[Third Party] Add Robolectric 3.0 build files.

[Third Party] Add Robolectric 3.0 build files.

BUG=522043
Committed: https://crrev.com/1153087482a5a26e662dc4afce662960cae8b15e
Cr-Commit-Position: refs/heads/master@{#407620}

[mikecase (-- gone --), 2016-06-09 18:51:29]

mikecase@chromium.org changed reviewers:
+ jbudorick@chromium.org, yolandyan@chromium.org

[mikecase (-- gone --), 2016-06-09 18:51:29]

Adding you guys in case you have any input. Only really difficult thing to build
is the shadows-core (most interesting part of this CL at least)

This CL will need all the third_party deps CL to land, and
https://codereview.chromium.org/2044223005, and will require me to land a pretty
big CL to change some of the Robolectric code.

[mikecase (-- gone --), 2016-06-09 18:52:36]

On 2016/06/09 at 18:51:29, mikecase wrote:
> Adding you guys in case you have any input. Only really difficult thing to
build is the shadows-core (most interesting part of this CL at least)
> 
> This CL will need all the third_party deps CL to land, and
https://codereview.chromium.org/2044223005, and will require me to land a pretty
big CL to change some of the Robolectric code.

There are some other built-in shadows in Robolectric shadows-multidex,
shadows-httpclient, and some other things that might be useful. Will add support
for these things after everything else lands.

[mikecase (-- gone --), 2016-06-28 23:15:20]

Added sha1 files for the jars specific to Robolectric that we need to pull in.
These are currently checked into the Robolectric repo with the prebuilt jars.
When we switch over to building Robolectric from source we are going to stop
DEPS'ing in this repo. So now we need to download them from GS.

[jbudorick, 2016-06-29 09:49:09]

lgtm

https://codereview.chromium.org/2051913002/diff/40001/third_party/robolectric...
File third_party/robolectric/BUILD.gn (right):

https://codereview.chromium.org/2051913002/diff/40001/third_party/robolectric...
third_party/robolectric/BUILD.gn:372: script =
"$root_build_dir/bin/robolectric_template_processor"
This is both neat and kind of crazy.

https://codereview.chromium.org/2051913002/diff/40001/third_party/robolectric...
third_party/robolectric/BUILD.gn:654:
"robolectric/robolectric-shadows/shadows-multidex/src/main/java/org/robolectric/shadows/multidex/ShadowMultiDex.java",
We can finally get rid of
https://chromium.googlesource.com/chromium/src/+/master/base/test/android/jun...
!

[mikecase (-- gone --), 2016-07-21 18:16:15]

Updated. PTAL

robolectric_3_java target builds! However, not used by anything at the moment.

Updates include...

==== Adding some robolectrics to whitelist of improperly named targets. ====

These targets have to have these names (for now) since Robolectric looks for
these specifically named jars in the classpath. This is on my list of things to
fix.

==== No longer pull in anything from the chromium/third_party/robolectric.git
repo ====

This repo contains the old Robolectric 2.4 jars. I want to keep things simple,
so I (for the time being) added the old jars to the gs://chromium-robolectric
bucket (this is where some new Robolectric 3.0 jars are downloaded from).
Nothing will change about these jars except where they are downloaded from. And
I will delete them permanently once all the tests have been updated.

==== Minor build file changes ====

Added testonly to a couple places where needed. Changed guava and hamcrest jars
from android_java_prebuilts to java_prebuilts. I think this is fine to do.

[mikecase (-- gone --), 2016-07-21 18:17:19]

On 2016/07/21 at 18:16:15, mikecase wrote:
> Updated. PTAL
> 
> robolectric_3_java target builds! However, not used by anything at the moment.
> 
> Updates include...
> 
> ==== Adding some robolectrics to whitelist of improperly named targets. ====
> 
> These targets have to have these names (for now) since Robolectric looks for
these specifically named jars in the classpath. This is on my list of things to
fix.
> 
> ==== No longer pull in anything from the chromium/third_party/robolectric.git
repo ====
> 
> This repo contains the old Robolectric 2.4 jars. I want to keep things simple,
so I (for the time being) added the old jars to the gs://chromium-robolectric
bucket (this is where some new Robolectric 3.0 jars are downloaded from).
Nothing will change about these jars except where they are downloaded from. And
I will delete them permanently once all the tests have been updated.
> 
> ==== Minor build file changes ====
> 
> Added testonly to a couple places where needed. Changed guava and hamcrest
jars from android_java_prebuilts to java_prebuilts. I think this is fine to do.

Also, didn't mention, now pull in my Robolectric commit via DEPS.

[mikecase (-- gone --), 2016-07-25 16:43:39]

ping, PTAL and tell me if this still looks good. Previous comment lists what
changed since 1gtm'ed it.

[jbudorick, 2016-07-25 17:21:56]

https://codereview.chromium.org/2051913002/diff/60001/third_party/robolectric...
File third_party/robolectric/BUILD.gn (right):

https://codereview.chromium.org/2051913002/diff/60001/third_party/robolectric...
third_party/robolectric/BUILD.gn:314: processed_template_files = [
Are these all a function of entries in template_files? Can we avoid listing them
twice if so?

https://codereview.chromium.org/2051913002/diff/60001/third_party/robolectric...
File third_party/robolectric/README.chromium (right):

https://codereview.chromium.org/2051913002/diff/60001/third_party/robolectric...
third_party/robolectric/README.chromium:9: Local Modifications: None
This should probably list the changes you made to remove the XML library.

[mikecase (-- gone --), 2016-07-25 17:50:24]

https://codereview.chromium.org/2051913002/diff/60001/third_party/robolectric...
File third_party/robolectric/BUILD.gn (right):

https://codereview.chromium.org/2051913002/diff/60001/third_party/robolectric...
third_party/robolectric/BUILD.gn:314: processed_template_files = [
On 2016/07/25 at 17:21:55, jbudorick wrote:
> Are these all a function of entries in template_files? Can we avoid listing
them twice if so?

I don't know how to do that in GN :/

https://codereview.chromium.org/2051913002/diff/60001/third_party/robolectric...
File third_party/robolectric/README.chromium (right):

https://codereview.chromium.org/2051913002/diff/60001/third_party/robolectric...
third_party/robolectric/README.chromium:9: Local Modifications: None
On 2016/07/25 at 17:21:56, jbudorick wrote:
> This should probably list the changes you made to remove the XML library.

Done

[jbudorick, 2016-07-25 18:07:40]

https://codereview.chromium.org/2051913002/diff/60001/third_party/robolectric...
File third_party/robolectric/BUILD.gn (right):

https://codereview.chromium.org/2051913002/diff/60001/third_party/robolectric...
third_party/robolectric/BUILD.gn:314: processed_template_files = [
On 2016/07/25 17:50:24, mikecase wrote:
> On 2016/07/25 at 17:21:55, jbudorick wrote:
> > Are these all a function of entries in template_files? Can we avoid listing
> them twice if so?
> 
> I don't know how to do that in GN :/

Maybe something like this:

processed_template_files = []
foreach(template_files, t) {
  gen_dir = get_path_info(t, "gen_dir")
  name = get_path_info(, "name")
  processed_template_files += ["$gen_dir/$name.java"]
}

https://chromium.googlesource.com/chromium/src/+/master/tools/gn/docs/referen...
https://chromium.googlesource.com/chromium/src/+/master/tools/gn/docs/referen...

[mikecase (-- gone --), 2016-07-25 19:55:58]

https://codereview.chromium.org/2051913002/diff/100001/third_party/robolectri...
File third_party/robolectric/BUILD.gn (right):

https://codereview.chromium.org/2051913002/diff/100001/third_party/robolectri...
third_party/robolectric/BUILD.gn:319: }
Done! Didnt know about this get_path_info magic.

[jbudorick, 2016-07-25 20:01:05]

lgtm w/ q

https://codereview.chromium.org/2051913002/diff/100001/third_party/robolectri...
File third_party/robolectric/BUILD.gn (right):

https://codereview.chromium.org/2051913002/diff/100001/third_party/robolectri...
third_party/robolectric/BUILD.gn:319: }
On 2016/07/25 19:55:58, mikecase wrote:
> Done! Didnt know about this get_path_info magic.

Does this get turned into the right name? Does "name" only strip the last
extension (leaving "Foo.java")?

[mikecase (-- gone --), 2016-07-25 20:03:58]

On 2016/07/25 at 20:01:05, jbudorick wrote:
> lgtm w/ q
> 
>
https://codereview.chromium.org/2051913002/diff/100001/third_party/robolectri...
> File third_party/robolectric/BUILD.gn (right):
> 
>
https://codereview.chromium.org/2051913002/diff/100001/third_party/robolectri...
> third_party/robolectric/BUILD.gn:319: }
> On 2016/07/25 19:55:58, mikecase wrote:
> > Done! Didnt know about this get_path_info magic.
> 
> Does this get turned into the right name? Does "name" only strip the last
extension (leaving "Foo.java")?

Yes. I initially had it the way you put in your comment and it was like "no file
foo.java.java" found.
So it seems to only strip the last extension.

[mikecase (-- gone --), 2016-07-25 20:08:10]

The CQ bit was checked by mikecase@chromium.org

[commit-bot: I haz the power, 2016-07-25 20:08:53]

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, 2016-07-25 20:53:02]

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

[commit-bot: I haz the power, 2016-07-25 20:53:02]

Try jobs failed on following builders:
  android_arm64_dbg_recipe on master.tryserver.chromium.android (JOB_FAILED,
https://build.chromium.org/p/tryserver.chromium.android/builders/android_arm6...)

[mikecase (-- gone --), 2016-07-25 21:09:48]

The CQ bit was checked by mikecase@chromium.org

[mikecase (-- gone --), 2016-07-25 21:09:49]

The patchset sent to the CQ was uploaded after l-g-t-m from
jbudorick@chromium.org
Link to the patchset: https://codereview.chromium.org/2051913002/#ps120001
(title: "Make some variables final.")

[commit-bot: I haz the power, 2016-07-25 21:10:32]

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, 2016-07-25 22:54:40]

Committed patchset #7 (id:120001)

[commit-bot: I haz the power, 2016-07-25 22:55:57]

Description was changed from

==========
[Third Party] Add Robolectric 3.0 build files.

BUG=522043
==========

to

==========
[Third Party] Add Robolectric 3.0 build files.

BUG=522043

Committed: https://crrev.com/1153087482a5a26e662dc4afce662960cae8b15e
Cr-Commit-Position: refs/heads/master@{#407620}
==========

[commit-bot: I haz the power, 2016-07-25 22:55:58]

Patchset 7 (id:??) landed as
https://crrev.com/1153087482a5a26e662dc4afce662960cae8b15e
Cr-Commit-Position: refs/heads/master@{#407620}

[ehmaldonado_chromium, 2016-07-27 08:13:32]

I think this CL causes WebRTC to break when we roll the chromium revision
(https://codereview.webrtc.org/2186543002/), making our android gn bots to fail.

android_gn_rel 
https://build.chromium.org/p/tryserver.webrtc/builders/android_gn_rel/builds/...

ERROR at //build/config/android/internal_rules.gni:139:23: Can't load input
file.
            deps += [ "${_target_label}__build_config" ]
                      ^-------------------------------
Unable to load:
  /b/build/slave/android_gn/build/src/third_party/ow2_asm/BUILD.gn
I also checked in the secondary tree for:
 
/b/build/slave/android_gn/build/src/build/secondary/third_party/ow2_asm/BUILD.gn


android_gn_dbg
https://build.chromium.org/p/tryserver.webrtc/builders/android_gn_dbg/builds/...

ERROR at //build/config/android/internal_rules.gni:139:23: Can't load input
file.
            deps += [ "${_target_label}__build_config" ]
                      ^-------------------------------
Unable to load:
  /b/build/slave/android_gn/build/src/third_party/bouncycastle/BUILD.gn
I also checked in the secondary tree for:
 
/b/build/slave/android_gn/build/src/build/secondary/third_party/bouncycastle/BUILD.gn


These two are dependencies of Robolectric 3.0, which is added by this CL.
Could be that we need additional config in WebRTC.

[mikecase (-- gone --), 2016-07-27 15:19:01]

On 2016/07/27 at 08:13:32, ehmaldonado wrote:
> I think this CL causes WebRTC to break when we roll the chromium revision
(https://codereview.webrtc.org/2186543002/), making our android gn bots to fail.
> 
> android_gn_rel 
>
https://build.chromium.org/p/tryserver.webrtc/builders/android_gn_rel/builds/...
> 
> ERROR at //build/config/android/internal_rules.gni:139:23: Can't load input
file.
>             deps += [ "${_target_label}__build_config" ]
>                       ^-------------------------------
> Unable to load:
>   /b/build/slave/android_gn/build/src/third_party/ow2_asm/BUILD.gn
> I also checked in the secondary tree for:
>  
/b/build/slave/android_gn/build/src/build/secondary/third_party/ow2_asm/BUILD.gn
> 
> 
> android_gn_dbg
>
https://build.chromium.org/p/tryserver.webrtc/builders/android_gn_dbg/builds/...
> 
> ERROR at //build/config/android/internal_rules.gni:139:23: Can't load input
file.
>             deps += [ "${_target_label}__build_config" ]
>                       ^-------------------------------
> Unable to load:
>   /b/build/slave/android_gn/build/src/third_party/bouncycastle/BUILD.gn
> I also checked in the secondary tree for:
>  
/b/build/slave/android_gn/build/src/build/secondary/third_party/bouncycastle/BUILD.gn
> 
> 
> These two are dependencies of Robolectric 3.0, which is added by this CL.
> Could be that we need additional config in WebRTC.

Im not too familiar with how the WebRTC bots are configured. I looked at the bot
and it doesn't seem to have all of the third_party Chromium directories. Not
sure what is up, but ill look into fixing this.

[mikecase (-- gone --), 2016-07-27 16:29:35]

On 2016/07/27 at 15:19:01, mikecase wrote:
> On 2016/07/27 at 08:13:32, ehmaldonado wrote:
> > I think this CL causes WebRTC to break when we roll the chromium revision
(https://codereview.webrtc.org/2186543002/), making our android gn bots to fail.
> > 
> > android_gn_rel 
> >
https://build.chromium.org/p/tryserver.webrtc/builders/android_gn_rel/builds/...
> > 
> > ERROR at //build/config/android/internal_rules.gni:139:23: Can't load input
file.
> >             deps += [ "${_target_label}__build_config" ]
> >                       ^-------------------------------
> > Unable to load:
> >   /b/build/slave/android_gn/build/src/third_party/ow2_asm/BUILD.gn
> > I also checked in the secondary tree for:
> >  
/b/build/slave/android_gn/build/src/build/secondary/third_party/ow2_asm/BUILD.gn
> > 
> > 
> > android_gn_dbg
> >
https://build.chromium.org/p/tryserver.webrtc/builders/android_gn_dbg/builds/...
> > 
> > ERROR at //build/config/android/internal_rules.gni:139:23: Can't load input
file.
> >             deps += [ "${_target_label}__build_config" ]
> >                       ^-------------------------------
> > Unable to load:
> >   /b/build/slave/android_gn/build/src/third_party/bouncycastle/BUILD.gn
> > I also checked in the secondary tree for:
> >  
/b/build/slave/android_gn/build/src/build/secondary/third_party/bouncycastle/BUILD.gn
> > 
> > 
> > These two are dependencies of Robolectric 3.0, which is added by this CL.
> > Could be that we need additional config in WebRTC.
> 
> Im not too familiar with how the WebRTC bots are configured. I looked at the
bot and it doesn't seem to have all of the third_party Chromium directories. Not
sure what is up, but ill look into fixing this.

I have a CL here for my attempt to fix it.
https://codereview.chromium.org/2183923005
I'm not familiar with WebRTC or the setup_links thing, but this feels like the
right fix.

[phoglund_chromium, 2016-07-28 10:27:16]

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

[phoglund_chromium, 2016-07-28 10:27:17]

Ok, we've fixed one problem by setting up the links, on to the next one:

1963/3627] ACTION
//third_party/robolectric:shadows-core-3.0__compile_java__javac(//build/toolchain/android:arm)
FAILED: gen/third_party/robolectric/shadows-core-3.0__compile_java__javac.d
gen/third_party/robolectric/shadows-core-3.0__compile_java.javac.jar
gen/third_party/robolectric/shadows-core-3.0__compile_java.javac.jar.md5.stamp 
python ../../build/android/gyp/javac.py
--depfile=gen/third_party/robolectric/shadows-core-3.0__compile_java__javac.d
--jar-path=gen/third_party/robolectric/shadows-core-3.0__compile_java.javac.jar
--java-srcjars=\[\]
--java-srcjars=@FileArg\(gen/third_party/robolectric/shadows-core-3.0.build_config:javac:srcjars\)
--classpath=@FileArg\(gen/third_party/robolectric/shadows-core-3.0.build_config:javac:classpath\)
--provider-configuration=../../third_party/robolectric/robolectric/robolectric-shadows/shadows-core/src/main/resources/META-INF/services/org.robolectric.ShadowsAdapter
--provider-configuration=../../third_party/robolectric/robolectric/robolectric-processor/src/test/resources/META-INF/services/org.robolectric.internal.ShadowProvider
--processor=org.robolectric.annotation.processing.RobolectricProcessor
--processor-arg=org.robolectric.annotation.processing.shadowPackage=org.robolectric
--additional-jar-file=/b/build/slave/android_gn/build/src/third_party/sqlite4java/lib/linux-x86_64/libsqlite4java.so:linux-x86_64/libsqlite4java.so
--additional-jar-file=/b/build/slave/android_gn/build/src/third_party/sqlite4java/lib/linux-x86/libsqlite4java.so:linux-x86/libsqlite4java.so
--additional-jar-file=/b/build/slave/android_gn/build/src/third_party/sqlite4java/lib/mac-x86_64/libsqlite4java.jnilib:mac-x86_64/libsqlite4java.jnilib
--additional-jar-file=/b/build/slave/android_gn/build/src/third_party/sqlite4java/lib/windows-x86/sqlite4java.dll:windows-x86/sqlite4java.dll
--additional-jar-file=/b/build/slave/android_gn/build/src/third_party/sqlite4java/lib/windows-x86_64/sqlite4java.dll:windows-x86_64/sqlite4java.dll
@gen/third_party/robolectric/shadows-core-3.0.sources
Traceback (most recent call last):
  File "../../build/android/gyp/javac.py", line 441, in <module>
    sys.exit(main(sys.argv[1:]))
  File "../../build/android/gyp/javac.py", line 437, in main
    pass_changes=True)
  File
"/b/build/slave/android_gn/build/src/chromium/src/build/android/gyp/util/build_utils.py",
line 536, in CallAndWriteDepfileIfStale
    pass_changes=True)
  File
"/b/build/slave/android_gn/build/src/chromium/src/build/android/gyp/util/md5_check.py",
line 63, in CallAndRecordIfStale
    new_metadata.AddFile(path, _Md5ForPath(path))
  File
"/b/build/slave/android_gn/build/src/chromium/src/build/android/gyp/util/md5_check.py",
line 381, in _Md5ForPath
    _UpdateMd5ForFile(md5, path)
  File
"/b/build/slave/android_gn/build/src/chromium/src/build/android/gyp/util/md5_check.py",
line 362, in _UpdateMd5ForFile
    with open(path, 'rb') as infile:
IOError: [Errno 2] No such file or directory:
'gen/third_party/robolectric/robolectric/robolectric-shadows/shadows-core/src/main/resources/android/app/RobolectricActivityManager.java'

Are we missing some step for generating robolectric code?

[phoglund_chromium, 2016-07-28 10:31:56]

We still use gyp for WebRTC so you probably need to port your gn changes to gyp
here.

[jbudorick, 2016-07-28 12:55:13]

On 2016/07/28 10:31:56, phoglund_chrome wrote:
> We still use gyp for WebRTC so you probably need to port your gn changes to
gyp
> here.

You do?!

[phoglund_chromium, 2016-07-28 13:01:41]

On 2016/07/28 12:55:13, jbudorick wrote:
> On 2016/07/28 10:31:56, phoglund_chrome wrote:
> > We still use gyp for WebRTC so you probably need to port your gn changes to
> gyp
> > here.
> 
> You do?!

Yes! Gaze upon for instance
https://build.chromium.org/p/client.webrtc/builders/Android64%20Builder%20%28...
and tremble in horror, for it invokes gyp!

We're obviously trying to switch, and I think this effort will accelerate when
kjellander@ gets back from vacation next week. Gyp is dying pretty fast.

[jbudorick, 2016-07-28 13:03:59]

On 2016/07/28 13:01:41, phoglund_chrome wrote:
> On 2016/07/28 12:55:13, jbudorick wrote:
> > On 2016/07/28 10:31:56, phoglund_chrome wrote:
> > > We still use gyp for WebRTC so you probably need to port your gn changes
to
> > gyp
> > > here.
> > 
> > You do?!
> 
> Yes! Gaze upon for instance
>
https://build.chromium.org/p/client.webrtc/builders/Android64%20Builder%20%28...
> and tremble in horror, for it invokes gyp!
> 

Tremble in horror, indeed. I'm surprised that the gyp build (or the portion of
it that webrtc uses) hasn't decayed yet.

> We're obviously trying to switch, and I think this effort will accelerate when
> kjellander@ gets back from vacation next week. Gyp is dying pretty fast.

[jbudorick, 2016-07-28 16:17:12]

On 2016/07/28 13:03:59, jbudorick wrote:
> On 2016/07/28 13:01:41, phoglund_chrome wrote:
> > On 2016/07/28 12:55:13, jbudorick wrote:
> > > On 2016/07/28 10:31:56, phoglund_chrome wrote:
> > > > We still use gyp for WebRTC so you probably need to port your gn changes
> to
> > > gyp
> > > > here.
> > > 
> > > You do?!
> > 
> > Yes! Gaze upon for instance
> >
>
https://build.chromium.org/p/client.webrtc/builders/Android64%20Builder%20%28...
> > and tremble in horror, for it invokes gyp!
> > 
> 
> Tremble in horror, indeed. I'm surprised that the gyp build (or the portion of
> it that webrtc uses) hasn't decayed yet.

On a somewhat more serious note: I'd strongly prefer that we not make further
gyp changes in chromium. I'll be happy to work w/ kjellander to get WebRTC
switched over to gn upon his return.

> 
> > We're obviously trying to switch, and I think this effort will accelerate
when
> > kjellander@ gets back from vacation next week. Gyp is dying pretty fast.

[phoglund_chromium, 2016-07-28 16:22:32]

> On a somewhat more serious note: I'd strongly prefer that we not make further
> gyp changes in chromium. I'll be happy to work w/ kjellander to get WebRTC
> switched over to gn upon his return.
> 

Sure. He's back on Monday. Meanwhile, I've tasked people to grind out as many
of the missing GN targets as possible so we can switch.

Any chance we can get some throwaway gyp code in until then? Feels like it's
an unknown amount of work until the switch is complete, and we can't keep the
project healthy if we can't roll chromium into WebRTC.

[phoglund_chromium, 2016-07-29 13:05:13]

On 2016/07/28 16:22:32, phoglund_chrome wrote:
> > On a somewhat more serious note: I'd strongly prefer that we not make
further
> > gyp changes in chromium. I'll be happy to work w/ kjellander to get WebRTC
> > switched over to gn upon his return.
> > 
> 
> Sure. He's back on Monday. Meanwhile, I've tasked people to grind out as many
> of the missing GN targets as possible so we can switch.
> 
> Any chance we can get some throwaway gyp code in until then? Feels like it's
> an unknown amount of work until the switch is complete, and we can't keep the
> project healthy if we can't roll chromium into WebRTC.

I'm going on vacation now so talk to kjellander@ about this. It's really
unacceptable
that we're staying in this broken state, so please consider adding at least some
workaround while we scramble to get on GN. If you can't do that, maybe we should
disable our Android GN bots for now (but that would be bad timing since we'll
work
a lot on GN now).

[jbudorick, 2016-07-29 14:27:30]

On 2016/07/29 13:05:13, phoglund_chrome_OOO_sep_05 wrote:
> On 2016/07/28 16:22:32, phoglund_chrome wrote:
> > > On a somewhat more serious note: I'd strongly prefer that we not make
> further
> > > gyp changes in chromium. I'll be happy to work w/ kjellander to get WebRTC
> > > switched over to gn upon his return.
> > > 
> > 
> > Sure. He's back on Monday. Meanwhile, I've tasked people to grind out as
many
> > of the missing GN targets as possible so we can switch.
> > 
> > Any chance we can get some throwaway gyp code in until then? Feels like it's
> > an unknown amount of work until the switch is complete, and we can't keep
the
> > project healthy if we can't roll chromium into WebRTC.
> 
> I'm going on vacation now so talk to kjellander@ about this. It's really
> unacceptable
> that we're staying in this broken state, so please consider adding at least
some
> workaround while we scramble to get on GN. If you can't do that, maybe we
should
> disable our Android GN bots for now (but that would be bad timing since we'll
> work
> a lot on GN now).

I think a gyp workaround is out of the question. We deprecated chromium's
Android gyp build almost three months ago:
https://groups.google.com/a/chromium.org/d/msg/chromium-dev/9FLgT0LPfQ8/Ze94q...

I'll talk to kjellander about this next week.