Rename RenderObject::setNeedsLayout to setNeedsLayoutAndFullRepaint

Rename RenderObject::setNeedsLayout to setNeedsLayoutAndFullRepaint

As we do full repaint on selfNeedsLayout, this renaming makes the
requirement explicit and into opting-in style. After this,
we can
- add opting-out-repaint version of setNeedsLayout;
- change setNeedsLayoutForcingRepaint calls to the new setNeedsLayout
  if repaint is not needed.

BUG=358460
Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=174329

[esprehn, 2014-05-19 21:06:02]

What's the new method? As is this change seems to make things worse.

[Xianzhu, 2014-05-19 21:11:31]

On 2014/05/19 21:06:02, esprehn wrote:
> What's the new method? As is this change seems to make things worse.

This change just renames the current method and doesn't change any behavior. The
new method is in https://codereview.chromium.org/265703012/. Separated the
renaming into this to avoid https://codereview.chromium.org/265703012/ from
being too big.

Previously I named the new method setNeedsLayoutWithoutForcingRepaint() (along
with the original setNeedsLayout()). After Julien's comment I changed to this
opting-in style, that is new setNeedsLayout() without forcing repaint and
original setNeedsLayout() renamed to setNeedsLayoutAndFullRepaint().

[esprehn, 2014-05-19 21:14:27]

On 2014/05/19 21:11:31, Xianzhu wrote:
> On 2014/05/19 21:06:02, esprehn wrote:
> > What's the new method? As is this change seems to make things worse.
> 
> This change just renames the current method and doesn't change any behavior.
The
> new method is in https://codereview.chromium.org/265703012/. Separated the
> renaming into this to avoid https://codereview.chromium.org/265703012/ from
> being too big.
> 
> Previously I named the new method setNeedsLayoutWithoutForcingRepaint() (along
> with the original setNeedsLayout()). After Julien's comment I changed to this
> opting-in style, that is new setNeedsLayout() without forcing repaint and
> original setNeedsLayout() renamed to setNeedsLayoutAndFullRepaint().

lgtm, I agree with Julien. We need to end up with setNeedsLayout() again.

[Xianzhu, 2014-05-19 21:24:40]

The CQ bit was checked by wangxianzhu@chromium.org

[commit-bot: I haz the power, 2014-05-19 21:25:06]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/wangxianzhu@chromium.org/280633002/60001

[eseidel, 2014-05-19 21:39:49]

Why does a caller ever want to force both layout and repaint at the same time? 
Shoudn't layout cause the repaint if needed?

[Xianzhu, 2014-05-19 21:44:35]

On 2014/05/19 21:39:49, eseidel wrote:
> Why does a caller ever want to force both layout and repaint at the same time?

> Shoudn't layout cause the repaint if needed?

Though some caller don't want to, RenderObject::setNeedsLayout() will force
repaint. This renaming let the caller know there will be repaint even it doesn't
want. Next step is to implement a no-repaint version of
RenderObject::setNeedsLayout() and let callers that don't want repaint use.

[eseidel, 2014-05-19 22:09:39]

My belief is that although there may be callers who want this repaint behavior
today, they're all wrong.  So once you add the new function, this function
should be "deprecated" and all callers should move off of it and onto your new
function.

[commit-bot: I haz the power, 2014-05-19 22:59:32]

FYI, CQ is re-trying this CL (attempt #1).
Please consider checking whether the failures are real,
and report flakes to chrome-troopers@google.com.
The failing builders are:
  linux_blink_dbg on tryserver.blink
(http://build.chromium.org/p/tryserver.blink/builders/linux_blink_dbg/builds/8902)
  win_blink_rel on tryserver.blink
(http://build.chromium.org/p/tryserver.blink/builders/win_blink_rel/builds/8285)

[commit-bot: I haz the power, 2014-05-20 00:13:50]

Change committed as 174329