Rebaseline a few fast/repaint tests that we are now passing....

Rebaseline a few fast/repaint tests that we are now passing.

The rendering of repaint areas is still different from upstream due to differences in repaint logic.

R=jparent
TEST=none
BUG=8630
Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=25663

[Julie Parent, 2009-09-08 21:01:13]

I noticed that some of the grey regions do not match with upstream, for example:

theirs:
http://trac.webkit.org/browser/trunk/LayoutTests/platform/mac/fast/repaint/bu...
ours: http://codereview.chromium.org/196047/diff/1/16?column_width=80

Is this just the differences in repaint logic you were referring to?

LGTM

http://codereview.chromium.org/196047/diff/1/2
File webkit/tools/layout_tests/test_expectations.txt (right):

http://codereview.chromium.org/196047/diff/1/2#newcode1481
Line 1481: // Implement layoutTestController.display() and
display() is now implemented right?

[dglazkov, 2009-09-08 21:11:36]

On 2009/09/08 21:01:13, Julie Parent wrote:
> I noticed that some of the grey regions do not match with upstream, for
example:
> 
> theirs:
>
http://trac.webkit.org/browser/trunk/LayoutTests/platform/mac/fast/repaint/bu...
> ours: http://codereview.chromium.org/196047/diff/1/16?column_width=80
> 
> Is this just the differences in repaint logic you were referring to?

Right. They use NSView's native rectangle collection/unification, we just unify
all into one damage rect.

> LGTM
> 
> http://codereview.chromium.org/196047/diff/1/2
> File webkit/tools/layout_tests/test_expectations.txt (right):
> 
> http://codereview.chromium.org/196047/diff/1/2#newcode1481
> Line 1481: // Implement layoutTestController.display() and
> display() is now implemented right?

Yep. I'll remove this once all tests are rebaselined. This is just a small batch
:)

[ojan, 2009-09-08 22:27:47]

On Tue, Sep 8, 2009 at 2:11 PM, <dglazkov@chromium.org> wrote:

> On 2009/09/08 21:01:13, Julie Parent wrote:
>
>> I noticed that some of the grey regions do not match with upstream,
>>
> for example:
>
>  theirs:
>>
>
>
>
http://trac.webkit.org/browser/trunk/LayoutTests/platform/mac/fast/repaint/bu...
>
>> ours: http://codereview.chromium.org/196047/diff/1/16?column_width=80
>>
>
>  Is this just the differences in repaint logic you were referring to?
>>
>
> Right. They use NSView's native rectangle collection/unification, we
> just unify all into one damage rect.


Nit: would be nice to include this description in the change description.
Makes it more clear what the "differences" are for the next person who
stumbles upon this CL while tracking down a layout test failure.

[dglazkov, 2009-09-08 22:32:11]

Very good point. I am adding this to the next batch.

:DG<

On Tue, Sep 8, 2009 at 3:27 PM, Ojan Vafai<ojan@chromium.org> wrote:
> On Tue, Sep 8, 2009 at 2:11 PM, <dglazkov@chromium.org> wrote:
>>
>> On 2009/09/08 21:01:13, Julie Parent wrote:
>>>
>>> I noticed that some of the grey regions do not match with upstream,
>>
>> for example:
>>
>>> theirs:
>>
>>
>>
http://trac.webkit.org/browser/trunk/LayoutTests/platform/mac/fast/repaint/bu...
>>>
>>> ours: http://codereview.chromium.org/196047/diff/1/16?column_width=80
>>
>>> Is this just the differences in repaint logic you were referring to?
>>
>> Right. They use NSView's native rectangle collection/unification, we
>> just unify all into one damage rect.
>
> Nit: would be nice to include this description in the change description.
> Makes it more clear what the "differences" are for the next person who
> stumbles upon this CL while tracking down a layout test failure.