichford]

There will be a source directory that contains all the implementation files for WebKitApollo at WebKit/apollo/source.  Headers private to WebKitApollo will live in WebKit/apollo/source.  Public headers will all be in WebKitApollo/apollo/include/WebKitApollo, WebKit/apollo/include/mac/WebKitApollo, WebKit/apollo/include/linux/WebKitApollo, or WebKit/apollo/include/win/WebKitApollo.

I also realized I forgot to put in a copyright notice in WebConfig.h.

[Mark Rowe (bdash)]

Comment on attachment 13838 [details]
Initial patch of a few web kit apollo interface files.

About the directory naming:  I'm not sure what having the extra level "WebKitApollo" adds to things.  On the Mac at least the <WebKitApollo/HeaderName.h> framework-style references would work for free due to how Xcode's header copying build phase works.  You may want to take a look at how we handle the Windows WebKit layer accessing WebCore headers using the <WebCore/HeaderName.h> style import and use a similar approach here.

Miscellaneous coding style points:
*) There is a lot of extraneous whitespace inside parentheses in your typedef's.
*) The *-placement inside your typedefs isn't consistent.
*) The name m_pVTable in WebKitApolloBitmap seems like it's bordering on Hungarian notation.

What is the purpose of the WEBKIT_APOLLO_PROTO1 and WEBKIT_APOLLO_PROTO2 macros?  Their names provide no information as to the reason for their existence, and their empty definitions are not particularly enlightening either.

[Robert Blaut]

Chris, do you plan to proceed with the bug?

[Chris Brichford]

(In reply to comment #4)
> Chris, do you plan to proceed with the bug?
> 

Not at this time.  You can kill the bug.