Reland of Removing references to webrtc::PortAllocatorFactoryInterface.

Reland of Removing references to webrtc::PortAllocatorFactoryInterface.

Relanding after fixing conflict with:
https://codereview.chromium.org/1506383004

Original description:

Now using a new CreatePeerConnection method that takes a
PortAllocator as input. This removes the need for some
boilerplate code and code duplication between webrtc and chromium.
It also fixes an issue with TURN candidate priorities.

BUG=webrtc:5209
Committed: https://crrev.com/c563e691a061f90d9a52c4ebcea7157339b386d6
Cr-Commit-Position: refs/heads/master@{#365085}

[Taylor_Brandstetter, 2015-12-11 02:02:03]

deadbeef@chromium.org changed reviewers:
+ sergeyu@chromium.org, tommi@chromium.org

[Taylor_Brandstetter, 2015-12-11 02:02:03]

sergeyu@: Looks like I had bad timing hitting the CQ checkbox. :)
This is the same as the CL you just finished reviewing, but with a conflict with
remoting_me2me_host.cc fixed.

tommi@: If you don't mind, an extra pair of eyes on 
peer_connection_dependency_factory.cc wouldn't hurt.

[Sergey Ulanov, 2015-12-11 05:13:44]

lgtm.
When relanding reverted change it's usually preferred to reopen the original
review issue and reland from there. If the fix the trivial (like in this case)
then you can just land without waiting for another review cycle. If you do open
a separate issue then it's usually better to upload two separate patch sets, so
it's easy to see what was changed to fix the breakage.

[Taylor_Brandstetter, 2015-12-11 17:53:49]

On 2015/12/11 05:13:44, Sergey Ulanov wrote:
> lgtm.
> When relanding reverted change it's usually preferred to reopen the original
> review issue and reland from there. If the fix the trivial (like in this case)
> then you can just land without waiting for another review cycle. If you do
open
> a separate issue then it's usually better to upload two separate patch sets,
so
> it's easy to see what was changed to fix the breakage.

Ah, I didn't realize it was possible to reopen an issue. Thanks for pointing
that out; I'll do it next time.

[Sergey Ulanov, 2015-12-12 00:36:06]

On 2015/12/11 17:53:49, Taylor_Brandstetter wrote:
> On 2015/12/11 05:13:44, Sergey Ulanov wrote:
> > lgtm.
> > When relanding reverted change it's usually preferred to reopen the original
> > review issue and reland from there. If the fix the trivial (like in this
case)
> > then you can just land without waiting for another review cycle. If you do
> open
> > a separate issue then it's usually better to upload two separate patch sets,
> so
> > it's easy to see what was changed to fix the breakage.
> 
> Ah, I didn't realize it was possible to reopen an issue. Thanks for pointing
> that out; I'll do it next time.

can this be landed now or are you still waiting for something? I'm making other
changes in webrtc_transport and would like to avoid conflicts.

[Taylor_Brandstetter, 2015-12-14 19:12:50]

On 2015/12/12 00:36:06, Sergey Ulanov wrote:
> On 2015/12/11 17:53:49, Taylor_Brandstetter wrote:
> > On 2015/12/11 05:13:44, Sergey Ulanov wrote:
> > > lgtm.
> > > When relanding reverted change it's usually preferred to reopen the
original
> > > review issue and reland from there. If the fix the trivial (like in this
> case)
> > > then you can just land without waiting for another review cycle. If you do
> > open
> > > a separate issue then it's usually better to upload two separate patch
sets,
> > so
> > > it's easy to see what was changed to fix the breakage.
> > 
> > Ah, I didn't realize it was possible to reopen an issue. Thanks for pointing
> > that out; I'll do it next time.
> 
> can this be landed now or are you still waiting for something? I'm making
other
> changes in webrtc_transport and would like to avoid conflicts.

I'll land it today. I was partly waiting to see if tommi wanted to review, and
partly waiting until after the weekend, so I'd be available to fix anything that
broke.

[Taylor_Brandstetter, 2015-12-14 19:20:17]

The CQ bit was checked by deadbeef@chromium.org

[commit-bot: I haz the power, 2015-12-14 19:21:13]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1516163002/1
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1516163002/1

[commit-bot: I haz the power, 2015-12-14 20:30:26]

Committed patchset #1 (id:1)

[commit-bot: I haz the power, 2015-12-14 20:31:20]

Description was changed from

==========
Reland of Removing references to webrtc::PortAllocatorFactoryInterface.

Relanding after fixing conflict with:
https://codereview.chromium.org/1506383004

Original description:

Now using a new CreatePeerConnection method that takes a
PortAllocator as input. This removes the need for some
boilerplate code and code duplication between webrtc and chromium.
It also fixes an issue with TURN candidate priorities.

BUG=webrtc:5209
==========

to

==========
Reland of Removing references to webrtc::PortAllocatorFactoryInterface.

Relanding after fixing conflict with:
https://codereview.chromium.org/1506383004

Original description:

Now using a new CreatePeerConnection method that takes a
PortAllocator as input. This removes the need for some
boilerplate code and code duplication between webrtc and chromium.
It also fixes an issue with TURN candidate priorities.

BUG=webrtc:5209

Committed: https://crrev.com/c563e691a061f90d9a52c4ebcea7157339b386d6
Cr-Commit-Position: refs/heads/master@{#365085}
==========

[commit-bot: I haz the power, 2015-12-14 20:31:21]

Patchset 1 (id:??) landed as
https://crrev.com/c563e691a061f90d9a52c4ebcea7157339b386d6
Cr-Commit-Position: refs/heads/master@{#365085}