WebKit Bugzilla
New
Browse
Log In
×
Sign in with GitHub
or
Remember my login
Create Account
·
Forgot Password
Forgotten password account recovery
RESOLVED FIXED
89837
[WK2][Qt][EFL] Extract common code from LayerTreeHostQt
https://bugs.webkit.org/show_bug.cgi?id=89837
Summary
[WK2][Qt][EFL] Extract common code from LayerTreeHostQt
YoungTaeck Song
Reported
2012-06-24 17:07:30 PDT
Extract common code from LayerTreeHostQt to be used by both Qt and Efl. Because Extracted codes use WebGraphicsLayer, these are named 'LayerTreeHostWeb'.
Attachments
patch
(60.55 KB, patch)
2012-06-24 17:11 PDT
,
YoungTaeck Song
no flags
Details
Formatted Diff
Diff
patch
(60.59 KB, patch)
2012-06-24 20:39 PDT
,
YoungTaeck Song
noam
: review-
noam
: commit-queue-
Details
Formatted Diff
Diff
patch
(173.85 KB, patch)
2012-06-25 04:38 PDT
,
YoungTaeck Song
gyuyoung.kim
: commit-queue-
Details
Formatted Diff
Diff
patch
(173.84 KB, patch)
2012-06-25 04:49 PDT
,
YoungTaeck Song
gyuyoung.kim
: commit-queue-
Details
Formatted Diff
Diff
patch
(175.97 KB, patch)
2012-06-25 05:34 PDT
,
YoungTaeck Song
no flags
Details
Formatted Diff
Diff
Patch
(179.23 KB, patch)
2012-06-25 18:28 PDT
,
YoungTaeck Song
no flags
Details
Formatted Diff
Diff
Show Obsolete
(5)
View All
Add attachment
proposed patch, testcase, etc.
YoungTaeck Song
Comment 1
2012-06-24 17:11:06 PDT
Created
attachment 149209
[details]
patch
YoungTaeck Song
Comment 2
2012-06-24 20:39:21 PDT
Created
attachment 149231
[details]
patch rebase
Noam Rosenthal
Comment 3
2012-06-24 21:23:31 PDT
Comment on
attachment 149231
[details]
patch View in context:
https://bugs.webkit.org/attachment.cgi?id=149231&action=review
LayerTreeHostWeb doesn't fit with the rest of the names around. How about we create a LayerTreeCoordinator directory under WebProcess/WebPage, and also move WebGraphicsLayer under that directory? We should rename LayerTreeHostProxy to LayerTreeCoordinatorProxy while we're at it.
> Source/WebKit2/WebProcess/WebPage/web/qt/LayerTreeHostQt.h:38 > +class LayerTreeHostQt : public LayerTreeHostWeb { > +public: > + static PassRefPtr<LayerTreeHostQt> create(WebPage*); > + virtual ~LayerTreeHostQt(); > + > + virtual int64_t directlyCompositedImageKey(WebCore::Image*); > + > +private: > + explicit LayerTreeHostQt(WebPage*); > +};
Virtualizing for this seems like an overkill... I'd rather have non-virtual functions, and implement directlyCompositedImageKey inside a platform-specific #ifdef.
YoungTaeck Song
Comment 4
2012-06-25 04:38:44 PDT
Created
attachment 149275
[details]
patch
WebKit Review Bot
Comment 5
2012-06-25 04:41:51 PDT
Attachment 149275
[details]
did not pass style-queue: Failed to run "['Tools/Scripts/check-webkit-style', '--diff-files', u'Source/WebKit2/ChangeLog', u'Source/WebKit..." exit_code: 1 Source/WebKit2/WebProcess/WebPage/LayerTreeCoordinator/WebGraphicsLayer.cpp:459: Weird number of spaces at line-start. Are you using a 4-space indent? [whitespace/indent] [3] Source/WebKit2/WebProcess/WebPage/LayerTreeCoordinator/WebGraphicsLayer.h:78: This { should be at the end of the previous line [whitespace/braces] [4] Total errors found: 2 in 27 files If any of these errors are false positives, please file a bug against check-webkit-style.
Gyuyoung Kim
Comment 6
2012-06-25 04:42:05 PDT
Comment on
attachment 149275
[details]
patch
Attachment 149275
[details]
did not pass efl-ews (efl): Output:
http://queues.webkit.org/results/13097077
YoungTaeck Song
Comment 7
2012-06-25 04:49:06 PDT
Created
attachment 149276
[details]
patch
Alexis Menard (darktears)
Comment 8
2012-06-25 04:52:44 PDT
Comment on
attachment 149276
[details]
patch To make this refactor more simple, I think you should do the renames of the classes first, then the move and then extract the code. It will be less horrible to review and easier to rollout if needed as it will be simple patches.
Gyuyoung Kim
Comment 9
2012-06-25 04:55:24 PDT
Comment on
attachment 149276
[details]
patch
Attachment 149276
[details]
did not pass efl-ews (efl): Output:
http://queues.webkit.org/results/13088165
YoungTaeck Song
Comment 10
2012-06-25 05:34:17 PDT
Created
attachment 149277
[details]
patch fix build error of EFL
Gustavo Noronha (kov)
Comment 11
2012-06-25 05:43:29 PDT
Comment on
attachment 149277
[details]
patch
Attachment 149277
[details]
did not pass gtk-ews (gtk): Output:
http://queues.webkit.org/results/13090134
Alexis Menard (darktears)
Comment 12
2012-06-25 05:55:00 PDT
(In reply to
comment #11
)
> (From update of
attachment 149277
[details]
) >
Attachment 149277
[details]
did not pass gtk-ews (gtk): > Output:
http://queues.webkit.org/results/13090134
And you would not fight that much with EWS :).
YoungTaeck Song
Comment 13
2012-06-25 06:03:12 PDT
(In reply to
comment #12
)
> (In reply to
comment #11
) > > (From update of
attachment 149277
[details]
[details]) > >
Attachment 149277
[details]
[details] did not pass gtk-ews (gtk): > > Output:
http://queues.webkit.org/results/13090134
> > And you would not fight that much with EWS :).
I'm very sorry. I thought My patch would not affect to a different port.
Noam Rosenthal
Comment 14
2012-06-25 06:24:41 PDT
Comment on
attachment 149277
[details]
patch Looks great! Though the GTK build files need to be updated.
Noam Rosenthal
Comment 15
2012-06-25 13:07:45 PDT
Comment on
attachment 149277
[details]
patch Need to fix build issues.
YoungTaeck Song
Comment 16
2012-06-25 16:43:11 PDT
(In reply to
comment #15
)
> (From update of
attachment 149277
[details]
) > Need to fix build issues.
I'm sorry too late. That's why I'm first time at building GTK. I'll reupload ASAP.
YoungTaeck Song
Comment 17
2012-06-25 18:28:15 PDT
Created
attachment 149417
[details]
Patch
Noam Rosenthal
Comment 18
2012-06-25 18:46:30 PDT
Comment on
attachment 149417
[details]
Patch LGTM!
WebKit Review Bot
Comment 19
2012-06-25 20:41:45 PDT
Comment on
attachment 149417
[details]
Patch Clearing flags on attachment: 149417 Committed
r121221
: <
http://trac.webkit.org/changeset/121221
>
WebKit Review Bot
Comment 20
2012-06-25 20:41:51 PDT
All reviewed patches have been landed. Closing bug.
Note
You need to
log in
before you can comment on or make changes to this bug.
Top of Page
Format For Printing
XML
Clone This Bug