Add several repaint tests in case of table resize

Add several repaint tests in case of table resize

BUG=391177
R=dsinclair@chromium.org

Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=179060

[Xianzhu, 2014-07-23 16:47:29]

ping...

[dsinclair, 2014-07-23 16:57:29]

https://codereview.chromium.org/403723004/diff/40001/LayoutTests/fast/table/r...
File LayoutTests/fast/table/resize-table-repaint-percent-size-cell.html (right):

https://codereview.chromium.org/403723004/diff/40001/LayoutTests/fast/table/r...
LayoutTests/fast/table/resize-table-repaint-percent-size-cell.html:25: <table
id="table">
Can you add a description of what it means for this test to be successful?

https://codereview.chromium.org/403723004/diff/40001/LayoutTests/fast/table/r...
File LayoutTests/fast/table/resize-table-repaint-vertical-align-cell.html
(right):

https://codereview.chromium.org/403723004/diff/40001/LayoutTests/fast/table/r...
LayoutTests/fast/table/resize-table-repaint-vertical-align-cell.html:22: <table
id="table">
Can you add a description of what it means for this test to be successful?

https://codereview.chromium.org/403723004/diff/40001/LayoutTests/fast/table/r...
File LayoutTests/fast/table/resize-table-row-repaint.html (right):

https://codereview.chromium.org/403723004/diff/40001/LayoutTests/fast/table/r...
LayoutTests/fast/table/resize-table-row-repaint.html:20: <table>
Success description?

https://codereview.chromium.org/403723004/diff/40001/Source/core/rendering/Re...
File Source/core/rendering/RenderTable.cpp (right):

https://codereview.chromium.org/403723004/diff/40001/Source/core/rendering/Re...
Source/core/rendering/RenderTable.cpp:558: // FIXME: It would be better to let
the changed cells invalidate themselves.
crbug?

Do you have an idea of how hard this would be do to the correct way of having
the cells do their own invalidation?

[Xianzhu, 2014-07-23 17:31:48]

https://codereview.chromium.org/403723004/diff/40001/LayoutTests/fast/table/r...
File LayoutTests/fast/table/resize-table-repaint-percent-size-cell.html (right):

https://codereview.chromium.org/403723004/diff/40001/LayoutTests/fast/table/r...
LayoutTests/fast/table/resize-table-repaint-percent-size-cell.html:25: <table
id="table">
On 2014/07/23 16:57:29, dsinclair wrote:
> Can you add a description of what it means for this test to be successful?

Done.

https://codereview.chromium.org/403723004/diff/40001/LayoutTests/fast/table/r...
File LayoutTests/fast/table/resize-table-repaint-vertical-align-cell.html
(right):

https://codereview.chromium.org/403723004/diff/40001/LayoutTests/fast/table/r...
LayoutTests/fast/table/resize-table-repaint-vertical-align-cell.html:22: <table
id="table">
On 2014/07/23 16:57:29, dsinclair wrote:
> Can you add a description of what it means for this test to be successful?

Done.

https://codereview.chromium.org/403723004/diff/40001/LayoutTests/fast/table/r...
File LayoutTests/fast/table/resize-table-row-repaint.html (right):

https://codereview.chromium.org/403723004/diff/40001/LayoutTests/fast/table/r...
LayoutTests/fast/table/resize-table-row-repaint.html:20: <table>
On 2014/07/23 16:57:29, dsinclair wrote:
> Success description?

Done.

https://codereview.chromium.org/403723004/diff/40001/Source/core/rendering/Re...
File Source/core/rendering/RenderTable.cpp (right):

https://codereview.chromium.org/403723004/diff/40001/Source/core/rendering/Re...
Source/core/rendering/RenderTable.cpp:558: // FIXME: It would be better to let
the changed cells invalidate themselves.
On 2014/07/23 16:57:29, dsinclair wrote:
> crbug?

Filed crbug.com/396674.

> 
> Do you have an idea of how hard this would be do to the correct way of having
> the cells do their own invalidation?

Table rendering code looks complex to me and I'm unlikely to work out an optimal
solution before M38 branch. I think if I missed the branch even if I could work
out an optimal solution it would be difficult and risky to merge because of the
complexity.

[dsinclair, 2014-07-25 14:45:27]

lgtm

[Julien - ping for review, 2014-07-28 17:23:19]

https://codereview.chromium.org/403723004/diff/60001/LayoutTests/fast/table/r...
File LayoutTests/fast/table/resize-table-repaint-vertical-align-cell.html
(right):

https://codereview.chromium.org/403723004/diff/60001/LayoutTests/fast/table/r...
LayoutTests/fast/table/resize-table-repaint-vertical-align-cell.html:23: The
result repaint rects should cover the middle-vertical-aligned table cells. -->
Let's dump this into the output.

https://codereview.chromium.org/403723004/diff/60001/LayoutTests/fast/table/r...
File LayoutTests/fast/table/resize-table-row-repaint.html (right):

https://codereview.chromium.org/403723004/diff/60001/LayoutTests/fast/table/r...
LayoutTests/fast/table/resize-table-row-repaint.html:21: The result repaint
rects should cover the changed table cells. -->
Same comment here. This is a blocker for me not to dump important information
into the test output when we can. Do NOT land this patch this fixed.

https://codereview.chromium.org/403723004/diff/60001/Source/core/rendering/Re...
File Source/core/rendering/RenderTable.cpp (right):

https://codereview.chromium.org/403723004/diff/60001/Source/core/rendering/Re...
Source/core/rendering/RenderTable.cpp:558: // FIXME: It would be better to
invalidate the changed part only. crbug.com/396674.
We talked and I would rather disable the optimization for tables than work
around it during layout. That makes it easier to see what's going on.

[Xianzhu, 2014-07-28 20:35:37]

Just tested on ToT and found that resizing table can correctly invalidate table
cells. It has been working since
https://src.chromium.org/viewvc/blink?view=rev&revision=178719. Previously only
the table is incrementally invalidated because the invalidation flags of table
cells were not checked in invalidateTreeIfNeeded().

So the code change is not needed. Will just land the layout tests.

[Xianzhu, 2014-07-28 20:55:28]

https://codereview.chromium.org/403723004/diff/60001/LayoutTests/fast/table/r...
File LayoutTests/fast/table/resize-table-repaint-vertical-align-cell.html
(right):

https://codereview.chromium.org/403723004/diff/60001/LayoutTests/fast/table/r...
LayoutTests/fast/table/resize-table-repaint-vertical-align-cell.html:23: The
result repaint rects should cover the middle-vertical-aligned table cells. -->
On 2014/07/28 17:23:19, Julien Chaffraix - PST wrote:
> Let's dump this into the output.

Done.

https://codereview.chromium.org/403723004/diff/60001/LayoutTests/fast/table/r...
File LayoutTests/fast/table/resize-table-row-repaint.html (right):

https://codereview.chromium.org/403723004/diff/60001/LayoutTests/fast/table/r...
LayoutTests/fast/table/resize-table-row-repaint.html:21: The result repaint
rects should cover the changed table cells. -->
On 2014/07/28 17:23:19, Julien Chaffraix - PST wrote:
> Same comment here. This is a blocker for me not to dump important information
> into the test output when we can. Do NOT land this patch this fixed.

Done.

[Xianzhu, 2014-07-28 20:55:38]

The CQ bit was checked by wangxianzhu@chromium.org

[commit-bot: I haz the power, 2014-07-28 20:55:54]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/wangxianzhu@chromium.org/403723004/80001

[commit-bot: I haz the power, 2014-07-28 22:07:17]

FYI, CQ is re-trying this CL (attempt #1).
The failing builders are:
  win_blink_compile_dbg on tryserver.blink
(http://build.chromium.org/p/tryserver.blink/builders/win_blink_compile_dbg/bu...)
  mac_gpu_retina_triggered_tests on tryserver.chromium.gpu
(http://build.chromium.org/p/tryserver.chromium.gpu/builders/mac_gpu_retina_tr...)
  mac_gpu_triggered_tests on tryserver.chromium.gpu
(http://build.chromium.org/p/tryserver.chromium.gpu/builders/mac_gpu_triggered...)
  win_gpu_triggered_tests on tryserver.chromium.gpu
(http://build.chromium.org/p/tryserver.chromium.gpu/builders/win_gpu_triggered...)

[Xianzhu, 2014-07-28 22:21:45]

Committed patchset #5 manually as r179060 (presubmit successful).