Add cc/ to banned_directories in chromium clang style checker

Add cc/ to banned_directories in chromium clang style checker

The chromium compositor code will be imported from WebKit into this directory
soon. Since it's WebKit style, it doesn't currently adhere to all of the clang
style checks. This adds a temporary exemption so we don't spam the heck out of
everyone while we clean things up.

BUG=none
NOTRY=true

Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=150013

[Nico, 2012-08-04 00:51:09]

https://chromiumcodereview.appspot.com/10853012/diff/1/tools/clang/plugins/Ch...
File tools/clang/plugins/ChromeClassTester.cpp (right):

https://chromiumcodereview.appspot.com/10853012/diff/1/tools/clang/plugins/Ch...
tools/clang/plugins/ChromeClassTester.cpp:165:
banned_directories_.push_back("cc/");
Do you want to make this conditional on a flag? If so, you could remove this by
changing a flag, which is possible with a normal chromium CL. That way you
wouldn't be blocked on another clang roll when you want to turn this on. You
could also try turning this on in tryjobs to see how much work is left etc

[jamesr, 2012-08-04 00:52:10]

https://chromiumcodereview.appspot.com/10853012/diff/1/tools/clang/plugins/Ch...
File tools/clang/plugins/ChromeClassTester.cpp (right):

https://chromiumcodereview.appspot.com/10853012/diff/1/tools/clang/plugins/Ch...
tools/clang/plugins/ChromeClassTester.cpp:165:
banned_directories_.push_back("cc/");
On 2012/08/04 00:51:09, Nico wrote:
> Do you want to make this conditional on a flag? If so, you could remove this
by
> changing a flag, which is possible with a normal chromium CL. That way you
> wouldn't be blocked on another clang roll when you want to turn this on. You
> could also try turning this on in tryjobs to see how much work is left etc

Sure, how do I get access to flags from a clang plugin?

[Nico, 2012-08-04 00:57:22]

On 2012/08/04 00:52:10, jamesr wrote:
>
https://chromiumcodereview.appspot.com/10853012/diff/1/tools/clang/plugins/Ch...
> File tools/clang/plugins/ChromeClassTester.cpp (right):
> 
>
https://chromiumcodereview.appspot.com/10853012/diff/1/tools/clang/plugins/Ch...
> tools/clang/plugins/ChromeClassTester.cpp:165:
> banned_directories_.push_back("cc/");
> On 2012/08/04 00:51:09, Nico wrote:
> > Do you want to make this conditional on a flag? If so, you could remove this
> by
> > changing a flag, which is possible with a normal chromium CL. That way you
> > wouldn't be blocked on another clang roll when you want to turn this on. You
> > could also try turning this on in tryjobs to see how much work is left etc
> 
> Sure, how do I get access to flags from a clang plugin?

See https://chromiumcodereview.appspot.com/10828057

[jamesr, 2012-08-04 01:45:26]

This appears to work, based on some basic local checks.  I like it that you have
to pass 4 command line arguments to get 1 arg into a plugin (-Xclang
-plugin-arg-... -Xclang real-argument...).

[Nico, 2012-08-04 02:01:39]

lgtm

Yes, the 4 arguments kinda suck! That's because plugins are currently only a cc1
feature, the driver knowns nothing about them (driver := the program that
decides what needs to be done for a certain input – eg compile, then link – and
that forks / execs for these tasks to happen. cc1 := the actual compiler. in
clang, cc1 is the same binary, and passing -cc1 makes it run in cc1 mode. in
gcc, they are distinct binaries afaik </deviation>), and -Xclang is needed to
pass arguments through to cc1. There's a patch floating on the cfe-dev list to
add a real supported plugin api to clang. Maybe it'll happen eventually.

[Nico, 2012-08-04 02:01:58]

(ps: there's no point in cq'ing this as the cqs don't test plugin source
changes)

[commit-bot: I haz the power, 2012-08-04 02:54:02]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/jamesr@chromium.org/10853012/4002

[jamesr, 2012-08-04 02:54:20]

On 2012/08/04 02:01:58, Nico wrote:
> (ps: there's no point in cq'ing this as the cqs don't test plugin source
> changes)

I'm just a wild rebel!

[commit-bot: I haz the power, 2012-08-04 04:31:46]

Try job failure for 10853012-4002 (retry) (retry) on win_rel for step "compile"
(clobber build).
It's a second try, previously, step "compile" failed.
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=win_rel&nu...

[Nico, 2012-08-04 04:37:38]

lolololol

You can add NOTRY=true when you cq to be even more a rebel.

[commit-bot: I haz the power, 2012-08-04 04:40:25]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/jamesr@chromium.org/10853012/4002

[commit-bot: I haz the power, 2012-08-04 04:40:35]

Change committed as 150013