[Chromecast] Depend on Chromium's freetype-android

[Chromecast] Depend on Chromium's freetype-android

This change helps to reduce the external dependencies for OEMs to support cast_shell.

Updates third_party/libpng so it can be used as a dependency for freetype on Chromecast.

Remove the gyp variable use_custom_freetype because it is now unused.

BUG=internal b/26249831
TEST=builds and runs with gyp and gn

Committed: https://crrev.com/0ddf78a46654c13f9ee7de921a820087f45375e3
Cr-Commit-Position: refs/heads/master@{#372996}

[bcf, 2016-01-22 23:41:46]

Description was changed from

==========
[Chromecast] Depend on Chromium's freetype

BUG=internal b/26249831
TEST=builds and runs with gyp and gn
==========

to

==========
[Chromecast] Depend on Chromium's freetype

BUG=internal b/26249831
TEST=builds and runs with gyp and gn
==========

[bcf, 2016-01-22 23:41:46]

bcf@chromium.org changed reviewers:
+ mbjorge@chromium.org, slan@chromium.org, wzhong@chromium.org

[bcf, 2016-01-22 23:44:41]

This CL removes cast_shell dependencies on freetype2 and libpng.

It uses third_party/freetype2-android which is security critical

[bcf, 2016-01-22 23:45:06]

GN isn't updated yet. I'll update it after gyp reviewed.

[mbjorge, 2016-01-23 00:27:36]

lgtm

[mbjorge, 2016-01-23 00:28:05]

On 2016/01/23 at 00:27:36, mbjorge wrote:
In the commit message, you say 'builds and runs with gyp and gn'; which target
are you building in particular?

[bcf, 2016-01-23 00:37:54]

On 2016/01/23 00:28:05, mbjorge wrote:
> On 2016/01/23 at 00:27:36, mbjorge wrote:
> In the commit message, you say 'builds and runs with gyp and gn'; which target
> are you building in particular?

Building cast_shell and otapackage on gyp, cast_shell on gn. I can try the all
the other targets on GN after its updated.

I guess now the TEST is a lie, because this is a revival of the CL that used
third_party/freetype2 :).

[slan, 2016-01-25 17:35:16]

I know very little about freetype, but it is non-intuitive that a target called
"freetype-android" can be built for Linux. If this change works for us (and the
owners/experts for this code give us their blessing), I'm all for it, but we
should add comments.

Thanks for the CL, Bailey.

https://codereview.chromium.org/1627533002/diff/1/build/linux/BUILD.gn
File build/linux/BUILD.gn (right):

https://codereview.chromium.org/1627533002/diff/1/build/linux/BUILD.gn#newcode14
build/linux/BUILD.gn:14: use_system_fontconfig = !is_chromecast
Could you add a comment above these two vars explaining that Chromecast builds
cannot rely on system libraries for this functionality?

https://codereview.chromium.org/1627533002/diff/1/build/linux/BUILD.gn#newcod...
build/linux/BUILD.gn:125: "//third_party/freetype2",
This should be freetype-android target.

Also, there should be some kind of assert in here or a comment that explains
what's happening. Even with (some) context on the issue, this is confusing. Is
there a bug that we can link to to provide more context?

[slan, 2016-01-25 17:35:17]

I know very little about freetype, but it is non-intuitive that a target called
"freetype-android" can be built for Linux. If this change works for us (and the
owners/experts for this code give us their blessing), I'm all for it, but we
should add comments.

Thanks for the CL, Bailey.

[bcf, 2016-01-25 19:50:23]

On 2016/01/25 17:35:17, slan wrote:
> I know very little about freetype, but it is non-intuitive that a target
called
> "freetype-android" can be built for Linux. If this change works for us (and
the
> owners/experts for this code give us their blessing), I'm all for it, but we
> should add comments.
> 
> Thanks for the CL, Bailey.

Where do you suggest we add comments about usage of freetype-android? I added a
comment about it in build/linux/BUILD.gn.

FYI, this Cl also removes the cast_shell dependency on libpng.

[bcf, 2016-01-25 19:51:33]

https://codereview.chromium.org/1627533002/diff/1/build/linux/BUILD.gn
File build/linux/BUILD.gn (right):

https://codereview.chromium.org/1627533002/diff/1/build/linux/BUILD.gn#newcode14
build/linux/BUILD.gn:14: use_system_fontconfig = !is_chromecast
On 2016/01/25 17:35:16, slan wrote:
> Could you add a comment above these two vars explaining that Chromecast builds
> cannot rely on system libraries for this functionality?

Done.

In addition, I just changed freetype2 to check for chromecast directly, because
probably only chromecast wants to use freetype-android

https://codereview.chromium.org/1627533002/diff/1/build/linux/BUILD.gn#newcod...
build/linux/BUILD.gn:125: "//third_party/freetype2",
On 2016/01/25 17:35:16, slan wrote:
> This should be freetype-android target.
> 
> Also, there should be some kind of assert in here or a comment that explains
> what's happening. Even with (some) context on the issue, this is confusing. Is
> there a bug that we can link to to provide more context?

I haven't update GN yet in PS1. Its done now though.

I also added a comment for this.

[slan, 2016-01-27 23:57:06]

lgtm % nit

Please add OWNERS

https://codereview.chromium.org/1627533002/diff/1/build/linux/BUILD.gn
File build/linux/BUILD.gn (right):

https://codereview.chromium.org/1627533002/diff/1/build/linux/BUILD.gn#newcode14
build/linux/BUILD.gn:14: use_system_fontconfig = !is_chromecast
On 2016/01/25 19:51:33, bcf wrote:
> On 2016/01/25 17:35:16, slan wrote:
> > Could you add a comment above these two vars explaining that Chromecast
builds
> > cannot rely on system libraries for this functionality?
> 
> Done.
> 
> In addition, I just changed freetype2 to check for chromecast directly,
because
> probably only chromecast wants to use freetype-android

Though I understand that directly checking for Chromecast is fewer lines, I
think we should value consistency here. I would prefer we use the same
convention as is used with fontconfig.

[bcf, 2016-01-28 00:44:08]

bcf@chromium.org changed reviewers:
+ dpranke@chromium.org, thakis@chromium.org, thestig@chromium.org

[bcf, 2016-01-28 00:45:27]

+ OWNERS

Any thoughts on this?

thestig@ I had a similar CL last week, and we discussed how we shouldn't use
third_party/freetype because it is only used for testing. Would there be any
issues to using third_party/freetype-android?

[Dirk Pranke, 2016-01-28 01:09:21]

lgtm if the change in //content/shell/BUILD.gn is correct.

Can you please update the CL subject and description to say freetype-android
rather than just freetype (to minimize confusion) and also mention the libpng
change?

https://codereview.chromium.org/1627533002/diff/40001/content/shell/BUILD.gn
File content/shell/BUILD.gn (right):

https://codereview.chromium.org/1627533002/diff/40001/content/shell/BUILD.gn#...
content/shell/BUILD.gn:317: deps += [ "//third_party/freetype2" ]
is this right? you'll get //third_party/freetype2 in content_shell even for
chromecast, rather than //third_party/freetype-android ?

(just checking that that's the intent)

[bcf, 2016-01-28 21:36:44]

Description was changed from

==========
[Chromecast] Depend on Chromium's freetype

BUG=internal b/26249831
TEST=builds and runs with gyp and gn
==========

to

==========
[Chromecast] Depend on Chromium's freetype-android

This change helps to reduce the external dependencies for OEMs to support
cast_shell.

Updates third_party/libpng so it can be used as a dependency for freetype on
Chromecast.

Remove the gyp variable use_custom_freetype because it is now unused.

BUG=internal b/26249831
TEST=builds and runs with gyp and gn
==========

[bcf, 2016-01-28 21:37:28]

https://codereview.chromium.org/1627533002/diff/40001/content/shell/BUILD.gn
File content/shell/BUILD.gn (right):

https://codereview.chromium.org/1627533002/diff/40001/content/shell/BUILD.gn#...
content/shell/BUILD.gn:317: deps += [ "//third_party/freetype2" ]
On 2016/01/28 01:09:21, Dirk Pranke wrote:
> is this right? you'll get //third_party/freetype2 in content_shell even for
> chromecast, rather than //third_party/freetype-android ?
> 
> (just checking that that's the intent)

This logic is from this CL: https://codereview.chromium.org/1192933002

It seems like the issue from that CL no longer exists with statically linked
freetype

[Nico, 2016-01-28 21:42:00]

https://groups.google.com/a/chromium.org/d/msg/chromium-reviews/-wRFeRlSe2U/z...
suggests that Lei is pretty concerned about this. Don't land this before Lei has
commented.

[bcf, 2016-01-28 21:47:36]

bcf@chromium.org changed reviewers:
+ jam@chromium.org

[Lei Zhang, 2016-01-28 21:51:03]

On 2016/01/28 21:42:00, Nico wrote:
>
https://groups.google.com/a/chromium.org/d/msg/chromium-reviews/-wRFeRlSe2U/z...
> suggests that Lei is pretty concerned about this. Don't land this before Lei
has
> commented.

No issues from me since dpranke approved. I was concerned about the use of
third_party/freetype2.

[Nico, 2016-01-28 21:55:47]

ok, if Lei and Dirk are happy, then so am I. I did look over the code for a
second and I didn't understand your reply to one of Dirk's comments.

https://codereview.chromium.org/1627533002/diff/40001/content/shell/BUILD.gn
File content/shell/BUILD.gn (right):

https://codereview.chromium.org/1627533002/diff/40001/content/shell/BUILD.gn#...
content/shell/BUILD.gn:317: deps += [ "//third_party/freetype2" ]
On 2016/01/28 21:37:28, bcf wrote:
> On 2016/01/28 01:09:21, Dirk Pranke wrote:
> > is this right? you'll get //third_party/freetype2 in content_shell even for
> > chromecast, rather than //third_party/freetype-android ?
> > 
> > (just checking that that's the intent)
> 
> This logic is from this CL: https://codereview.chromium.org/1192933002
> 
> It seems like the issue from that CL no longer exists with statically linked
> freetype

This looks wrong to me too -- you probably don't want to depend on different
freetypes? Don't you want build/linux/freetype here as dep (and everywhere else
too)?

[bcf, 2016-01-28 22:31:04]

https://codereview.chromium.org/1627533002/diff/40001/content/shell/BUILD.gn
File content/shell/BUILD.gn (right):

https://codereview.chromium.org/1627533002/diff/40001/content/shell/BUILD.gn#...
content/shell/BUILD.gn:317: deps += [ "//third_party/freetype2" ]
On 2016/01/28 21:55:47, Nico wrote:
> On 2016/01/28 21:37:28, bcf wrote:
> > On 2016/01/28 01:09:21, Dirk Pranke wrote:
> > > is this right? you'll get //third_party/freetype2 in content_shell even
for
> > > chromecast, rather than //third_party/freetype-android ?
> > > 
> > > (just checking that that's the intent)
> > 
> > This logic is from this CL: https://codereview.chromium.org/1192933002
> > 
> > It seems like the issue from that CL no longer exists with statically linked
> > freetype
> 
> This looks wrong to me too -- you probably don't want to depend on different
> freetypes? Don't you want build/linux/freetype here as dep (and everywhere
else
> too)?

We discussed in the predecessor of this CL that third_party/freetype2 is only
for testing. (See third_party/freetype2/README.chromium). content_shell_lib
should always use third_party/freetype2 to ensure fonts always render the same
way for tests.

[Nico, 2016-01-28 22:33:01]

lgtm, don't wait with landing until I understand this :-) I'd appreciate a reply
though :-)

https://codereview.chromium.org/1627533002/diff/40001/content/shell/BUILD.gn
File content/shell/BUILD.gn (right):

https://codereview.chromium.org/1627533002/diff/40001/content/shell/BUILD.gn#...
content/shell/BUILD.gn:317: deps += [ "//third_party/freetype2" ]
On 2016/01/28 22:31:04, bcf wrote:
> On 2016/01/28 21:55:47, Nico wrote:
> > On 2016/01/28 21:37:28, bcf wrote:
> > > On 2016/01/28 01:09:21, Dirk Pranke wrote:
> > > > is this right? you'll get //third_party/freetype2 in content_shell even
> for
> > > > chromecast, rather than //third_party/freetype-android ?
> > > > 
> > > > (just checking that that's the intent)
> > > 
> > > This logic is from this CL: https://codereview.chromium.org/1192933002
> > > 
> > > It seems like the issue from that CL no longer exists with statically
linked
> > > freetype
> > 
> > This looks wrong to me too -- you probably don't want to depend on different
> > freetypes? Don't you want build/linux/freetype here as dep (and everywhere
> else
> > too)?
> 
> We discussed in the predecessor of this CL that third_party/freetype2 is only
> for testing. (See third_party/freetype2/README.chromium). content_shell_lib
> should always use third_party/freetype2 to ensure fonts always render the same
> way for tests. 

Does android use freetype2 for tests and freetype-android for production too?
Seems weird to test with one freetype and ship another, no?

[Lei Zhang, 2016-01-28 22:41:45]

On 2016/01/28 22:33:01, Nico wrote:
> Does android use freetype2 for tests and freetype-android for production too?
> Seems weird to test with one freetype and ship another, no?

Given the "Security Critical: yes" field, it's probably shipped.

It may be good to also update third_party/freetype-android/README.chromium and
get approval from third_party/freetype-android/OWNERS.

[bcf, 2016-01-28 22:44:51]

On 2016/01/28 22:33:01, Nico wrote:
> lgtm, don't wait with landing until I understand this :-) I'd appreciate a
reply
> though :-)
> 
> https://codereview.chromium.org/1627533002/diff/40001/content/shell/BUILD.gn
> File content/shell/BUILD.gn (right):
> 
>
https://codereview.chromium.org/1627533002/diff/40001/content/shell/BUILD.gn#...
> content/shell/BUILD.gn:317: deps += [ "//third_party/freetype2" ]
> On 2016/01/28 22:31:04, bcf wrote:
> > On 2016/01/28 21:55:47, Nico wrote:
> > > On 2016/01/28 21:37:28, bcf wrote:
> > > > On 2016/01/28 01:09:21, Dirk Pranke wrote:
> > > > > is this right? you'll get //third_party/freetype2 in content_shell
even
> > for
> > > > > chromecast, rather than //third_party/freetype-android ?
> > > > > 
> > > > > (just checking that that's the intent)
> > > > 
> > > > This logic is from this CL: https://codereview.chromium.org/1192933002
> > > > 
> > > > It seems like the issue from that CL no longer exists with statically
> linked
> > > > freetype
> > > 
> > > This looks wrong to me too -- you probably don't want to depend on
different
> > > freetypes? Don't you want build/linux/freetype here as dep (and everywhere
> > else
> > > too)?
> > 
> > We discussed in the predecessor of this CL that third_party/freetype2 is
only
> > for testing. (See third_party/freetype2/README.chromium). content_shell_lib
> > should always use third_party/freetype2 to ensure fonts always render the
same
> > way for tests. 
> 
> Does android use freetype2 for tests and freetype-android for production too?
> Seems weird to test with one freetype and ship another, no?

It looks like this is only for Linux builds. It's guarded with OS="linux" on gyp
and is_linux on gn.

Here's the rational from third_party/freetype2/README.chromium:

"We link this library into DumpRenderTree so that we can run the layout tests
on later versions of Ubuntu and still get the same font rendering so that
we don't have to support two sets of pixel test baselines."

I don't believe that chromecast actually uses freetype2 from content_shell_lib
in the tests, but this broke our build in the past.

[Dirk Pranke, 2016-01-28 22:51:52]

On 2016/01/28 22:44:51, bcf wrote:
> On 2016/01/28 22:33:01, Nico wrote:
> > lgtm, don't wait with landing until I understand this :-) I'd appreciate a
> reply
> > though :-)
> > 
> > https://codereview.chromium.org/1627533002/diff/40001/content/shell/BUILD.gn
> > File content/shell/BUILD.gn (right):
> > 
> >
>
https://codereview.chromium.org/1627533002/diff/40001/content/shell/BUILD.gn#...
> > content/shell/BUILD.gn:317: deps += [ "//third_party/freetype2" ]
> > On 2016/01/28 22:31:04, bcf wrote:
> > > On 2016/01/28 21:55:47, Nico wrote:
> > > > On 2016/01/28 21:37:28, bcf wrote:
> > > > > On 2016/01/28 01:09:21, Dirk Pranke wrote:
> > > > > > is this right? you'll get //third_party/freetype2 in content_shell
> even
> > > for
> > > > > > chromecast, rather than //third_party/freetype-android ?
> > > > > > 
> > > > > > (just checking that that's the intent)
> > > > > 
> > > > > This logic is from this CL: https://codereview.chromium.org/1192933002
> > > > > 
> > > > > It seems like the issue from that CL no longer exists with statically
> > linked
> > > > > freetype
> > > > 
> > > > This looks wrong to me too -- you probably don't want to depend on
> different
> > > > freetypes? Don't you want build/linux/freetype here as dep (and
everywhere
> > > else
> > > > too)?
> > > 
> > > We discussed in the predecessor of this CL that third_party/freetype2 is
> only
> > > for testing. (See third_party/freetype2/README.chromium).
content_shell_lib
> > > should always use third_party/freetype2 to ensure fonts always render the
> same
> > > way for tests. 
> > 
> > Does android use freetype2 for tests and freetype-android for production
too?
> > Seems weird to test with one freetype and ship another, no?
> 
> It looks like this is only for Linux builds. It's guarded with OS="linux" on
gyp
> and is_linux on gn.
> 
> Here's the rational from third_party/freetype2/README.chromium:
> 
> "We link this library into DumpRenderTree so that we can run the layout tests
> on later versions of Ubuntu and still get the same font rendering so that
> we don't have to support two sets of pixel test baselines."
> 
> I don't believe that chromecast actually uses freetype2 from content_shell_lib
> in the tests, but this broke our build in the past.

More background, since I'm the one responsible for this mess ... 

I believe Android always uses freetype-android and never uses freetype2.

Ideally we would use the same version for both ports, as that might allow us to
share
more layout test baselines. 

Eventually we want to switch linux to using freetype-android (i.e., statically
link
in a current version of freetype) to get the latest security and rendering fixes
regardless of what the system version is.

I don't think it matters much which version content_shell uses for chromecast
since
we don't run the layout tests.

So, this change still lgtm as long as the subject line and CL description are
updated.

[bcf, 2016-01-28 22:55:42]

On 2016/01/28 22:51:52, Dirk Pranke wrote:
> On 2016/01/28 22:44:51, bcf wrote:
> > On 2016/01/28 22:33:01, Nico wrote:
> > > lgtm, don't wait with landing until I understand this :-) I'd appreciate a
> > reply
> > > though :-)
> > > 
> > >
https://codereview.chromium.org/1627533002/diff/40001/content/shell/BUILD.gn
> > > File content/shell/BUILD.gn (right):
> > > 
> > >
> >
>
https://codereview.chromium.org/1627533002/diff/40001/content/shell/BUILD.gn#...
> > > content/shell/BUILD.gn:317: deps += [ "//third_party/freetype2" ]
> > > On 2016/01/28 22:31:04, bcf wrote:
> > > > On 2016/01/28 21:55:47, Nico wrote:
> > > > > On 2016/01/28 21:37:28, bcf wrote:
> > > > > > On 2016/01/28 01:09:21, Dirk Pranke wrote:
> > > > > > > is this right? you'll get //third_party/freetype2 in content_shell
> > even
> > > > for
> > > > > > > chromecast, rather than //third_party/freetype-android ?
> > > > > > > 
> > > > > > > (just checking that that's the intent)
> > > > > > 
> > > > > > This logic is from this CL:
https://codereview.chromium.org/1192933002
> > > > > > 
> > > > > > It seems like the issue from that CL no longer exists with
statically
> > > linked
> > > > > > freetype
> > > > > 
> > > > > This looks wrong to me too -- you probably don't want to depend on
> > different
> > > > > freetypes? Don't you want build/linux/freetype here as dep (and
> everywhere
> > > > else
> > > > > too)?
> > > > 
> > > > We discussed in the predecessor of this CL that third_party/freetype2 is
> > only
> > > > for testing. (See third_party/freetype2/README.chromium).
> content_shell_lib
> > > > should always use third_party/freetype2 to ensure fonts always render
the
> > same
> > > > way for tests. 
> > > 
> > > Does android use freetype2 for tests and freetype-android for production
> too?
> > > Seems weird to test with one freetype and ship another, no?
> > 
> > It looks like this is only for Linux builds. It's guarded with OS="linux" on
> gyp
> > and is_linux on gn.
> > 
> > Here's the rational from third_party/freetype2/README.chromium:
> > 
> > "We link this library into DumpRenderTree so that we can run the layout
tests
> > on later versions of Ubuntu and still get the same font rendering so that
> > we don't have to support two sets of pixel test baselines."
> > 
> > I don't believe that chromecast actually uses freetype2 from
content_shell_lib
> > in the tests, but this broke our build in the past.
> 
> More background, since I'm the one responsible for this mess ... 
> 
> I believe Android always uses freetype-android and never uses freetype2.
> 
> Ideally we would use the same version for both ports, as that might allow us
to
> share
> more layout test baselines. 
> 
> Eventually we want to switch linux to using freetype-android (i.e., statically
> link
> in a current version of freetype) to get the latest security and rendering
fixes
> regardless of what the system version is.
> 
> I don't think it matters much which version content_shell uses for chromecast
> since
> we don't run the layout tests.
> 
> So, this change still lgtm as long as the subject line and CL description are
> updated.

Dirk,

Are the current subject and description okay?

[bcf, 2016-01-28 22:56:40]

bcf@chromium.org changed reviewers:
+ wangxianzhu@chromium.org

[bcf, 2016-01-28 22:57:07]

On 2016/01/28 22:41:45, Lei Zhang wrote:
> On 2016/01/28 22:33:01, Nico wrote:
> > Does android use freetype2 for tests and freetype-android for production
too?
> > Seems weird to test with one freetype and ship another, no?
> 
> Given the "Security Critical: yes" field, it's probably shipped.
> 
> It may be good to also update third_party/freetype-android/README.chromium and
> get approval from third_party/freetype-android/OWNERS.

+ wangxianzhu@ for third_party/freetype-android/OWNERS

[Dirk Pranke, 2016-01-28 23:00:05]

On 2016/01/28 22:55:42, bcf wrote:
> Dirk,
> 
> Are the current subject and description okay?

Yup, thanks! Sorry if I didn't notice if had already been updated before ...

[Xianzhu, 2016-01-28 23:06:15]

On 2016/01/28 22:57:07, bcf wrote:
> On 2016/01/28 22:41:45, Lei Zhang wrote:
> > On 2016/01/28 22:33:01, Nico wrote:
> > > Does android use freetype2 for tests and freetype-android for production
> too?
> > > Seems weird to test with one freetype and ship another, no?
> > 

Android uses freetype-android for both tests and production.

> > Given the "Security Critical: yes" field, it's probably shipped.
> > 
> > It may be good to also update third_party/freetype-android/README.chromium
and
> > get approval from third_party/freetype-android/OWNERS.
> 
> + wangxianzhu@ for third_party/freetype-android/OWNERS

lgtm.

[jam, 2016-01-29 17:02:57]

did you add me for content? if so rubberstamp lgtm based on the other reviews

[bcf, 2016-01-29 23:39:40]

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

[commit-bot: I haz the power, 2016-01-29 23:40:44]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1627533002/80001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1627533002/80001

[bcf, 2016-01-29 23:44:28]

On 2016/01/29 17:02:57, jam wrote:
> did you add me for content? if so rubberstamp lgtm based on the other reviews

Thanks.

[bcf, 2016-01-29 23:45:59]

After discussing with slan@, PS4 removes the use_system_fontconfig from
build/linux/BUILD.gn to be consistent with the way freetype is handled.

[commit-bot: I haz the power, 2016-01-30 00:00:42]

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

[commit-bot: I haz the power, 2016-01-30 00:00:43]

Dry run: Try jobs failed on following builders:
  cast_shell_linux on tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/cast_shell_linu...)

[slan, 2016-01-30 00:38:13]

On 2016/01/30 00:00:43, commit-bot: I haz the power wrote:
> Dry run: Try jobs failed on following builders:
>   cast_shell_linux on tryserver.chromium.linux (JOB_FAILED,
>
http://build.chromium.org/p/tryserver.chromium.linux/builders/cast_shell_linu...)

Looks like the DEPS aren't bringing in the code for a Linux build :(
https://code.google.com/p/chromium/codesearch#chromium/src/DEPS&l=452

As discussed offline, let's update that in another patch.

[bcf, 2016-02-02 05:12:58]

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

[commit-bot: I haz the power, 2016-02-02 05:13:29]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1627533002/80001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1627533002/80001

[commit-bot: I haz the power, 2016-02-02 07:51:17]

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

[commit-bot: I haz the power, 2016-02-02 07:51:19]

Dry run: This issue passed the CQ dry run.

[bcf, 2016-02-02 18:41:53]

The CQ bit was checked by bcf@chromium.org

[bcf, 2016-02-02 18:41:53]

The patchset sent to the CQ was uploaded after l-g-t-m from
mbjorge@chromium.org, slan@chromium.org, thakis@chromium.org,
dpranke@chromium.org, jam@chromium.org, wangxianzhu@chromium.org
Link to the patchset: https://codereview.chromium.org/1627533002/#ps80001
(title: "Remove use_system_fontconfig flag from build/linux/BUILD.gn")

[commit-bot: I haz the power, 2016-02-02 18:44:17]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1627533002/80001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1627533002/80001

[commit-bot: I haz the power, 2016-02-02 18:52:33]

Description was changed from

==========
[Chromecast] Depend on Chromium's freetype-android

This change helps to reduce the external dependencies for OEMs to support
cast_shell.

Updates third_party/libpng so it can be used as a dependency for freetype on
Chromecast.

Remove the gyp variable use_custom_freetype because it is now unused.

BUG=internal b/26249831
TEST=builds and runs with gyp and gn
==========

to

==========
[Chromecast] Depend on Chromium's freetype-android

This change helps to reduce the external dependencies for OEMs to support
cast_shell.

Updates third_party/libpng so it can be used as a dependency for freetype on
Chromecast.

Remove the gyp variable use_custom_freetype because it is now unused.

BUG=internal b/26249831
TEST=builds and runs with gyp and gn
==========

[commit-bot: I haz the power, 2016-02-02 18:52:34]

Committed patchset #5 (id:80001)

[commit-bot: I haz the power, 2016-02-02 18:54:21]

Description was changed from

==========
[Chromecast] Depend on Chromium's freetype-android

This change helps to reduce the external dependencies for OEMs to support
cast_shell.

Updates third_party/libpng so it can be used as a dependency for freetype on
Chromecast.

Remove the gyp variable use_custom_freetype because it is now unused.

BUG=internal b/26249831
TEST=builds and runs with gyp and gn
==========

to

==========
[Chromecast] Depend on Chromium's freetype-android

This change helps to reduce the external dependencies for OEMs to support
cast_shell.

Updates third_party/libpng so it can be used as a dependency for freetype on
Chromecast.

Remove the gyp variable use_custom_freetype because it is now unused.

BUG=internal b/26249831
TEST=builds and runs with gyp and gn

Committed: https://crrev.com/0ddf78a46654c13f9ee7de921a820087f45375e3
Cr-Commit-Position: refs/heads/master@{#372996}
==========

[commit-bot: I haz the power, 2016-02-02 18:54:22]

Patchset 5 (id:??) landed as
https://crrev.com/0ddf78a46654c13f9ee7de921a820087f45375e3
Cr-Commit-Position: refs/heads/master@{#372996}