Reland of "Decoupling rtc_base from Obj-C code"

Reland of "Decoupling rtc_base from Obj-C code"

Reason for revert:
Trying to fix the link error on the chromium FYI trybots.

Original issue's description:
> Revert of Decoupling rtc_base from Obj-C code (patchset #1 id:1 of https://codereview.webrtc.org/3004373002/ )
> 
> Reason for revert:
> It breaks a chromium FYI trybot (the switch from rtc_static_library to rtc_source_set seems to be the problem).
> 
> Original issue's description:
> > Reland of Decoupling rtc_base from Obj-C code (patchset #1 id:1 of https://codereview.webrtc.org/3008103002/ )
> > 
> > Reason for revert:
> > Trying to reland after fixing downstream project.
> > 
> > Original issue's description:
> > > Revert of Decoupling rtc_base from Obj-C code (patchset #8 id:140001 of https://codereview.webrtc.org/3001623003/ )
> > > 
> > > Reason for revert:
> > > It breaks a downstream project.
> > > 
> > > Original issue's description:
> > > > Decoupling rtc_base from Obj-C code
> > > > 
> > > > The goal of this CL is to separate Obj-C/Obj-C++ code from targets which have
> > > > also C++ code (see https://bugs.chromium.org/p/webrtc/issues/detail?id=7743
> > > > for more information).
> > > > 
> > > > To achieve this we have created 2 targets (rtc_base_objc and rtc_base_generic)
> > > > and rtc_base will act as a proxy between these targets (this way we can avoid a
> > > > circular dependency between rtc_base_generic and rtc_base_objc).
> > > > 
> > > > BUG=webrtc:7743
> > > > 
> > > > Review-Url: https://codereview.webrtc.org/3001623003
> > > > Cr-Commit-Position: refs/heads/master@{#19661}
> > > > Committed: https://chromium.googlesource.com/external/webrtc/+/a0a5b98937de0f0bf90e17011d2cbc8c632cb409
> > > 
> > > TBR=kjellander@webrtc.org,kwiberg@webrtc.org
> > > # Skipping CQ checks because original CL landed less than 1 days ago.
> > > NOPRESUBMIT=true
> > > NOTREECHECKS=true
> > > NOTRY=true
> > > BUG=webrtc:7743
> > > 
> > > Review-Url: https://codereview.webrtc.org/3008103002
> > > Cr-Commit-Position: refs/heads/master@{#19664}
> > > Committed: https://chromium.googlesource.com/external/webrtc/+/791a8f611da8cc6b0668ccc03bf7525e6ba72a46
> > 
> > TBR=kjellander@webrtc.org,kwiberg@webrtc.org
> > # Skipping CQ checks because original CL landed less than 1 days ago.
> > NOPRESUBMIT=true
> > NOTREECHECKS=true
> > NOTRY=true
> > BUG=webrtc:7743
> > 
> > Review-Url: https://codereview.webrtc.org/3004373002
> > Cr-Commit-Position: refs/heads/master@{#19681}
> > Committed: https://chromium.googlesource.com/external/webrtc/+/a0c893f324da5c9f0773f8990a29cd24643864e4
> 
> TBR=kjellander@webrtc.org,kwiberg@webrtc.org
> # Skipping CQ checks because original CL landed less than 1 days ago.
> NOPRESUBMIT=true
> NOTREECHECKS=true
> NOTRY=true
> BUG=webrtc:7743
> 
> Review-Url: https://codereview.webrtc.org/3007183002
> Cr-Commit-Position: refs/heads/master@{#19688}
> Committed: https://chromium.googlesource.com/external/webrtc/+/9c5e511806ef64c4cede52722c8a7ae7ae50ff56

TBR=kjellander@webrtc.org,kwiberg@webrtc.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=webrtc:7743
BUG=webrtc:7743
Review-Url: https://codereview.webrtc.org/3012823002
Cr-Commit-Position: refs/heads/master@{#19726}
Committed: https://chromium.googlesource.com/external/webrtc/+/310e32b92a588ed31dad63c3b94a19cd025190f4

[mbonadei, 2017-09-06 08:52:20]

Description was changed from

==========
Revert "Revert of Decoupling rtc_base from Obj-C code (patchset #1 id:1 of
https://codereview.webrtc.org/3004373002/ )"

This reverts commit 9c5e511806ef64c4cede52722c8a7ae7ae50ff56.

BUG=webrtc:7743
==========

to

==========
Reland of "Decoupling rtc_base from Obj-C code"



BUG=webrtc:7743
==========

[mbonadei, 2017-09-06 15:15:07]

mbonadei@webrtc.org changed reviewers:
+ kjellander@webrtc.org

[mbonadei, 2017-09-06 15:15:08]

It seems that if we switch rtc_base and rtc_base_approved back to
rtc_static_library we are going to fix the frame_analyzer link error.

The problem is caused (probably) by the differences in how source_set and
static_library are linked.

Tested:
win_clang was failing in PS#2 and then it went green on PS#3.

[mbonadei, 2017-09-06 15:16:24]

Description was changed from

==========
Reland of "Decoupling rtc_base from Obj-C code"



BUG=webrtc:7743
==========

to

==========
Reland of "Decoupling rtc_base from Obj-C code"

Reason for revert:
Trying to fix the link error on the chromium FYI trybots.

Original issue's description:
> Revert of Decoupling rtc_base from Obj-C code (patchset #1 id:1 of
https://codereview.webrtc.org/3004373002/ )
> 
> Reason for revert:
> It breaks a chromium FYI trybot (the switch from rtc_static_library to
rtc_source_set seems to be the problem).
> 
> Original issue's description:
> > Reland of Decoupling rtc_base from Obj-C code (patchset #1 id:1 of
https://codereview.webrtc.org/3008103002/ )
> > 
> > Reason for revert:
> > Trying to reland after fixing downstream project.
> > 
> > Original issue's description:
> > > Revert of Decoupling rtc_base from Obj-C code (patchset #8 id:140001 of
https://codereview.webrtc.org/3001623003/ )
> > > 
> > > Reason for revert:
> > > It breaks a downstream project.
> > > 
> > > Original issue's description:
> > > > Decoupling rtc_base from Obj-C code
> > > > 
> > > > The goal of this CL is to separate Obj-C/Obj-C++ code from targets which
have
> > > > also C++ code (see
https://bugs.chromium.org/p/webrtc/issues/detail?id=7743
> > > > for more information).
> > > > 
> > > > To achieve this we have created 2 targets (rtc_base_objc and
rtc_base_generic)
> > > > and rtc_base will act as a proxy between these targets (this way we can
avoid a
> > > > circular dependency between rtc_base_generic and rtc_base_objc).
> > > > 
> > > > BUG=webrtc:7743
> > > > 
> > > > Review-Url: https://codereview.webrtc.org/3001623003
> > > > Cr-Commit-Position: refs/heads/master@{#19661}
> > > > Committed:
https://chromium.googlesource.com/external/webrtc/+/a0a5b98937de0f0bf90e17011...
> > > 
> > > TBR=kjellander@webrtc.org,kwiberg@webrtc.org
> > > # Skipping CQ checks because original CL landed less than 1 days ago.
> > > NOPRESUBMIT=true
> > > NOTREECHECKS=true
> > > NOTRY=true
> > > BUG=webrtc:7743
> > > 
> > > Review-Url: https://codereview.webrtc.org/3008103002
> > > Cr-Commit-Position: refs/heads/master@{#19664}
> > > Committed:
https://chromium.googlesource.com/external/webrtc/+/791a8f611da8cc6b0668ccc03...
> > 
> > TBR=kjellander@webrtc.org,kwiberg@webrtc.org
> > # Skipping CQ checks because original CL landed less than 1 days ago.
> > NOPRESUBMIT=true
> > NOTREECHECKS=true
> > NOTRY=true
> > BUG=webrtc:7743
> > 
> > Review-Url: https://codereview.webrtc.org/3004373002
> > Cr-Commit-Position: refs/heads/master@{#19681}
> > Committed:
https://chromium.googlesource.com/external/webrtc/+/a0c893f324da5c9f0773f8990...
> 
> TBR=kjellander@webrtc.org,kwiberg@webrtc.org
> # Skipping CQ checks because original CL landed less than 1 days ago.
> NOPRESUBMIT=true
> NOTREECHECKS=true
> NOTRY=true
> BUG=webrtc:7743
> 
> Review-Url: https://codereview.webrtc.org/3007183002
> Cr-Commit-Position: refs/heads/master@{#19688}
> Committed:
https://chromium.googlesource.com/external/webrtc/+/9c5e511806ef64c4cede52722...

TBR=kjellander@webrtc.org,kwiberg@webrtc.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=webrtc:7743



BUG=webrtc:7743
==========

[kjellander_webrtc, 2017-09-06 15:17:46]

lgtm assuming all the bots are happy. Will you run some Chromium trybots as
well?

[mbonadei, 2017-09-06 15:21:23]

On 2017/09/06 15:17:46, kjellander_webrtc wrote:
> lgtm assuming all the bots are happy. Will you run some Chromium trybots as
> well?

Right, I scheduled them right now.

[mbonadei, 2017-09-06 15:23:46]

On 2017/09/06 15:21:23, mbonadei wrote:
> On 2017/09/06 15:17:46, kjellander_webrtc wrote:
> > lgtm assuming all the bots are happy. Will you run some Chromium trybots as
> > well?
> 
> Right, I scheduled them right now.

Well, we have other link errors here. I will continue tomorrow.

[kjellander_webrtc, 2017-09-06 17:59:36]

https://codereview.webrtc.org/3012823002/diff/40001/webrtc/rtc_base/BUILD.gn
File webrtc/rtc_base/BUILD.gn (right):

https://codereview.webrtc.org/3012823002/diff/40001/webrtc/rtc_base/BUILD.gn#...
webrtc/rtc_base/BUILD.gn:431: public_deps = [
So the generic target is made to avoid the cyclic dependency error, right?

I think you have three options here if we want this to work then:
1) add a noop.cc file to satisfy linking for static library, it fails if there
isn't any source.
2) move only the parts that are involved in the cycle to generic and keep other
source in rtc_base, but that's quite confusing.
3) just disable gn check for the rtc_base_objc target.

[mbonadei, 2017-09-06 19:28:59]

On 2017/09/06 17:59:36, kjellander_webrtc wrote:
> https://codereview.webrtc.org/3012823002/diff/40001/webrtc/rtc_base/BUILD.gn
> File webrtc/rtc_base/BUILD.gn (right):
> 
>
https://codereview.webrtc.org/3012823002/diff/40001/webrtc/rtc_base/BUILD.gn#...
> webrtc/rtc_base/BUILD.gn:431: public_deps = [
> So the generic target is made to avoid the cyclic dependency error, right?
> 
> I think you have three options here if we want this to work then:
> 1) add a noop.cc file to satisfy linking for static library, it fails if there
> isn't any source.
> 2) move only the parts that are involved in the cycle to generic and keep
other
> source in rtc_base, but that's quite confusing.
> 3) just disable gn check for the rtc_base_objc target.

I think option N.1 is the best in this situation. I will try that one.

[mbonadei, 2017-09-07 07:11:32]

mbonadei@webrtc.org changed reviewers:
+ kwiberg@webrtc.org

[mbonadei, 2017-09-07 07:12:51]

On 2017/09/06 19:28:59, mbonadei wrote:
> On 2017/09/06 17:59:36, kjellander_webrtc wrote:
> > https://codereview.webrtc.org/3012823002/diff/40001/webrtc/rtc_base/BUILD.gn
> > File webrtc/rtc_base/BUILD.gn (right):
> > 
> >
>
https://codereview.webrtc.org/3012823002/diff/40001/webrtc/rtc_base/BUILD.gn#...
> > webrtc/rtc_base/BUILD.gn:431: public_deps = [
> > So the generic target is made to avoid the cyclic dependency error, right?
> > 
> > I think you have three options here if we want this to work then:
> > 1) add a noop.cc file to satisfy linking for static library, it fails if
there
> > isn't any source.
> > 2) move only the parts that are involved in the cycle to generic and keep
> other
> > source in rtc_base, but that's quite confusing.
> > 3) just disable gn check for the rtc_base_objc target.
> 
> I think option N.1 is the best in this situation. I will try that one.

kwiberg@: at the end of the day it turns out that we have to use
rtc_static_library
(at least for rtc_base and rtc_base_generic) and when we target mac/ios and win
we
also need to provide at least one source file.

[kwiberg-webrtc, 2017-09-07 09:20:28]

On 2017/09/07 07:12:51, mbonadei wrote:
> On 2017/09/06 19:28:59, mbonadei wrote:
> > On 2017/09/06 17:59:36, kjellander_webrtc wrote:
> > >
https://codereview.webrtc.org/3012823002/diff/40001/webrtc/rtc_base/BUILD.gn
> > > File webrtc/rtc_base/BUILD.gn (right):
> > > 
> > >
> >
>
https://codereview.webrtc.org/3012823002/diff/40001/webrtc/rtc_base/BUILD.gn#...
> > > webrtc/rtc_base/BUILD.gn:431: public_deps = [
> > > So the generic target is made to avoid the cyclic dependency error, right?
> > > 
> > > I think you have three options here if we want this to work then:
> > > 1) add a noop.cc file to satisfy linking for static library, it fails if
> there
> > > isn't any source.
> > > 2) move only the parts that are involved in the cycle to generic and keep
> > other
> > > source in rtc_base, but that's quite confusing.
> > > 3) just disable gn check for the rtc_base_objc target.
> > 
> > I think option N.1 is the best in this situation. I will try that one.
> 
> kwiberg@: at the end of the day it turns out that we have to use
> rtc_static_library
> (at least for rtc_base and rtc_base_generic) and when we target mac/ios and
win
> we
> also need to provide at least one source file.

OK. Too bad, but if we have to, we have to.

[mbonadei, 2017-09-07 14:36:21]

The CQ bit was checked by mbonadei@webrtc.org

[mbonadei, 2017-09-07 14:36:22]

The patchset sent to the CQ was uploaded after l-g-t-m from
kjellander@webrtc.org
Link to the patchset: https://codereview.webrtc.org/3012823002/#ps80001 (title:
"Fixing windows trybots")

[commit-bot: I haz the power, 2017-09-07 14:36:25]

CQ is trying da patch.

Follow status at:
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.webrtc.org/...

[commit-bot: I haz the power, 2017-09-07 14:36:28]

CQ is committing da patch.

Bot data: {"patchset_id": 80001, "attempt_start_ts": 1504794981704830,
"parent_rev": "05ec178808fc8b0da4511cfaac1f734304fd48eb", "commit_rev":
"310e32b92a588ed31dad63c3b94a19cd025190f4"}

[commit-bot: I haz the power, 2017-09-07 14:36:34]

Description was changed from

==========
Reland of "Decoupling rtc_base from Obj-C code"

Reason for revert:
Trying to fix the link error on the chromium FYI trybots.

Original issue's description:
> Revert of Decoupling rtc_base from Obj-C code (patchset #1 id:1 of
https://codereview.webrtc.org/3004373002/ )
> 
> Reason for revert:
> It breaks a chromium FYI trybot (the switch from rtc_static_library to
rtc_source_set seems to be the problem).
> 
> Original issue's description:
> > Reland of Decoupling rtc_base from Obj-C code (patchset #1 id:1 of
https://codereview.webrtc.org/3008103002/ )
> > 
> > Reason for revert:
> > Trying to reland after fixing downstream project.
> > 
> > Original issue's description:
> > > Revert of Decoupling rtc_base from Obj-C code (patchset #8 id:140001 of
https://codereview.webrtc.org/3001623003/ )
> > > 
> > > Reason for revert:
> > > It breaks a downstream project.
> > > 
> > > Original issue's description:
> > > > Decoupling rtc_base from Obj-C code
> > > > 
> > > > The goal of this CL is to separate Obj-C/Obj-C++ code from targets which
have
> > > > also C++ code (see
https://bugs.chromium.org/p/webrtc/issues/detail?id=7743
> > > > for more information).
> > > > 
> > > > To achieve this we have created 2 targets (rtc_base_objc and
rtc_base_generic)
> > > > and rtc_base will act as a proxy between these targets (this way we can
avoid a
> > > > circular dependency between rtc_base_generic and rtc_base_objc).
> > > > 
> > > > BUG=webrtc:7743
> > > > 
> > > > Review-Url: https://codereview.webrtc.org/3001623003
> > > > Cr-Commit-Position: refs/heads/master@{#19661}
> > > > Committed:
https://chromium.googlesource.com/external/webrtc/+/a0a5b98937de0f0bf90e17011...
> > > 
> > > TBR=kjellander@webrtc.org,kwiberg@webrtc.org
> > > # Skipping CQ checks because original CL landed less than 1 days ago.
> > > NOPRESUBMIT=true
> > > NOTREECHECKS=true
> > > NOTRY=true
> > > BUG=webrtc:7743
> > > 
> > > Review-Url: https://codereview.webrtc.org/3008103002
> > > Cr-Commit-Position: refs/heads/master@{#19664}
> > > Committed:
https://chromium.googlesource.com/external/webrtc/+/791a8f611da8cc6b0668ccc03...
> > 
> > TBR=kjellander@webrtc.org,kwiberg@webrtc.org
> > # Skipping CQ checks because original CL landed less than 1 days ago.
> > NOPRESUBMIT=true
> > NOTREECHECKS=true
> > NOTRY=true
> > BUG=webrtc:7743
> > 
> > Review-Url: https://codereview.webrtc.org/3004373002
> > Cr-Commit-Position: refs/heads/master@{#19681}
> > Committed:
https://chromium.googlesource.com/external/webrtc/+/a0c893f324da5c9f0773f8990...
> 
> TBR=kjellander@webrtc.org,kwiberg@webrtc.org
> # Skipping CQ checks because original CL landed less than 1 days ago.
> NOPRESUBMIT=true
> NOTREECHECKS=true
> NOTRY=true
> BUG=webrtc:7743
> 
> Review-Url: https://codereview.webrtc.org/3007183002
> Cr-Commit-Position: refs/heads/master@{#19688}
> Committed:
https://chromium.googlesource.com/external/webrtc/+/9c5e511806ef64c4cede52722...

TBR=kjellander@webrtc.org,kwiberg@webrtc.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=webrtc:7743



BUG=webrtc:7743
==========

to

==========
Reland of "Decoupling rtc_base from Obj-C code"

Reason for revert:
Trying to fix the link error on the chromium FYI trybots.

Original issue's description:
> Revert of Decoupling rtc_base from Obj-C code (patchset #1 id:1 of
https://codereview.webrtc.org/3004373002/ )
> 
> Reason for revert:
> It breaks a chromium FYI trybot (the switch from rtc_static_library to
rtc_source_set seems to be the problem).
> 
> Original issue's description:
> > Reland of Decoupling rtc_base from Obj-C code (patchset #1 id:1 of
https://codereview.webrtc.org/3008103002/ )
> > 
> > Reason for revert:
> > Trying to reland after fixing downstream project.
> > 
> > Original issue's description:
> > > Revert of Decoupling rtc_base from Obj-C code (patchset #8 id:140001 of
https://codereview.webrtc.org/3001623003/ )
> > > 
> > > Reason for revert:
> > > It breaks a downstream project.
> > > 
> > > Original issue's description:
> > > > Decoupling rtc_base from Obj-C code
> > > > 
> > > > The goal of this CL is to separate Obj-C/Obj-C++ code from targets which
have
> > > > also C++ code (see
https://bugs.chromium.org/p/webrtc/issues/detail?id=7743
> > > > for more information).
> > > > 
> > > > To achieve this we have created 2 targets (rtc_base_objc and
rtc_base_generic)
> > > > and rtc_base will act as a proxy between these targets (this way we can
avoid a
> > > > circular dependency between rtc_base_generic and rtc_base_objc).
> > > > 
> > > > BUG=webrtc:7743
> > > > 
> > > > Review-Url: https://codereview.webrtc.org/3001623003
> > > > Cr-Commit-Position: refs/heads/master@{#19661}
> > > > Committed:
https://chromium.googlesource.com/external/webrtc/+/a0a5b98937de0f0bf90e17011...
> > > 
> > > TBR=kjellander@webrtc.org,kwiberg@webrtc.org
> > > # Skipping CQ checks because original CL landed less than 1 days ago.
> > > NOPRESUBMIT=true
> > > NOTREECHECKS=true
> > > NOTRY=true
> > > BUG=webrtc:7743
> > > 
> > > Review-Url: https://codereview.webrtc.org/3008103002
> > > Cr-Commit-Position: refs/heads/master@{#19664}
> > > Committed:
https://chromium.googlesource.com/external/webrtc/+/791a8f611da8cc6b0668ccc03...
> > 
> > TBR=kjellander@webrtc.org,kwiberg@webrtc.org
> > # Skipping CQ checks because original CL landed less than 1 days ago.
> > NOPRESUBMIT=true
> > NOTREECHECKS=true
> > NOTRY=true
> > BUG=webrtc:7743
> > 
> > Review-Url: https://codereview.webrtc.org/3004373002
> > Cr-Commit-Position: refs/heads/master@{#19681}
> > Committed:
https://chromium.googlesource.com/external/webrtc/+/a0c893f324da5c9f0773f8990...
> 
> TBR=kjellander@webrtc.org,kwiberg@webrtc.org
> # Skipping CQ checks because original CL landed less than 1 days ago.
> NOPRESUBMIT=true
> NOTREECHECKS=true
> NOTRY=true
> BUG=webrtc:7743
> 
> Review-Url: https://codereview.webrtc.org/3007183002
> Cr-Commit-Position: refs/heads/master@{#19688}
> Committed:
https://chromium.googlesource.com/external/webrtc/+/9c5e511806ef64c4cede52722...

TBR=kjellander@webrtc.org,kwiberg@webrtc.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=webrtc:7743



BUG=webrtc:7743

Review-Url: https://codereview.webrtc.org/3012823002
Cr-Commit-Position: refs/heads/master@{#19726}
Committed:
https://chromium.googlesource.com/external/webrtc/+/310e32b92a588ed31dad63c3b...
==========

[commit-bot: I haz the power, 2017-09-07 14:36:35]

Committed patchset #5 (id:80001) as
https://chromium.googlesource.com/external/webrtc/+/310e32b92a588ed31dad63c3b...