Bug 68936

Summary: [Chromium] Seperate GTK specific Gyp rules from X11 Gyp rules
Product: WebKit Reporter: Fady Samuel <fsamuel>
Component: New BugsAssignee: Fady Samuel <fsamuel>
Status: RESOLVED FIXED    
Severity: Normal CC: tony, webkit.review.bot
Priority: P2    
Version: 528+ (Nightly build)   
Hardware: Unspecified   
OS: Unspecified   
Attachments:
Description Flags
Patch
none
Patch
none
Patch none

Description Fady Samuel 2011-09-27 14:29:47 PDT
[Chromium] Seperate GTK specific Gyp rules from X11 Gyp rules
Comment 1 Fady Samuel 2011-09-27 14:33:25 PDT
Created attachment 108899 [details]
Patch
Comment 2 Tony Chang 2011-09-27 14:56:29 PDT
Comment on attachment 108899 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=108899&action=review

> Source/WebCore/ChangeLog:14
> +        No new tests. (OOPS!)

Remove this line.

> Source/WebCore/WebCore.gyp/WebCore.gyp:-1373
> -            ['include', 'platform/chromium/KeyCodeConversionGtk\\.cpp$'],

Why can't we delete KeyCodeConversionGtk.cpp?

> Source/WebCore/WebCore.gyp/WebCore.gyp:1532
> +        ['use_x11 == 0', {
>            'sources/': [
>              ['exclude', '(Gtk|Linux)\\.cpp$'],
>              ['exclude', 'Harfbuzz[^/]+\\.(cpp|h)$'],
>            ],
>          }],

Can we move this to an else clause of the use_x11 == 1 above?

> Source/WebCore/WebCore.gyp/WebCore.gyp:1635
> +        ['use_x11 == 0', {
>            'sources/': [
>              ['exclude', '(Gtk|Linux)\\.cpp$'],
>            ],

This one is weird.  Should we split the Gtk and Linux cases into two conditions?

> Source/WebCore/WebCore.gyp/WebCore.gyp:1765
> -        ['toolkit_uses_gtk == 0', {
> +        ['use_x11 == 0', {
>            'sources/': [
>              ['exclude', '(Gtk|Linux)\\.cpp$'],

ditto.

> Source/WebKit/chromium/ChangeLog:8
> +        public/linux/WebFontInfo.h is a copy of public/gtk/WebFontInfo.h  
> +
> +        The original file will be deleted once the appropriate changes have landed in Chromium.

One way to make this more clear is to have the old header be a forwarding header.  It would avoid the confusion about having 2 identical headers in the tree.
Comment 3 Fady Samuel 2011-09-28 05:54:32 PDT
Created attachment 109011 [details]
Patch
Comment 4 WebKit Review Bot 2011-09-28 05:57:56 PDT
Attachment 109011 [details] did not pass style-queue:

Failed to run "['Tools/Scripts/check-webkit-style', '--diff-files', u'Source/WebCore/ChangeLog', u'Source/WebCor..." exit_code: 1

Source/WebKit/chromium/public/gtk/WebFontInfo.h:31:  #ifndef header guard has wrong style, please use: WebFontInfo_h  [build/header_guard] [5]
Total errors found: 1 in 6 files


If any of these errors are false positives, please file a bug against check-webkit-style.
Comment 5 Fady Samuel 2011-09-28 06:00:31 PDT
Comment on attachment 109011 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=109011&action=review

>> Source/WebKit/chromium/public/gtk/WebFontInfo.h:31
>> +#ifndef Gtk_WebFontInfo_h
> 
> #ifndef header guard has wrong style, please use: WebFontInfo_h  [build/header_guard] [5]

Hmm, if both WebFontInfo files use the same header guard, could that cause problems?
Comment 6 Fady Samuel 2011-09-28 08:42:19 PDT
Created attachment 109024 [details]
Patch
Comment 7 Tony Chang 2011-09-28 09:47:16 PDT
Comment on attachment 109024 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=109024&action=review

> Source/WebCore/WebCore.gyp/WebCore.gyp:1642
> +        ['use_x11 == 0', {
> +          'sources/': [
> +            ['exclude', 'Linux\\.cpp$'],

It might be more accurate to do 'os_posix and OS!="mac"', but this is OK too.
Comment 8 WebKit Review Bot 2011-09-28 10:35:13 PDT
Comment on attachment 109024 [details]
Patch

Clearing flags on attachment: 109024

Committed r96233: <http://trac.webkit.org/changeset/96233>
Comment 9 WebKit Review Bot 2011-09-28 10:35:17 PDT
All reviewed patches have been landed.  Closing bug.