Fix a bunch of generated file build flakes in //extensions

Fix a bunch of generated file build flakes in //extensions

Several files in //extensions could randomly fail to build due to
missing dependencies on header generator targets, mostly mojo
and grit. Add the dependencies so builds are not flaky.

BUG=655123
Committed: https://crrev.com/dc44a53ed3244a4a68cb1fd4ccfaacf1e97e5d23
Cr-Commit-Position: refs/heads/master@{#429543}

[tsniatowski, 2016-10-26 11:03:47]

Description was changed from

==========
Fix a bunch of generated file build flakes in //extensions

Several files in //extensions could randomly fail to build due to
missing dependencies on header generator targets, mostly mojo
and grit. Add the dependencies so builds are not flaky.

BUG=655123
==========

to

==========
Fix a bunch of generated file build flakes in //extensions

Several files in //extensions could randomly fail to build due to
missing dependencies on header generator targets, mostly mojo
and grit. Add the dependencies so builds are not flaky.

BUG=655123
==========

[tsniatowski, 2016-10-26 11:03:47]

tsniatowski@opera.com changed reviewers:
+ lazyboy@chromium.org, wjmaclean@chromium.org

[tsniatowski, 2016-10-26 11:06:10]

On 2016/10/26 11:03:47, tsniatowski wrote:
> mailto:tsniatowski@opera.com changed reviewers:
> + mailto:lazyboy@chromium.org, mailto:wjmaclean@chromium.org

PTAL (build flakes aren't fun).

Repro steps: build chrome and wait for a flake. That's slow, so simulate the
unfortunate ordering by:

gn gen out && gn clean out && ninja -C out FLAKY_FILE

where FLAKY_FILE is one of:
obj/extensions/browser/api/api/device_permissions_manager.o
obj/extensions/browser/api/api/device_permissions_prompt.o
obj/extensions/browser/api/bluetooth_low_energy/bluetooth_low_energy/bluetooth_api_advertisement.o
obj/extensions/browser/api/bluetooth_low_energy/bluetooth_low_energy/bluetooth_low_energy_api.o
obj/extensions/browser/api/bluetooth_low_energy/bluetooth_low_energy/bluetooth_low_energy_event_router.o
obj/extensions/browser/api/bluetooth_low_energy/bluetooth_low_energy/utils.o
obj/extensions/browser/api/serial/serial/serial_api.o
obj/extensions/browser/api/serial/serial/serial_connection.o
obj/extensions/browser/api/serial/serial/serial_event_dispatcher.o
obj/extensions/browser/api/web_request/web_request/web_request_api.o
obj/extensions/browser/api/web_request/web_request/web_request_api_helpers.o
obj/extensions/browser/app_window/app_window/app_window.o
obj/extensions/browser/guest_view/app_view/app_view/app_view_guest.o
obj/extensions/browser/guest_view/extension_options/extension_options/extension_options_guest.o
obj/extensions/browser/guest_view/extension_view/extension_view/extension_view_guest.o
obj/extensions/browser/guest_view/mime_handler_view/mime_handler_view/mime_handler_view_guest.o
obj/extensions/browser/guest_view/web_view/web_view/web_view_guest.o
obj/extensions/renderer/renderer/display_source_session.o
obj/extensions/renderer/renderer/dispatcher.o
obj/extensions/renderer/renderer/display_source_custom_bindings.o
obj/extensions/common/api/api_registration/generated_api_registration.o

[wjmaclean_google.com, 2016-10-26 12:22:59]

wjmaclean@google.com changed reviewers:
+ wjmaclean@google.com

[wjmaclean_google.com, 2016-10-26 12:22:59]

lgtm

[tsniatowski, 2016-10-26 13:20:55]

The CQ bit was checked by tsniatowski@opera.com

[commit-bot: I haz the power, 2016-10-26 13:21:13]

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, 2016-10-26 13:21:14]

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

[commit-bot: I haz the power, 2016-10-26 13:21:15]

No L-G-T-M from a valid reviewer yet. 
CQ run can only be started by full committers or once the patch has
received an L-G-T-M from a full committer.
Even if an L-G-T-M may have been provided, it was from a non-committer,
_not_ a full super star committer.
See http://www.chromium.org/getting-involved/become-a-committer
Note that this has nothing to do with OWNERS files.

[tsniatowski, 2016-10-26 14:07:44]

On 2016/10/26 13:21:15, commit-bot: I haz the power wrote:
> No L-G-T-M from a valid reviewer yet. 
> CQ run can only be started by full committers or once the patch has
> received an L-G-T-M from a full committer.
> Even if an L-G-T-M may have been provided, it was from a non-committer,
> _not_ a full super star committer.
> See http://www.chromium.org/getting-involved/become-a-committer
> Note that this has nothing to do with OWNERS files.

I'm ever so slightly confused, but no matter. Anyway; @lazyboy, PTAL?

[wjmaclean, 2016-10-26 14:23:41]

On 2016/10/26 14:07:44, tsniatowski wrote:
> On 2016/10/26 13:21:15, commit-bot: I haz the power wrote:
> > No L-G-T-M from a valid reviewer yet. 
> > CQ run can only be started by full committers or once the patch has
> > received an L-G-T-M from a full committer.
> > Even if an L-G-T-M may have been provided, it was from a non-committer,
> > _not_ a full super star committer.
> > See http://www.chromium.org/getting-involved/become-a-committer
> > Note that this has nothing to do with OWNERS files.
> 
> I'm ever so slightly confused, but no matter. Anyway; @lazyboy, PTAL?

I suspect this *is* an OWNERS file issue ... so far as I know, I've been a full
committer for years now ...

[tsniatowski, 2016-10-26 14:29:56]

On 2016/10/26 14:23:41, wjmaclean wrote:
> On 2016/10/26 14:07:44, tsniatowski wrote:
> > On 2016/10/26 13:21:15, commit-bot: I haz the power wrote:
> > > No L-G-T-M from a valid reviewer yet. 
> > > CQ run can only be started by full committers or once the patch has
> > > received an L-G-T-M from a full committer.
> > > Even if an L-G-T-M may have been provided, it was from a non-committer,
> > > _not_ a full super star committer.
> > > See http://www.chromium.org/getting-involved/become-a-committer
> > > Note that this has nothing to do with OWNERS files.
> > 
> > I'm ever so slightly confused, but no matter. Anyway; @lazyboy, PTAL?
> 
> I suspect this *is* an OWNERS file issue ... so far as I know, I've been a
full
> committer for years now ...

I think so too. I was going for a reviewer who'd reviewed recently-ish here, but
didn't double-chek OWNERS scope. Then again, it may have also been a
wjmaclean<->wjmaclean1 issue.

[wjmaclean, 2016-10-26 14:30:34]

On 2016/10/26 14:23:41, wjmaclean wrote:
> On 2016/10/26 14:07:44, tsniatowski wrote:
> > On 2016/10/26 13:21:15, commit-bot: I haz the power wrote:
> > > No L-G-T-M from a valid reviewer yet. 
> > > CQ run can only be started by full committers or once the patch has
> > > received an L-G-T-M from a full committer.
> > > Even if an L-G-T-M may have been provided, it was from a non-committer,
> > > _not_ a full super star committer.
> > > See http://www.chromium.org/getting-involved/become-a-committer
> > > Note that this has nothing to do with OWNERS files.
> > 
> > I'm ever so slightly confused, but no matter. Anyway; @lazyboy, PTAL?
> 
> I suspect this *is* an OWNERS file issue ... so far as I know, I've been a
full
> committer for years now ...

LGTM

[wjmaclean, 2016-10-26 14:31:24]

On 2016/10/26 14:30:34, wjmaclean wrote:
> On 2016/10/26 14:23:41, wjmaclean wrote:
> > On 2016/10/26 14:07:44, tsniatowski wrote:
> > > On 2016/10/26 13:21:15, commit-bot: I haz the power wrote:
> > > > No L-G-T-M from a valid reviewer yet. 
> > > > CQ run can only be started by full committers or once the patch has
> > > > received an L-G-T-M from a full committer.
> > > > Even if an L-G-T-M may have been provided, it was from a non-committer,
> > > > _not_ a full super star committer.
> > > > See http://www.chromium.org/getting-involved/become-a-committer
> > > > Note that this has nothing to do with OWNERS files.
> > > 
> > > I'm ever so slightly confused, but no matter. Anyway; @lazyboy, PTAL?
> > 
> > I suspect this *is* an OWNERS file issue ... so far as I know, I've been a
> full
> > committer for years now ...
> 
> LGTM

That should work, but you still need owners approval for the non-GuestView
stuff.

[lazyboy, 2016-10-26 17:22:22]

lazyboy@chromium.org changed reviewers:
+ rdevlin.cronin@chromium.org

[lazyboy, 2016-10-26 17:22:23]

Over to Devlin for review.

https://codereview.chromium.org/2452943003/diff/1/extensions/common/api/BUILD.gn
File extensions/common/api/BUILD.gn (right):

https://codereview.chromium.org/2452943003/diff/1/extensions/common/api/BUILD...
extensions/common/api/BUILD.gn:115: "//extensions/browser/api",
Not sure if you want this, over to Devlin for reviewing.

[Devlin, 2016-10-26 19:30:27]

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

[Devlin, 2016-10-26 19:30:28]

Sorry to pass the buck farther.

Dirk, can you take a look at the comment in extensions/common/api/BUILD.gn?

https://codereview.chromium.org/2452943003/diff/1/extensions/browser/app_wind...
File extensions/browser/app_window/BUILD.gn (right):

https://codereview.chromium.org/2452943003/diff/1/extensions/browser/app_wind...
extensions/browser/app_window/BUILD.gn:32: "//extensions:extensions_resources",
could this just be extensions_browser_resources?

https://codereview.chromium.org/2452943003/diff/1/extensions/browser/guest_vi...
File extensions/browser/guest_view/web_view/BUILD.gn (right):

https://codereview.chromium.org/2452943003/diff/1/extensions/browser/guest_vi...
extensions/browser/guest_view/web_view/BUILD.gn:28:
"//components/guest_view/browser",
Do none of the other extensions/browser/guest_view files need this?  (More
surprised than anything else)

https://codereview.chromium.org/2452943003/diff/1/extensions/common/api/BUILD.gn
File extensions/common/api/BUILD.gn (right):

https://codereview.chromium.org/2452943003/diff/1/extensions/common/api/BUILD...
extensions/common/api/BUILD.gn:115: "//extensions/browser/api",
On 2016/10/26 17:22:23, lazyboy wrote:
> Not sure if you want this, over to Devlin for reviewing.

Hmm....  So I'm assuming we need this because the generated files #include all
the browser apis in order to register the functions.  Ideally, this would just
go in extensions/browser/api, since the output file also goes there, but then we
can't use the variables here (like schema_files) as easily.  Dirk might know if
there's an easy way for this to expose schema_files as a variable, which could
then be included from the extensions/browser/api BUILD.gn, and if that's a
reasonable approach to take here.

[Devlin, 2016-10-26 19:31:00]

https://codereview.chromium.org/2452943003/diff/1/extensions/common/api/BUILD.gn
File extensions/common/api/BUILD.gn (right):

https://codereview.chromium.org/2452943003/diff/1/extensions/common/api/BUILD...
extensions/common/api/BUILD.gn:115: "//extensions/browser/api",
On 2016/10/26 19:30:28, Devlin wrote:
> On 2016/10/26 17:22:23, lazyboy wrote:
> > Not sure if you want this, over to Devlin for reviewing.
> 
> Hmm....  So I'm assuming we need this because the generated files #include all
> the browser apis in order to register the functions.  Ideally, this would just
> go in extensions/browser/api, since the output file also goes there, but then
we
> can't use the variables here (like schema_files) as easily.  Dirk might know
if
> there's an easy way for this to expose schema_files as a variable, which could
> then be included from the extensions/browser/api BUILD.gn, and if that's a
> reasonable approach to take here.

(fwiw, I think this actually works, but it's a little messy to reference
"browser" from a common at all - even in a BUILD.gn file)

[tsniatowski, 2016-10-26 20:43:42]

https://codereview.chromium.org/2452943003/diff/1/extensions/browser/app_wind...
File extensions/browser/app_window/BUILD.gn (right):

https://codereview.chromium.org/2452943003/diff/1/extensions/browser/app_wind...
extensions/browser/app_window/BUILD.gn:32: "//extensions:extensions_resources",
On 2016/10/26 19:30:28, Devlin wrote:
> could this just be extensions_browser_resources?

Checked -- yes. I can change this if you prefer.

https://codereview.chromium.org/2452943003/diff/1/extensions/browser/guest_vi...
File extensions/browser/guest_view/web_view/BUILD.gn (right):

https://codereview.chromium.org/2452943003/diff/1/extensions/browser/guest_vi...
extensions/browser/guest_view/web_view/BUILD.gn:28:
"//components/guest_view/browser",
On 2016/10/26 19:30:28, Devlin wrote:
> Do none of the other extensions/browser/guest_view files need this?  (More
> surprised than anything else)

Apparently not from an "are there missing dependencies that can break the build"
perspective according to my test; from a gn check perspective -- maybe, I'm not
handling it.

Actually after a recheck this particular dep is not needed to fix the flakiness.
Keep it?

https://codereview.chromium.org/2452943003/diff/1/extensions/common/api/BUILD.gn
File extensions/common/api/BUILD.gn (right):

https://codereview.chromium.org/2452943003/diff/1/extensions/common/api/BUILD...
extensions/common/api/BUILD.gn:115: "//extensions/browser/api",
On 2016/10/26 19:31:00, Devlin wrote:
> On 2016/10/26 19:30:28, Devlin wrote:
> > On 2016/10/26 17:22:23, lazyboy wrote:
> > > Not sure if you want this, over to Devlin for reviewing.
> > 
> > Hmm....  So I'm assuming we need this because the generated files #include
all
> > the browser apis in order to register the functions.  Ideally, this would
just
> > go in extensions/browser/api, since the output file also goes there, but
then
> we
> > can't use the variables here (like schema_files) as easily.  Dirk might know
> if
> > there's an easy way for this to expose schema_files as a variable, which
could
> > then be included from the extensions/browser/api BUILD.gn, and if that's a
> > reasonable approach to take here.
> 
> (fwiw, I think this actually works, but it's a little messy to reference
> "browser" from a common at all - even in a BUILD.gn file)

Yes, what ends up happening is that the build of
obj/extensions/common/api/api_registration/generated_api_registration.o includes
gen/content/public/common/window_container_type.mojom.h and friends via one of
the api headers.

FWIW, the dependency / reference is already there, this CL only makes it
explicitly listed (so the build is not flaky). Seemed better than adding a
direct //content dep just because it happens to fix the flake.

[Dirk Pranke, 2016-10-26 21:49:21]

Let me know if I didn't answer (or misunderstood) the question.

https://codereview.chromium.org/2452943003/diff/1/extensions/common/api/BUILD.gn
File extensions/common/api/BUILD.gn (right):

https://codereview.chromium.org/2452943003/diff/1/extensions/common/api/BUILD...
extensions/common/api/BUILD.gn:115: "//extensions/browser/api",
On 2016/10/26 20:43:42, tsniatowski wrote:
> On 2016/10/26 19:31:00, Devlin wrote:
> > > can't use the variables here (like schema_files) as easily.  Dirk might
know
> > if
> > > there's an easy way for this to expose schema_files as a variable, which
> could
> > > then be included from the extensions/browser/api BUILD.gn, and if that's a
> > > reasonable approach to take here.

The way to share variables between files is to move the variable into a .gni
file; that gni file can then be imported by anything that needs it. In this
particular case, that looks easy enough.

[Devlin, 2016-10-26 21:57:52]

https://codereview.chromium.org/2452943003/diff/1/extensions/browser/app_wind...
File extensions/browser/app_window/BUILD.gn (right):

https://codereview.chromium.org/2452943003/diff/1/extensions/browser/app_wind...
extensions/browser/app_window/BUILD.gn:32: "//extensions:extensions_resources",
On 2016/10/26 20:43:41, tsniatowski wrote:
> On 2016/10/26 19:30:28, Devlin wrote:
> > could this just be extensions_browser_resources?
> 
> Checked -- yes. I can change this if you prefer.

Please.  extensions_resources includes renderer resources, which conceptually
should not be required in the browser.

https://codereview.chromium.org/2452943003/diff/1/extensions/browser/guest_vi...
File extensions/browser/guest_view/web_view/BUILD.gn (right):

https://codereview.chromium.org/2452943003/diff/1/extensions/browser/guest_vi...
extensions/browser/guest_view/web_view/BUILD.gn:28:
"//components/guest_view/browser",
On 2016/10/26 20:43:41, tsniatowski wrote:
> On 2016/10/26 19:30:28, Devlin wrote:
> > Do none of the other extensions/browser/guest_view files need this?  (More
> > surprised than anything else)
> 
> Apparently not from an "are there missing dependencies that can break the
build"
> perspective according to my test; from a gn check perspective -- maybe, I'm
not
> handling it.
> 
> Actually after a recheck this particular dep is not needed to fix the
flakiness.
> Keep it?

Let's hoist this up to extensions/browser/guest_view/BUILD.gn - we *do* depend
on it.

https://codereview.chromium.org/2452943003/diff/1/extensions/common/api/BUILD.gn
File extensions/common/api/BUILD.gn (right):

https://codereview.chromium.org/2452943003/diff/1/extensions/common/api/BUILD...
extensions/common/api/BUILD.gn:115: "//extensions/browser/api",
On 2016/10/26 21:49:21, Dirk Pranke wrote:
> On 2016/10/26 20:43:42, tsniatowski wrote:
> > On 2016/10/26 19:31:00, Devlin wrote:
> > > > can't use the variables here (like schema_files) as easily.  Dirk might
> know
> > > if
> > > > there's an easy way for this to expose schema_files as a variable, which
> > could
> > > > then be included from the extensions/browser/api BUILD.gn, and if that's
a
> > > > reasonable approach to take here.
> 
> The way to share variables between files is to move the variable into a .gni
> file; that gni file can then be imported by anything that needs it. In this
> particular case, that looks easy enough.

tsniatowski@, can we try doing this and moving this action to
extensions/browser/api/BUILD.gn?  I think that would be a lot cleaner.

(I realize this is only making the BUILD.gn reflect reality, but it's a good
opportunity to make it better. :))

[tsniatowski, 2016-10-27 08:18:59]

https://codereview.chromium.org/2452943003/diff/1/extensions/common/api/BUILD.gn
File extensions/common/api/BUILD.gn (right):

https://codereview.chromium.org/2452943003/diff/1/extensions/common/api/BUILD...
extensions/common/api/BUILD.gn:115: "//extensions/browser/api",
On 2016/10/26 21:57:52, Devlin wrote:
> On 2016/10/26 21:49:21, Dirk Pranke wrote:
> > On 2016/10/26 20:43:42, tsniatowski wrote:
> > > On 2016/10/26 19:31:00, Devlin wrote:
> > > > > can't use the variables here (like schema_files) as easily.  Dirk
might
> > know
> > > > if
> > > > > there's an easy way for this to expose schema_files as a variable,
which
> > > could
> > > > > then be included from the extensions/browser/api BUILD.gn, and if
that's
> a
> > > > > reasonable approach to take here.
> > 
> > The way to share variables between files is to move the variable into a .gni
> > file; that gni file can then be imported by anything that needs it. In this
> > particular case, that looks easy enough.
> 
> tsniatowski@, can we try doing this and moving this action to
> extensions/browser/api/BUILD.gn?  I think that would be a lot cleaner.
> 
> (I realize this is only making the BUILD.gn reflect reality, but it's a good
> opportunity to make it better. :))

OK, I'll have a look. For consistency, do you want an equivalent change in
//extensions/shell: extract schema args to .gni, move api registration to
browser/?

[tsniatowski, 2016-10-27 10:58:35]

On 2016/10/27 08:18:59, tsniatowski wrote:
>
https://codereview.chromium.org/2452943003/diff/1/extensions/common/api/BUILD.gn
> File extensions/common/api/BUILD.gn (right):
> 
>
https://codereview.chromium.org/2452943003/diff/1/extensions/common/api/BUILD...
> extensions/common/api/BUILD.gn:115: "//extensions/browser/api",
> On 2016/10/26 21:57:52, Devlin wrote:
> > On 2016/10/26 21:49:21, Dirk Pranke wrote:
> > > On 2016/10/26 20:43:42, tsniatowski wrote:
> > > > On 2016/10/26 19:31:00, Devlin wrote:
> > > > > > can't use the variables here (like schema_files) as easily.  Dirk
> might
> > > know
> > > > > if
> > > > > > there's an easy way for this to expose schema_files as a variable,
> which
> > > > could
> > > > > > then be included from the extensions/browser/api BUILD.gn, and if
> that's
> > a
> > > > > > reasonable approach to take here.
> > > 
> > > The way to share variables between files is to move the variable into a
.gni
> > > file; that gni file can then be imported by anything that needs it. In
this
> > > particular case, that looks easy enough.
> > 
> > tsniatowski@, can we try doing this and moving this action to
> > extensions/browser/api/BUILD.gn?  I think that would be a lot cleaner.
> > 
> > (I realize this is only making the BUILD.gn reflect reality, but it's a good
> > opportunity to make it better. :))
> 
> OK, I'll have a look. For consistency, do you want an equivalent change in
> //extensions/shell: extract schema args to .gni, move api registration to
> browser/?

Skipped shell for now. I'm not entirely convinced it turned out "better" this
way TBH.

[Devlin, 2016-10-27 16:00:59]

I think this lgtm.  I feel better about having the clearer browser/common
separation; thanks for doing that.  One more question for Dirk that might clean
this up a bit more.  Also, Dirk, if you can take a quick skim over this to make
sure it makes sense, that'd be great - I'm picking up gn, but am not quite
comfortable enough to assert that this won't cause problems. :)

https://codereview.chromium.org/2452943003/diff/20001/extensions/browser/api/...
File extensions/browser/api/BUILD.gn (right):

https://codereview.chromium.org/2452943003/diff/20001/extensions/browser/api/...
extensions/browser/api/BUILD.gn:96: rebase_path(extensions_api_schema_files,
".", "//extensions/common/api")
Can we assert() some of these as a sanity check?

https://codereview.chromium.org/2452943003/diff/20001/extensions/browser/api/...
extensions/browser/api/BUILD.gn:103: "//extensions/common/api")
gn accepts fully qualified paths (e.g. //extensions/common/api/foo.json).  I
wonder if we could, in the schema.gni file, export the variables like that? 
Then we could just use them in their raw form here.

dpranke@, is that possible?  Something like an "source_path" function or the
like?  rebase_path(path, "//", "//extensions/common/api") wouldn't quite work
because it wouldn't produce the leading //.

(We could also hypothetically just do string addition -
"//extensions/common/api/" + path - but that seems a little more hacky.)

[Dirk Pranke, 2016-10-27 21:28:44]

https://codereview.chromium.org/2452943003/diff/20001/extensions/browser/api/...
File extensions/browser/api/BUILD.gn (right):

https://codereview.chromium.org/2452943003/diff/20001/extensions/browser/api/...
extensions/browser/api/BUILD.gn:96: rebase_path(extensions_api_schema_files,
".", "//extensions/common/api")
On 2016/10/27 16:00:59, Devlin wrote:
> Can we assert() some of these as a sanity check?

What would you be asserting?

https://codereview.chromium.org/2452943003/diff/20001/extensions/browser/api/...
extensions/browser/api/BUILD.gn:103: "//extensions/common/api")
On 2016/10/27 16:00:59, Devlin wrote:
> gn accepts fully qualified paths (e.g. //extensions/common/api/foo.json).  I
> wonder if we could, in the schema.gni file, export the variables like that? 
> Then we could just use them in their raw form here.
> 
> dpranke@, is that possible?  Something like an "source_path" function or the
> like?  rebase_path(path, "//", "//extensions/common/api") wouldn't quite work
> because it wouldn't produce the leading //.
> 
> (We could also hypothetically just do string addition -
> "//extensions/common/api/" + path - but that seems a little more hacky.)

Use get_path_info(extension_api_schema_files, "abspath"), etc., in schema.gni 
(see `gn help get_path_info` for more).

[tsniatowski, 2016-10-31 17:25:39]

The CQ bit was checked by tsniatowski@opera.com to run a CQ dry run

[commit-bot: I haz the power, 2016-10-31 17:26:01]

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, 2016-10-31 17:56:31]

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

[commit-bot: I haz the power, 2016-10-31 17:56:33]

Dry run: Try jobs failed on following builders:
  ios-simulator on master.tryserver.chromium.mac (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.mac/builders/ios-simulator/bui...)

[tsniatowski, 2016-11-03 06:11:11]

On 2016/10/27 21:28:44, Dirk Pranke wrote:
>
https://codereview.chromium.org/2452943003/diff/20001/extensions/browser/api/...
> File extensions/browser/api/BUILD.gn (right):
> 
>
https://codereview.chromium.org/2452943003/diff/20001/extensions/browser/api/...
> extensions/browser/api/BUILD.gn:96: rebase_path(extensions_api_schema_files,
> ".", "//extensions/common/api")
> On 2016/10/27 16:00:59, Devlin wrote:
> > Can we assert() some of these as a sanity check?
> 
> What would you be asserting?
> 
>
https://codereview.chromium.org/2452943003/diff/20001/extensions/browser/api/...
> extensions/browser/api/BUILD.gn:103: "//extensions/common/api")
> On 2016/10/27 16:00:59, Devlin wrote:
> > gn accepts fully qualified paths (e.g. //extensions/common/api/foo.json).  I
> > wonder if we could, in the schema.gni file, export the variables like that? 
> > Then we could just use them in their raw form here.
> > 
> > dpranke@, is that possible?  Something like an "source_path" function or the
> > like?  rebase_path(path, "//", "//extensions/common/api") wouldn't quite
work
> > because it wouldn't produce the leading //.
> > 
> > (We could also hypothetically just do string addition -
> > "//extensions/common/api/" + path - but that seems a little more hacky.)
> 
> Use get_path_info(extension_api_schema_files, "abspath"), etc., in schema.gni 
> (see `gn help get_path_info` for more).

(back after local holidays) I'll assume the last fixup is fine, let me know if
you want some further tweaks and I'll do them in a followup.

[tsniatowski, 2016-11-03 06:11:20]

The CQ bit was checked by tsniatowski@opera.com

[tsniatowski, 2016-11-03 06:11:20]

The patchset sent to the CQ was uploaded after l-g-t-m from
wjmaclean@google.com, wjmaclean@chromium.org, rdevlin.cronin@chromium.org
Link to the patchset: https://codereview.chromium.org/2452943003/#ps40001
(title: "get_path_info in schema.gni")

[commit-bot: I haz the power, 2016-11-03 06:11:40]

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, 2016-11-03 08:12:30]

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

[commit-bot: I haz the power, 2016-11-03 08:12:31]

Try jobs failed on following builders:
  linux_chromium_asan_rel_ng on master.tryserver.chromium.linux (JOB_TIMED_OUT,
no build URL)

[tsniatowski, 2016-11-03 08:20:35]

The CQ bit was checked by tsniatowski@opera.com

[commit-bot: I haz the power, 2016-11-03 08:20:50]

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, 2016-11-03 10:21:11]

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

[commit-bot: I haz the power, 2016-11-03 10:21:12]

Try jobs failed on following builders:
  linux_chromium_asan_rel_ng on master.tryserver.chromium.linux (JOB_TIMED_OUT,
no build URL)

[tsniatowski, 2016-11-03 10:29:19]

The CQ bit was checked by tsniatowski@opera.com

[commit-bot: I haz the power, 2016-11-03 10:29:35]

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, 2016-11-03 11:52:14]

Description was changed from

==========
Fix a bunch of generated file build flakes in //extensions

Several files in //extensions could randomly fail to build due to
missing dependencies on header generator targets, mostly mojo
and grit. Add the dependencies so builds are not flaky.

BUG=655123
==========

to

==========
Fix a bunch of generated file build flakes in //extensions

Several files in //extensions could randomly fail to build due to
missing dependencies on header generator targets, mostly mojo
and grit. Add the dependencies so builds are not flaky.

BUG=655123
==========

[commit-bot: I haz the power, 2016-11-03 11:52:16]

Committed patchset #3 (id:40001)

[commit-bot: I haz the power, 2016-11-03 11:54:44]

Description was changed from

==========
Fix a bunch of generated file build flakes in //extensions

Several files in //extensions could randomly fail to build due to
missing dependencies on header generator targets, mostly mojo
and grit. Add the dependencies so builds are not flaky.

BUG=655123
==========

to

==========
Fix a bunch of generated file build flakes in //extensions

Several files in //extensions could randomly fail to build due to
missing dependencies on header generator targets, mostly mojo
and grit. Add the dependencies so builds are not flaky.

BUG=655123

Committed: https://crrev.com/dc44a53ed3244a4a68cb1fd4ccfaacf1e97e5d23
Cr-Commit-Position: refs/heads/master@{#429543}
==========

[commit-bot: I haz the power, 2016-11-03 11:54:45]

Patchset 3 (id:??) landed as
https://crrev.com/dc44a53ed3244a4a68cb1fd4ccfaacf1e97e5d23
Cr-Commit-Position: refs/heads/master@{#429543}