Sort Java AIDL input filenames for makefile stability.

Sort Java AIDL input filenames for makefile stability.

Sort the list of input java files for the aidl build step so that
the contents of generated makefiles in WebView have a stable ordering.

BUG=

Committed: https://crrev.com/8eff03f99038a90198095d0bab3363a249357cb5
Cr-Commit-Position: refs/heads/master@{#292686}

[Torne, 2014-06-09 14:07:48]

Hi Chris, David,

r173333 introduced this list of files which is in no particular order and thus
causes the webview makefiles to change every time they are regenerated. Since we
have to check them into the android tree this makes a lot of churn :(

Is it okay to just sort them here?

[cjhopman, 2014-06-09 14:51:51]

On 2014/06/09 14:07:48, Torne wrote:
> Hi Chris, David,
> 
> r173333 introduced this list of files which is in no particular order and thus
> causes the webview makefiles to change every time they are regenerated. Since
we
> have to check them into the android tree this makes a lot of churn :(
> 
> Is it okay to just sort them here?

This is just added to 'inputs' which to the ninja generator has no important
order. Could you sort all 'inputs' when writing the makefiles? If not for
whatever reason, this lgtm.

[Torne, 2014-06-09 14:59:15]

On 2014/06/09 14:51:51, cjhopman wrote:
> On 2014/06/09 14:07:48, Torne wrote:
> > Hi Chris, David,
> > 
> > r173333 introduced this list of files which is in no particular order and
thus
> > causes the webview makefiles to change every time they are regenerated.
Since
> we
> > have to check them into the android tree this makes a lot of churn :(
> > 
> > Is it okay to just sort them here?
> 
> This is just added to 'inputs' which to the ninja generator has no important
> order. Could you sort all 'inputs' when writing the makefiles? If not for
> whatever reason, this lgtm.

We could sort this in the android generator, yeah. It doesn't do that at the
moment because it hasn't really come up; the gyp build doesn't often do this
kind of thing.

[David Trainor- moved to gerrit, 2014-06-09 17:59:51]

On 2014/06/09 14:59:15, Torne wrote:
> On 2014/06/09 14:51:51, cjhopman wrote:
> > On 2014/06/09 14:07:48, Torne wrote:
> > > Hi Chris, David,
> > > 
> > > r173333 introduced this list of files which is in no particular order and
> thus
> > > causes the webview makefiles to change every time they are regenerated.
> Since
> > we
> > > have to check them into the android tree this makes a lot of churn :(
> > > 
> > > Is it okay to just sort them here?
> > 
> > This is just added to 'inputs' which to the ninja generator has no important
> > order. Could you sort all 'inputs' when writing the makefiles? If not for
> > whatever reason, this lgtm.
> 
> We could sort this in the android generator, yeah. It doesn't do that at the
> moment because it hasn't really come up; the gyp build doesn't often do this
> kind of thing.

lgtm, but deferring to Chris wrt alternative solutions.

[Torne, 2014-08-29 17:17:08]

I had a look at doing it in the generator, but lots of other parts of the
chromium build actually depend on the ordering of inputs/outputs (e.g.
everywhere that does "python <(inputs)" where the first input is the script to
run), so it looks like if I want stable ordering I have to sort it in each place
that dynamically generates a list in a nondeterministic way, and I really want
stable ordering as it makes it very hard to debug webview build issues when
stuff changes randomly unrelated to what I'm doing.

So I'm gonna go ahead and land this.. sorry ;)

[Torne, 2014-08-29 17:17:12]

The CQ bit was checked by torne@chromium.org

[commit-bot: I haz the power, 2014-08-29 17:18:35]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/torne@chromium.org/323803003/1

[Torne, 2014-08-29 17:27:23]

The CQ bit was unchecked by torne@chromium.org

[Torne, 2014-08-29 17:28:31]

The CQ bit was checked by torne@chromium.org

[commit-bot: I haz the power, 2014-08-29 17:29:56]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/torne@chromium.org/323803003/20001

[commit-bot: I haz the power, 2014-08-29 18:43:27]

FYI, CQ is re-trying this CL (attempt #1).
The failing builders are:
  win_chromium_rel_swarming on tryserver.chromium.win
(http://build.chromium.org/p/tryserver.chromium.win/builders/win_chromium_rel_...)

[commit-bot: I haz the power, 2014-08-29 22:16:41]

Committed patchset #2 (id:20001) as 813d3d9165125da4c738c2ce17762a204057344e

[commit-bot: I haz the power, 2014-09-10 03:10:39]

Patchset 2 (id:??) landed as
https://crrev.com/8eff03f99038a90198095d0bab3363a249357cb5
Cr-Commit-Position: refs/heads/master@{#292686}