android: Create a GN template for create_dist_jar.py

android: Create a GN template for create_dist_jar.py

BUG=648244
Review-Url: https://codereview.chromium.org/2623243002
Cr-Commit-Position: refs/heads/master@{#444054}
Committed: https://chromium.googlesource.com/chromium/src/+/77ad9ee9d1c3ca1ef984900a31cb206bf885b936

[mbonadei1, 2017-01-12 16:04:32]

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

[mbonadei1, 2017-01-12 16:04:32]

I tried this and it works fine.

In this CL I tried to add public_deps to android_library:
https://codereview.chromium.org/2627223003 (assuming that this CL will land).

https://codereview.chromium.org/2623243002/diff/1/build/android/gyp/write_bui...
File build/android/gyp/write_build_config.py (left):

https://codereview.chromium.org/2623243002/diff/1/build/android/gyp/write_bui...
build/android/gyp/write_build_config.py:635: dependency_jars = [c['jar_path']
for c in all_library_deps]
For the "android_library" use case I think that we should just include
direct_library_deps instead of all_library_deps.
If we go with all_library_deps we end up with a huge jar wit lots of stuff not
directly related. Do you agree?

[agrieve, 2017-01-13 13:15:20]

On 2017/01/12 16:04:32, mbonadei1 wrote:
> I tried this and it works fine.
> 
> In this CL I tried to add public_deps to android_library:
> https://codereview.chromium.org/2627223003 (assuming that this CL will land).
> 
>
https://codereview.chromium.org/2623243002/diff/1/build/android/gyp/write_bui...
> File build/android/gyp/write_build_config.py (left):
> 
>
https://codereview.chromium.org/2623243002/diff/1/build/android/gyp/write_bui...
> build/android/gyp/write_build_config.py:635: dependency_jars = [c['jar_path']
> for c in all_library_deps]
> For the "android_library" use case I think that we should just include
> direct_library_deps instead of all_library_deps.
> If we go with all_library_deps we end up with a huge jar wit lots of stuff not
> directly related. Do you agree?

Updated with direct_deps_only option. PTAL.

[agrieve, 2017-01-13 13:15:38]

The CQ bit was checked by agrieve@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2017-01-13 13:15:54]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2017-01-13 14:22:48]

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

[commit-bot: I haz the power, 2017-01-13 14:22:49]

Dry run: This issue passed the CQ dry run.

[mbonadei1, 2017-01-16 09:07:40]

I tried the updated CL: https://codereview.webrtc.org/2637723003/

This creates a libjingle_peerconnection_java jar with ThreadUtils and all the
classes in 'webrtc/base:base_java'.

But I have three concerns:
1. We lose files creation date (it defaults to 2001/01/01 00:00:00)
2. Files dimensions are different
3. The dist_jar has less files than the android_library (look at
org/webrtc/RendererCommon$1.class which is not present in dist_jar)

Old target:
$ jar tvf
out/android/lib.java/webrtc/sdk/android/libjingle_peerconnection_java.jar | grep
RendererCommon
   830 Mon Jan 16 09:39:56 CET 2017 org/webrtc/RendererCommon$1.class
  1994 Mon Jan 16 09:39:56 CET 2017
org/webrtc/RendererCommon$VideoLayoutMeasure.class
  1659 Mon Jan 16 09:39:56 CET 2017 org/webrtc/RendererCommon$YuvUploader.class
   311 Mon Jan 16 09:39:56 CET 2017 org/webrtc/RendererCommon$GlDrawer.class
  1199 Mon Jan 16 09:39:56 CET 2017 org/webrtc/RendererCommon$ScalingType.class
  3326 Mon Jan 16 09:39:56 CET 2017 org/webrtc/RendererCommon.class
   295 Mon Jan 16 09:39:56 CET 2017
org/webrtc/RendererCommon$RendererEvents.class

New target (dist_jar):
$ jar tvf out/android/obj/webrtc/sdk/android/libjingle_peerconnection_java.jar |
grep RendererCommon
   480 Mon Jan 01 00:00:00 CET 2001
org/webrtc/RendererCommon$VideoLayoutMeasure.class
   246 Mon Jan 01 00:00:00 CET 2001 org/webrtc/RendererCommon$YuvUploader.class
   268 Mon Jan 01 00:00:00 CET 2001 org/webrtc/RendererCommon$GlDrawer.class
   514 Mon Jan 01 00:00:00 CET 2001 org/webrtc/RendererCommon$ScalingType.class
   764 Mon Jan 01 00:00:00 CET 2001 org/webrtc/RendererCommon.class
   252 Mon Jan 01 00:00:00 CET 2001
org/webrtc/RendererCommon$RendererEvents.class

[agrieve, 2017-01-16 18:00:48]

On 2017/01/16 09:07:40, mbonadei1 wrote:
> I tried the updated CL: https://codereview.webrtc.org/2637723003/
> 
> This creates a libjingle_peerconnection_java jar with ThreadUtils and all the
> classes in 'webrtc/base:base_java'.
> 
> But I have three concerns:
> 1. We lose files creation date (it defaults to 2001/01/01 00:00:00)
> 2. Files dimensions are different
> 3. The dist_jar has less files than the android_library (look at
> org/webrtc/RendererCommon$1.class which is not present in dist_jar)
> 
> Old target:
> $ jar tvf
> out/android/lib.java/webrtc/sdk/android/libjingle_peerconnection_java.jar |
grep
> RendererCommon
>    830 Mon Jan 16 09:39:56 CET 2017 org/webrtc/RendererCommon$1.class
>   1994 Mon Jan 16 09:39:56 CET 2017
> org/webrtc/RendererCommon$VideoLayoutMeasure.class
>   1659 Mon Jan 16 09:39:56 CET 2017
org/webrtc/RendererCommon$YuvUploader.class
>    311 Mon Jan 16 09:39:56 CET 2017 org/webrtc/RendererCommon$GlDrawer.class
>   1199 Mon Jan 16 09:39:56 CET 2017
org/webrtc/RendererCommon$ScalingType.class
>   3326 Mon Jan 16 09:39:56 CET 2017 org/webrtc/RendererCommon.class
>    295 Mon Jan 16 09:39:56 CET 2017
> org/webrtc/RendererCommon$RendererEvents.class
> 
> New target (dist_jar):
> $ jar tvf out/android/obj/webrtc/sdk/android/libjingle_peerconnection_java.jar
|
> grep RendererCommon
>    480 Mon Jan 01 00:00:00 CET 2001
> org/webrtc/RendererCommon$VideoLayoutMeasure.class
>    246 Mon Jan 01 00:00:00 CET 2001
org/webrtc/RendererCommon$YuvUploader.class
>    268 Mon Jan 01 00:00:00 CET 2001 org/webrtc/RendererCommon$GlDrawer.class
>    514 Mon Jan 01 00:00:00 CET 2001
org/webrtc/RendererCommon$ScalingType.class
>    764 Mon Jan 01 00:00:00 CET 2001 org/webrtc/RendererCommon.class
>    252 Mon Jan 01 00:00:00 CET 2001
> org/webrtc/RendererCommon$RendererEvents.class

Doh! One of the ifs was backwards and it was using .interface.jars. The sizes &
missing files should be fixed now.

As for timestamps, we always intentionally null them out in order to achieve
hermetic builds (building the same thing on two different machines should
produce the same byte-for-byte artifact). git doesn't maintain mtimes, so I
doubt you'd want to?

[mbonadei1, 2017-01-17 08:45:36]

On 2017/01/16 18:00:48, agrieve wrote:
> On 2017/01/16 09:07:40, mbonadei1 wrote:
> > I tried the updated CL: https://codereview.webrtc.org/2637723003/
> > 
> > This creates a libjingle_peerconnection_java jar with ThreadUtils and all
the
> > classes in 'webrtc/base:base_java'.
> > 
> > But I have three concerns:
> > 1. We lose files creation date (it defaults to 2001/01/01 00:00:00)
> > 2. Files dimensions are different
> > 3. The dist_jar has less files than the android_library (look at
> > org/webrtc/RendererCommon$1.class which is not present in dist_jar)
> > 
> > Old target:
> > $ jar tvf
> > out/android/lib.java/webrtc/sdk/android/libjingle_peerconnection_java.jar |
> grep
> > RendererCommon
> >    830 Mon Jan 16 09:39:56 CET 2017 org/webrtc/RendererCommon$1.class
> >   1994 Mon Jan 16 09:39:56 CET 2017
> > org/webrtc/RendererCommon$VideoLayoutMeasure.class
> >   1659 Mon Jan 16 09:39:56 CET 2017
> org/webrtc/RendererCommon$YuvUploader.class
> >    311 Mon Jan 16 09:39:56 CET 2017 org/webrtc/RendererCommon$GlDrawer.class
> >   1199 Mon Jan 16 09:39:56 CET 2017
> org/webrtc/RendererCommon$ScalingType.class
> >   3326 Mon Jan 16 09:39:56 CET 2017 org/webrtc/RendererCommon.class
> >    295 Mon Jan 16 09:39:56 CET 2017
> > org/webrtc/RendererCommon$RendererEvents.class
> > 
> > New target (dist_jar):
> > $ jar tvf
out/android/obj/webrtc/sdk/android/libjingle_peerconnection_java.jar
> |
> > grep RendererCommon
> >    480 Mon Jan 01 00:00:00 CET 2001
> > org/webrtc/RendererCommon$VideoLayoutMeasure.class
> >    246 Mon Jan 01 00:00:00 CET 2001
> org/webrtc/RendererCommon$YuvUploader.class
> >    268 Mon Jan 01 00:00:00 CET 2001 org/webrtc/RendererCommon$GlDrawer.class
> >    514 Mon Jan 01 00:00:00 CET 2001
> org/webrtc/RendererCommon$ScalingType.class
> >    764 Mon Jan 01 00:00:00 CET 2001 org/webrtc/RendererCommon.class
> >    252 Mon Jan 01 00:00:00 CET 2001
> > org/webrtc/RendererCommon$RendererEvents.class
> 
> Doh! One of the ifs was backwards and it was using .interface.jars. The sizes
&
> missing files should be fixed now.
> 
> As for timestamps, we always intentionally null them out in order to achieve
> hermetic builds (building the same thing on two different machines should
> produce the same byte-for-byte artifact). git doesn't maintain mtimes, so I
> doubt you'd want to?

Yeah, the sizes and the missing files reported in the previous comment are now
fixed:

Old target:
$ jar tvf
./out/android/lib.java/webrtc/sdk/android/libjingle_peerconnection_java.jar |
grep RendererComm
   830 Tue Jan 17 08:45:58 CET 2017 org/webrtc/RendererCommon$1.class
  1994 Tue Jan 17 08:45:58 CET 2017
org/webrtc/RendererCommon$VideoLayoutMeasure.class
  1659 Tue Jan 17 08:45:58 CET 2017 org/webrtc/RendererCommon$YuvUploader.class
   311 Tue Jan 17 08:45:58 CET 2017 org/webrtc/RendererCommon$GlDrawer.class
  1199 Tue Jan 17 08:46:00 CET 2017 org/webrtc/RendererCommon$ScalingType.class
  3326 Tue Jan 17 08:46:00 CET 2017 org/webrtc/RendererCommon.class
   295 Tue Jan 17 08:46:00 CET 2017
org/webrtc/RendererCommon$RendererEvents.class

New target:
$ jar tvf ./out/android/obj/webrtc/sdk/android/libjingle_peerconnection_java.jar
| grep RendererCommon
   830 Mon Jan 01 00:00:00 CET 2001 org/webrtc/RendererCommon$1.class
  1994 Mon Jan 01 00:00:00 CET 2001
org/webrtc/RendererCommon$VideoLayoutMeasure.class
  1659 Mon Jan 01 00:00:00 CET 2001 org/webrtc/RendererCommon$YuvUploader.class
   311 Mon Jan 01 00:00:00 CET 2001 org/webrtc/RendererCommon$GlDrawer.class
  1199 Mon Jan 01 00:00:00 CET 2001 org/webrtc/RendererCommon$ScalingType.class
  3326 Mon Jan 01 00:00:00 CET 2001 org/webrtc/RendererCommon.class
   295 Mon Jan 01 00:00:00 CET 2001
org/webrtc/RendererCommon$RendererEvents.class

Ok, I agree on the timestamps. It makes lots of sense.

lgtm for me.

[sakal-chromium, 2017-01-17 10:35:41]

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

[sakal-chromium, 2017-01-17 10:35:41]

I tested this CL and it seems to work. lgtm

[agrieve, 2017-01-17 14:33:22]

On 2017/01/17 10:35:41, sakal-chromium wrote:
> I tested this CL and it seems to work. lgtm

Great! Thanks for your patience on this.

[agrieve, 2017-01-17 14:33:25]

The CQ bit was checked by agrieve@chromium.org

[commit-bot: I haz the power, 2017-01-17 14:33:38]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2017-01-17 15:33:27]

CQ is committing da patch.

Bot data: {"patchset_id": 40001, "attempt_start_ts": 1484663605894930,
"parent_rev": "899e87225508ac196d057b8cff1d9ff06d33d1bc", "commit_rev":
"77ad9ee9d1c3ca1ef984900a31cb206bf885b936"}

[commit-bot: I haz the power, 2017-01-17 15:34:01]

Description was changed from

==========
android: Create a GN template for create_dist_jar.py

BUG=648244
==========

to

==========
android: Create a GN template for create_dist_jar.py

BUG=648244

Review-Url: https://codereview.chromium.org/2623243002
Cr-Commit-Position: refs/heads/master@{#444054}
Committed:
https://chromium.googlesource.com/chromium/src/+/77ad9ee9d1c3ca1ef984900a31cb...
==========

[commit-bot: I haz the power, 2017-01-17 15:34:02]

Committed patchset #3 (id:40001) as
https://chromium.googlesource.com/chromium/src/+/77ad9ee9d1c3ca1ef984900a31cb...