Remove localized App Restrictions resources.

Remove localized App Restrictions resources.

This is causing a ~1.46 mb increase in the binary size of chrome on android.
Preemptively reverting this to facilitate further discussion.

This change introduces a dummy target in gyp that deletes existing localized resources. The target as well as the python script can be removed once all buildbots have run this once (~ a day).
Thankfully this is is not required in gn.

BUG=450548
Committed: https://crrev.com/287196bac815e47ec142a77644a0b3848bb2b33f
Cr-Commit-Position: refs/heads/master@{#324432}

[knn, 2015-04-07 18:14:12]

knn@chromium.org changed reviewers:
+ bauerb@chromium.org

[knn, 2015-04-08 10:16:06]

knn@chromium.org changed reviewers:
+ bartfab@chromium.org

[knn, 2015-04-08 10:16:06]

Oops, forgot to send this. PTAL.

[Bernhard Bauer, 2015-04-08 10:24:19]

https://codereview.chromium.org/1062213002/diff/1/components/policy/tools/rem...
File components/policy/tools/remove_localized_app_restrictions.py (right):

https://codereview.chromium.org/1062213002/diff/1/components/policy/tools/rem...
components/policy/tools/remove_localized_app_restrictions.py:8: resources_dir
specifies the root directory of the localized  app restrictions
Remove one space before "app"

https://codereview.chromium.org/1062213002/diff/1/components/policy/tools/rem...
components/policy/tools/remove_localized_app_restrictions.py:21: os.rmdir(path)
What happens if this fails?

[knn, 2015-04-08 10:40:50]

PTAL.

https://codereview.chromium.org/1062213002/diff/1/components/policy/tools/rem...
File components/policy/tools/remove_localized_app_restrictions.py (right):

https://codereview.chromium.org/1062213002/diff/1/components/policy/tools/rem...
components/policy/tools/remove_localized_app_restrictions.py:8: resources_dir
specifies the root directory of the localized  app restrictions
On 2015/04/08 10:24:18, Bernhard Bauer wrote:
> Remove one space before "app"

Done.

https://codereview.chromium.org/1062213002/diff/1/components/policy/tools/rem...
components/policy/tools/remove_localized_app_restrictions.py:21: os.rmdir(path)
On 2015/04/08 10:24:18, Bernhard Bauer wrote:
> What happens if this fails?

The stamp file will not be written and build will stop.

[Bernhard Bauer, 2015-04-08 12:36:17]

https://codereview.chromium.org/1062213002/diff/1/components/policy/tools/rem...
File components/policy/tools/remove_localized_app_restrictions.py (right):

https://codereview.chromium.org/1062213002/diff/1/components/policy/tools/rem...
components/policy/tools/remove_localized_app_restrictions.py:21: os.rmdir(path)
On 2015/04/08 10:40:50, knn wrote:
> On 2015/04/08 10:24:18, Bernhard Bauer wrote:
> > What happens if this fails?
> 
> The stamp file will not be written and build will stop.

What I meant to ask was, what happens if e.g. the directory is non-empty? I
assume we will throw an exception? Is that going to have an effect on the rest
of the build? Do we want to gracefully handle that case (if only to get a nicer
error message when this is an expected failure mode)?

[knn, 2015-04-08 12:52:54]

https://codereview.chromium.org/1062213002/diff/1/components/policy/tools/rem...
File components/policy/tools/remove_localized_app_restrictions.py (right):

https://codereview.chromium.org/1062213002/diff/1/components/policy/tools/rem...
components/policy/tools/remove_localized_app_restrictions.py:21: os.rmdir(path)
On 2015/04/08 12:36:17, Bernhard Bauer wrote:
> On 2015/04/08 10:40:50, knn wrote:
> > On 2015/04/08 10:24:18, Bernhard Bauer wrote:
> > > What happens if this fails?
> > 
> > The stamp file will not be written and build will stop.
> 
> What I meant to ask was, what happens if e.g. the directory is non-empty? I
> assume we will throw an exception? Is that going to have an effect on the rest
> of the build? Do we want to gracefully handle that case (if only to get a
nicer
> error message when this is an expected failure mode)?

If the directory is not empty, we are packing junk resource files which
shouldn't be happening. I do not expect this script to run more than once with
the old resources. If someone manually fiddles with the temporary generated
files isn't it fair to expect them to clean up as well?
Since there should be no sane way to get to an non-empty directory, I am not
handling it. The error message is the only thing we know about the failure as
well.

[Bernhard Bauer, 2015-04-08 12:58:54]

lgtm

https://codereview.chromium.org/1062213002/diff/1/components/policy/tools/rem...
File components/policy/tools/remove_localized_app_restrictions.py (right):

https://codereview.chromium.org/1062213002/diff/1/components/policy/tools/rem...
components/policy/tools/remove_localized_app_restrictions.py:21: os.rmdir(path)
On 2015/04/08 12:52:54, knn wrote:
> On 2015/04/08 12:36:17, Bernhard Bauer wrote:
> > On 2015/04/08 10:40:50, knn wrote:
> > > On 2015/04/08 10:24:18, Bernhard Bauer wrote:
> > > > What happens if this fails?
> > > 
> > > The stamp file will not be written and build will stop.
> > 
> > What I meant to ask was, what happens if e.g. the directory is non-empty? I
> > assume we will throw an exception? Is that going to have an effect on the
rest
> > of the build? Do we want to gracefully handle that case (if only to get a
> nicer
> > error message when this is an expected failure mode)?
> 
> If the directory is not empty, we are packing junk resource files which
> shouldn't be happening. I do not expect this script to run more than once with
> the old resources. If someone manually fiddles with the temporary generated
> files isn't it fair to expect them to clean up as well?
> Since there should be no sane way to get to an non-empty directory, I am not
> handling it. The error message is the only thing we know about the failure as
> well.

Okay... I'm just worried that someone might rightfully add something here, and
then suddenly something would break, but I guess that's not going to be likely
if this is only temporary anyway.

[bartfab (slow), 2015-04-09 12:04:14]

LGTM with a question: Why is it necessary to remove these generated files on
build machines? We have plenty of obsolete object files etc. lying around. Why
can we not just stop generating i18n-ized resources but leave the stale ones
lying around?

https://codereview.chromium.org/1062213002/diff/20001/components/policy.gypi
File components/policy.gypi (right):

https://codereview.chromium.org/1062213002/diff/20001/components/policy.gypi#...
components/policy.gypi:382: # TODO(knn): Remove after all build bots have run
Nit: Please reference a bug number.

https://codereview.chromium.org/1062213002/diff/20001/components/policy/tools...
File components/policy/tools/remove_localized_app_restrictions.py (right):

https://codereview.chromium.org/1062213002/diff/20001/components/policy/tools...
components/policy/tools/remove_localized_app_restrictions.py:2: # Copyright (c)
2015 The Chromium Authors. All rights reserved.
Nit: s/(c) //

[knn, 2015-04-09 13:07:43]

We cannot leave the stale resource files as Gyp only knows about the folder
where the resources exist and not the individual files. Any existing file would
be used without warning.
This is not the case in GN thankfully :)

https://codereview.chromium.org/1062213002/diff/20001/components/policy.gypi
File components/policy.gypi (right):

https://codereview.chromium.org/1062213002/diff/20001/components/policy.gypi#...
components/policy.gypi:382: # TODO(knn): Remove after all build bots have run
On 2015/04/09 12:04:14, bartfab wrote:
> Nit: Please reference a bug number.

Done.

https://codereview.chromium.org/1062213002/diff/20001/components/policy/tools...
File components/policy/tools/remove_localized_app_restrictions.py (right):

https://codereview.chromium.org/1062213002/diff/20001/components/policy/tools...
components/policy/tools/remove_localized_app_restrictions.py:2: # Copyright (c)
2015 The Chromium Authors. All rights reserved.
On 2015/04/09 12:04:14, bartfab wrote:
> Nit: s/(c) //

Done.

[knn, 2015-04-09 14:09:09]

The CQ bit was checked by knn@chromium.org

[knn, 2015-04-09 14:09:09]

The patchset sent to the CQ was uploaded after l-g-t-m from bauerb@chromium.org,
bartfab@chromium.org
Link to the patchset: https://codereview.chromium.org/1062213002/#ps40001
(title: "nit")

[commit-bot: I haz the power, 2015-04-09 14:09:54]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1062213002/40001

[commit-bot: I haz the power, 2015-04-09 14:14:23]

Committed patchset #3 (id:40001)

[commit-bot: I haz the power, 2015-04-09 14:15:15]

Patchset 3 (id:??) landed as
https://crrev.com/287196bac815e47ec142a77644a0b3848bb2b33f
Cr-Commit-Position: refs/heads/master@{#324432}