WebKit Bugzilla
New
Browse
Log In
×
Sign in with GitHub
or
Remember my login
Create Account
·
Forgot Password
Forgotten password account recovery
RESOLVED FIXED
Bug 71249
Switch RenderObject to layout abstraction
https://bugs.webkit.org/show_bug.cgi?id=71249
Summary
Switch RenderObject to layout abstraction
Emil A Eklund
Reported
2011-10-31 15:54:59 PDT
Switch RenderObject to layout type abstraction.
Attachments
Patch
(18.35 KB, patch)
2011-10-31 15:56 PDT
,
Emil A Eklund
no flags
Details
Formatted Diff
Diff
Patch
(17.46 KB, patch)
2011-10-31 16:47 PDT
,
Emil A Eklund
no flags
Details
Formatted Diff
Diff
Patch
(16.66 KB, patch)
2011-10-31 17:32 PDT
,
Emil A Eklund
no flags
Details
Formatted Diff
Diff
Patch
(16.23 KB, patch)
2011-11-01 10:38 PDT
,
Emil A Eklund
no flags
Details
Formatted Diff
Diff
Show Obsolete
(3)
View All
Add attachment
proposed patch, testcase, etc.
Emil A Eklund
Comment 1
2011-10-31 15:56:40 PDT
Created
attachment 113097
[details]
Patch
Darin Adler
Comment 2
2011-10-31 16:06:51 PDT
Comment on
attachment 113097
[details]
Patch View in context:
https://bugs.webkit.org/attachment.cgi?id=113097&action=review
> Source/WebCore/rendering/RenderObject.cpp:794 > + int width = roundf((side == BSTop || side == BSBottom ? y2 - y1 : x2 - x1));
The round function specifically takes a “float” so is not appropriate to round a LayoutUnit to an int. If LayoutUnit was a double or int it would be quite inefficient!
> Source/WebCore/rendering/RenderObject.h:1052 > -inline int adjustForAbsoluteZoom(int value, RenderObject* renderer) > +inline float adjustForAbsoluteZoom(int value, RenderObject* renderer)
I think this change should be landed separately with its own review and rationale.
Emil A Eklund
Comment 3
2011-10-31 16:47:53 PDT
Created
attachment 113107
[details]
Patch
Emil A Eklund
Comment 4
2011-10-31 16:49:12 PDT
(In reply to
comment #2
)
> The round function specifically takes a “float” so is not appropriate to round a LayoutUnit to an int. If LayoutUnit was a double or int it would be quite inefficient!
Good catch, I should have caught that. Thanks!
> > > Source/WebCore/rendering/RenderObject.h:1052 > > -inline int adjustForAbsoluteZoom(int value, RenderObject* renderer) > > +inline float adjustForAbsoluteZoom(int value, RenderObject* renderer) > > I think this change should be landed separately with its own review and rationale.
Reverted this change, leaving it for later. Please take another look.
Darin Adler
Comment 5
2011-10-31 16:51:14 PDT
Comment on
attachment 113107
[details]
Patch Patch looks good, but does not apply in the EWS.
Emil A Eklund
Comment 6
2011-10-31 17:32:53 PDT
Created
attachment 113115
[details]
Patch
Daniel Bates
Comment 7
2011-10-31 20:28:13 PDT
Comment on
attachment 113115
[details]
Patch
Attachment 113115
[details]
did not pass mac-ews (mac): Output:
http://queues.webkit.org/results/10241949
Emil A Eklund
Comment 8
2011-11-01 10:38:41 PDT
Created
attachment 113191
[details]
Patch
Emil A Eklund
Comment 9
2011-11-01 10:41:55 PDT
Thanks Darin!
WebKit Review Bot
Comment 10
2011-11-01 16:02:37 PDT
Comment on
attachment 113191
[details]
Patch Clearing flags on attachment: 113191 Committed
r99008
: <
http://trac.webkit.org/changeset/99008
>
WebKit Review Bot
Comment 11
2011-11-01 16:02:43 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