Add code style check error for using static_cast instead of toFoo function.

Add code style check error for using static_cast instead of toFoo function.

The original CL https://chromiumcodereview.appspot.com/23654034/ was reverted as -
1. It effected pushing of new CL with static_cast, though this issue exists at many places
   in current code base. It was agreed to first fix all pain points in current codebase,
   make it warning for now and as devs become ready for this guideline, make it an error.
2. It was causing a regression (http://crbug.com/295673).
3. There was a (perf) issue in unittests execution.
3. It was causing a regression (probably, not sure) in test-webkitpy and WebKit XP bots.

In this modified CL, I have addressed issues 1, 2 & 3. As I am not aware what was broken on
bots, I haven't addressed it yet. Based on input for issue 4, I will correct the patch.

R=dpranke,inferno,ojan,eseidel

BUG=295673
TESTS=Unit tests added in cpp_unittest.py; no behavior changes;

Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=158628

[r.kasibhatla, 2013-09-24 12:55:38]

PTAL.

On crrev.com/23654034, it was suggested to make it a warning for now and
sometime later to make it as error. I am not sure how to make it as warning,
hence I have changed the category to runtime/casting (which is by default
filtered). We can report these errors by removing runtime/casting from filter
list. I have also corrected the regression reported on http://crbug.com/295673
as part of this.

@dpranke: As per your comments on crrev.com/23654034, I have corrected 2 issues
- getting absolute path for the header file & unittests should avoid the file
search and instead mock the header.
I couldn't avoid os.walk to find the exact header file, so need your help to
correct it if possible. Also, unittests test-webkitpy didn't fail for me locally
nor I observed any failure on bots (when the earlier patch was tried on
trybots). So, if can you tell me the error you saw I will fix it (if need to).

[r.kasibhatla, 2013-09-26 16:44:15]

On 2013/09/24 12:55:38, r.kasibhatla wrote:
> PTAL.
> 
> On crrev.com/23654034, it was suggested to make it a warning for now and
> sometime later to make it as error. I am not sure how to make it as warning,
> hence I have changed the category to runtime/casting (which is by default
> filtered). We can report these errors by removing runtime/casting from filter
> list. I have also corrected the regression reported on http://crbug.com/295673
> as part of this.
> 
> @dpranke: As per your comments on crrev.com/23654034, I have corrected 2
issues
> - getting absolute path for the header file & unittests should avoid the file
> search and instead mock the header.
> I couldn't avoid os.walk to find the exact header file, so need your help to
> correct it if possible. Also, unittests test-webkitpy didn't fail for me
locally
> nor I observed any failure on bots (when the earlier patch was tried on
> trybots). So, if can you tell me the error you saw I will fix it (if need to).
Can anybody PTAL on this modified version? I addressed couple of errors and
comments given after commit of previous version.
Also, I would like to know the failures in unit tests with earlier committed
version (so as to address if any).

[eseidel, 2013-09-26 16:46:18]

Dirk is really your best reviewer here.  Sorry, we've been distracted by BlinkON
these last couple days.  Today is the deadline for internal performance reviews,
so I expect Dirk may be distracted until tomorrow as well. :(

[Dirk Pranke, 2013-09-26 19:01:35]

This is getting closer. Make the minor suggested changes and run the try jobs
and see if everything runs cleanly and I'm okay with this for now.

This is a somewhat cursory review, however. I'm actually OOO today and for the
next few days, so eseidel or ojan or inferno can pick this up if it needs more
work.

Thanks (and sorry for the delay).

https://codereview.chromium.org/24406002/diff/1/Tools/Scripts/webkitpy/style/...
File Tools/Scripts/webkitpy/style/checkers/cpp.py (right):

https://codereview.chromium.org/24406002/diff/1/Tools/Scripts/webkitpy/style/...
Tools/Scripts/webkitpy/style/checkers/cpp.py:3475: io =
_unit_test_config.get(INCLUDE_IO_INJECTION_KEY, codecs)
Ick. Can you file a separate bug to clean up all of this stuff to use a host
object and get rid of this unit_test_config() scaffolding?

[r.kasibhatla, 2013-09-26 19:07:43]

Also, I don't have permission to run try bots, so somebody has to run on my
patch. :).

https://codereview.chromium.org/24406002/diff/1/Tools/Scripts/webkitpy/style/...
File Tools/Scripts/webkitpy/style/checkers/cpp.py (right):

https://codereview.chromium.org/24406002/diff/1/Tools/Scripts/webkitpy/style/...
Tools/Scripts/webkitpy/style/checkers/cpp.py:3475: io =
_unit_test_config.get(INCLUDE_IO_INJECTION_KEY, codecs)
On 2013/09/26 19:01:35, Dirk Pranke wrote:
> Ick. Can you file a separate bug to clean up all of this stuff to use a host
> object and get rid of this unit_test_config() scaffolding?
Sorry for re-asking, but should I remove usage of unit_test_config() from this
patch also, or it is fine for now in this patch and file a new bug for complete
removal of unit_test_config() at all places (to be replaced by host object).

[Dirk Pranke, 2013-09-26 19:08:58]

I just kicked off some try jobs. Do not attempt to remove unit_test_config() as
part of this patch, that should be a separate thing :).

[r.kasibhatla, 2013-09-27 03:49:26]

On 2013/09/26 19:08:58, Dirk Pranke wrote:
> I just kicked off some try jobs. Do not attempt to remove unit_test_config()
as
> part of this patch, that should be a separate thing :).
It seems strange that the failure seen is in layout test
(virtual/gpu/fast/canvas/canvas-blending-pattern-over-pattern.html), though
there is no code change at all in this CL. Can somebody please trigger trybot
again?

[inferno, 2013-09-27 03:50:09]

On 2013/09/27 03:49:26, r.kasibhatla wrote:
> On 2013/09/26 19:08:58, Dirk Pranke wrote:
> > I just kicked off some try jobs. Do not attempt to remove unit_test_config()
> as
> > part of this patch, that should be a separate thing :).
> It seems strange that the failure seen is in layout test
> (virtual/gpu/fast/canvas/canvas-blending-pattern-over-pattern.html), though
> there is no code change at all in this CL. Can somebody please trigger trybot
> again?

done!

[r.kasibhatla, 2013-09-27 05:35:41]

On 2013/09/27 03:50:09, inferno wrote:
> On 2013/09/27 03:49:26, r.kasibhatla wrote:
> > On 2013/09/26 19:08:58, Dirk Pranke wrote:
> > > I just kicked off some try jobs. Do not attempt to remove
unit_test_config()
> > as
> > > part of this patch, that should be a separate thing :).
> > It seems strange that the failure seen is in layout test
> > (virtual/gpu/fast/canvas/canvas-blending-pattern-over-pattern.html), though
> > there is no code change at all in this CL. Can somebody please trigger
trybot
> > again?
> 
> done!
I am not sure what is happening but seems windows build is stopping suddenly for
some unknown reason. In the logs I dont find any failure but result seems
failure. Can somebody help me with whats going?

[inferno, 2013-09-27 16:48:26]

On 2013/09/27 05:35:41, r.kasibhatla wrote:
> On 2013/09/27 03:50:09, inferno wrote:
> > On 2013/09/27 03:49:26, r.kasibhatla wrote:
> > > On 2013/09/26 19:08:58, Dirk Pranke wrote:
> > > > I just kicked off some try jobs. Do not attempt to remove
> unit_test_config()
> > > as
> > > > part of this patch, that should be a separate thing :).
> > > It seems strange that the failure seen is in layout test
> > > (virtual/gpu/fast/canvas/canvas-blending-pattern-over-pattern.html),
though
> > > there is no code change at all in this CL. Can somebody please trigger
> trybot
> > > again?
> > 
> > done!
> I am not sure what is happening but seems windows build is stopping suddenly
for
> some unknown reason. In the logs I dont find any failure but result seems
> failure. Can somebody help me with whats going?

Lets try it one more time. I don't understand why these win bots are failing :(

[r.kasibhatla, 2013-09-28 16:12:55]

On 2013/09/27 16:48:26, inferno wrote:
> On 2013/09/27 05:35:41, r.kasibhatla wrote:
> > On 2013/09/27 03:50:09, inferno wrote:
> > > On 2013/09/27 03:49:26, r.kasibhatla wrote:
> > > > On 2013/09/26 19:08:58, Dirk Pranke wrote:
> > > > > I just kicked off some try jobs. Do not attempt to remove
> > unit_test_config()
> > > > as
> > > > > part of this patch, that should be a separate thing :).
> > > > It seems strange that the failure seen is in layout test
> > > > (virtual/gpu/fast/canvas/canvas-blending-pattern-over-pattern.html),
> though
> > > > there is no code change at all in this CL. Can somebody please trigger
> > trybot
> > > > again?
> > > 
> > > done!
> > I am not sure what is happening but seems windows build is stopping suddenly
> for
> > some unknown reason. In the logs I dont find any failure but result seems
> > failure. Can somebody help me with whats going?
> 
> Lets try it one more time. I don't understand why these win bots are failing
:(
@inferno, @dpranke: I am out of reasons for failures on windows bots. The
failures and the patch have no connection at all. Any suggestions?

[r.kasibhatla, 2013-10-01 03:06:25]

On 2013/09/30 21:28:28, inferno wrote:
win_blink trybot stopped for some reason - I don't see any failure and
win_layout is failing for few layout tests (not sure why). I guess this is again
due to flakiness of win try bots... Should I ask for re-kick of win bots or can
we go ahead with review & commit?

[inferno, 2013-10-01 05:29:38]

lgtm

lgtm. win bots look flaky, you can go ahead and commit with the nit fixed.

https://codereview.chromium.org/24406002/diff/1/Tools/Scripts/webkitpy/style/...
File Tools/Scripts/webkitpy/style/checkers/cpp.py (right):

https://codereview.chromium.org/24406002/diff/1/Tools/Scripts/webkitpy/style/...
Tools/Scripts/webkitpy/style/checkers/cpp.py:3439: _sep =
'src/third_party/WebKit'
This will be overkill since we have ton of layout tests. Restrict it to
src/third_party/WebKit/Source

[r.kasibhatla, 2013-10-01 09:53:46]

Fixed the nit.

https://codereview.chromium.org/24406002/diff/1/Tools/Scripts/webkitpy/style/...
File Tools/Scripts/webkitpy/style/checkers/cpp.py (right):

https://codereview.chromium.org/24406002/diff/1/Tools/Scripts/webkitpy/style/...
Tools/Scripts/webkitpy/style/checkers/cpp.py:3439: _sep =
'src/third_party/WebKit'
On 2013/10/01 05:29:38, inferno wrote:
> This will be overkill since we have ton of layout tests. Restrict it to
> src/third_party/WebKit/Source

Done.

[commit-bot: I haz the power, 2013-10-01 09:54:11]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/r.kasibhatla@samsung.com/24406002/25001

[commit-bot: I haz the power, 2013-10-01 11:41:59]

Retried try job too often on win_blink_rel for step(s) webkit_tests
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=win_blink_...

[commit-bot: I haz the power, 2013-10-01 11:52:33]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/r.kasibhatla@samsung.com/24406002/25001

[commit-bot: I haz the power, 2013-10-01 12:20:42]

Change committed as 158628