Rewrite plugin_flags.sh Bash script in Python.

Rewrite plugin_flags.sh Bash script in Python.

This is necessary because GN's exec_script only runs Python scripts.

This was requested by Nico in https://codereview.chromium.org/971593003/.

BUG=462972
TEST=run ./tools/clang/scripts/plugin_flags.py and compare the output
with the bash version. They should be exactly the same.
R=thakis@chromium.org

Committed: https://crrev.com/070329d32a77224bb62bf9f11fc94ae58a279a79
Cr-Commit-Position: refs/heads/master@{#318924}

[tfarina, 2015-03-02 20:42:32]

Remember when reviewing this that I'm still crawling in Python.

This was formatted with:

./third_party/WebKit/Tools/Scripts/format-webkitpy --chromium
tools/clang/scripts/plugin_flags.py

[Nico, 2015-03-02 20:45:31]

This looks fine (in general, we don't use webkit formatting scripts outside of
webkit, but it has done a decent job here). Can you also delete the old .sh file
and update common.gypi to reference this new file?

https://codereview.chromium.org/966723003/diff/40001/tools/clang/scripts/plug...
File tools/clang/scripts/plugin_flags.py (right):

https://codereview.chromium.org/966723003/diff/40001/tools/clang/scripts/plug...
tools/clang/scripts/plugin_flags.py:35: ' -Xclang strict-virtual-specifiers' %
LIB_PATH
nit: i think the style guide suggests

print ('asdf'
    'asdf'
    asdf') % FOO

for multi-line strings

[tfarina, 2015-03-02 21:11:41]

https://codereview.chromium.org/966723003/diff/40001/tools/clang/scripts/plug...
File tools/clang/scripts/plugin_flags.py (right):

https://codereview.chromium.org/966723003/diff/40001/tools/clang/scripts/plug...
tools/clang/scripts/plugin_flags.py:35: ' -Xclang strict-virtual-specifiers' %
LIB_PATH
On 2015/03/02 20:45:31, Nico wrote:
> nit: i think the style guide suggests
> 
> print ('asdf'
>     'asdf'
>     asdf') % FOO
> 
> for multi-line strings

Done.

[Nico, 2015-03-02 21:20:42]

lgtm, thanks!

[tfarina, 2015-03-03 12:54:34]

The CQ bit was checked by tfarina@chromium.org

[tfarina, 2015-03-03 12:54:34]

The patchset sent to the CQ was uploaded after l-g-t-m from thakis@chromium.org
Link to the patchset: https://codereview.chromium.org/966723003/#ps80001 (title:
"set executable bit")

[commit-bot: I haz the power, 2015-03-03 12:55:25]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/966723003/80001

[commit-bot: I haz the power, 2015-03-03 13:12:17]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2015-03-03 13:12:18]

Try jobs failed on following builders:
  chromium_presubmit on tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/chromium_presub...)

[dcheng, 2015-03-03 13:14:43]

dcheng@chromium.org changed reviewers:
+ dcheng@chromium.org

[dcheng, 2015-03-03 13:14:44]

https://codereview.chromium.org/966723003/diff/80001/tools/clang/scripts/plug...
File tools/clang/scripts/plugin_flags.py (right):

https://codereview.chromium.org/966723003/diff/80001/tools/clang/scripts/plug...
tools/clang/scripts/plugin_flags.py:26: 'libFindBadConstructs' +
Put this all on one line?

[tfarina, 2015-03-03 14:28:55]

The CQ bit was checked by tfarina@chromium.org

[tfarina, 2015-03-03 14:28:56]

The patchset sent to the CQ was uploaded after l-g-t-m from thakis@chromium.org
Link to the patchset: https://codereview.chromium.org/966723003/#ps120001
(title: "git update-index --chmod=+x")

[commit-bot: I haz the power, 2015-03-03 14:30:15]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/966723003/120001

[dcheng, 2015-03-03 14:47:20]

The CQ bit was unchecked by dcheng@chromium.org

[dcheng, 2015-03-03 14:47:40]

On 2015/03/03 at 14:30:15, commit-bot wrote:
> CQ is trying da patch. Follow status at
>  https://chromium-cq-status.appspot.com/patch-status/966723003/120001

Please address my comment?

[tfarina, 2015-03-03 17:11:24]

On 2015/03/03 14:47:40, dcheng wrote:
> On 2015/03/03 at 14:30:15, commit-bot wrote:
> > CQ is trying da patch. Follow status at
> >  https://chromium-cq-status.appspot.com/patch-status/966723003/120001
> 
> Please address my comment?

Why? It was formatted by a tool as I said earlier, and it does not fit in one
line. And Nico was already happy with it.

[dcheng, 2015-03-03 17:16:19]

On 2015/03/03 at 17:11:24, tfarina wrote:
> On 2015/03/03 14:47:40, dcheng wrote:
> > On 2015/03/03 at 14:30:15, commit-bot wrote:
> > > CQ is trying da patch. Follow status at
> > >  https://chromium-cq-status.appspot.com/patch-status/966723003/120001
> > 
> > Please address my comment?
> 
> Why? It was formatted by a tool as I said earlier, and it does not fit in one
line. And Nico was already happy with it.

The tool has generated suboptimal formatting (hmm... where have I heard that
before).

The reason is:
LIB_PATH = os.path.join(
    CLANG_LIB_PATH,
    'libFindBadConstructs' +
    '.' +
    LIBSUFFIX)

looks like it's joining a list of:
CLANG_LIB_PATH
libFindBadConstructs
.
LIBSUFFIX

I would have expected
LIB_PATH = os.path.join(
    CLANG_LIB_PATH,
    'libFindBadConstructs' + '.' + LIBSUFFIX)

But in this case, we should at least combine the . and the libFindBadConstructs
bit, even if we're going to go with the (IMO) unfortunate auto formatting.

In general, even if you disagree with a review comment, it's best to reply so
the reviewer knows you've seen it =)

[tfarina, 2015-03-03 17:20:32]

Done. PTAL!

[dcheng, 2015-03-03 17:22:17]

lgtm, thanks.

[tfarina, 2015-03-03 17:24:15]

The CQ bit was checked by tfarina@chromium.org

[tfarina, 2015-03-03 17:24:15]

The patchset sent to the CQ was uploaded after l-g-t-m from thakis@chromium.org
Link to the patchset: https://codereview.chromium.org/966723003/#ps140001
(title: "dcheng")

[commit-bot: I haz the power, 2015-03-03 17:25:03]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/966723003/140001

[Dirk Pranke, 2015-03-03 17:25:03]

dpranke@chromium.org changed reviewers:
+ dpranke@chromium.org

[Dirk Pranke, 2015-03-03 17:25:04]

I thought I had approved this earlier, but apparently I never pressed 'send'.

Generally speaking I would rather have our source code be auto-formatted than
perfect, and so I would rather not get into the habit of reformatting code that
(a) passes the style guide and (b) was auto-formatted.

However, if you can *rewrite* the code to be more legible and the auto-formatter
is still happy with it, all the better. In this particular case I agree with
Daniel that that line is better rewritten.

(lgtm).

[dcheng, 2015-03-03 17:28:01]

On 2015/03/03 at 17:25:04, dpranke wrote:
> I thought I had approved this earlier, but apparently I never pressed 'send'.
> 
> Generally speaking I would rather have our source code be auto-formatted than
perfect, and so I would rather not get into the habit of reformatting code that
(a) passes the style guide and (b) was auto-formatted.
> 
> However, if you can *rewrite* the code to be more legible and the
auto-formatter is still happy with it, all the better. In this particular case I
agree with Daniel that that line is better rewritten.
> 
> (lgtm).

Right, as a heavy proponent of clang-format, I'm OK with the autoformatting
here. To be clear, my primary objection was the '.' on a standalone line
(especially because it has a special meaning in paths and looks somewhat
deceptive otherwise).

[commit-bot: I haz the power, 2015-03-03 19:13:04]

Committed patchset #8 (id:140001)

[commit-bot: I haz the power, 2015-03-03 19:14:45]

Patchset 8 (id:??) landed as
https://crrev.com/070329d32a77224bb62bf9f11fc94ae58a279a79
Cr-Commit-Position: refs/heads/master@{#318924}