rtc_base drags in a bunch of unwieldly dependencies (e.g. nss and
json) not required for standalone webrtc (aka rtc/media). The root of
the problem appears to be that MessageQueue depends on a socket server.
(And since common.h -> logging.h -> thread.h -> messagequeue.h, this
dependency spreads quickly.)
This starts a new target for a "purified" subset of rtc_base. It adds
the files which are already being used, replacing the use of common.h
with checks.h. desktop_capture is a lost cause, and retains its
dependency on the full rtc_base.
The hope is that as additional components are desired they will be
cleaned and added to rtc_base_approved.
BUG=3806
R=andresp@webrtc.org, henrike@webrtc.org
Review URL: https://webrtc-codereview.appspot.com/22649004
git-svn-id: http://webrtc.googlecode.com/svn/trunk@7188 4adac7df-926f-26a2-2b94-8c16560cd09d
- Change hw video decoder wrapper to allow to feed multiple input
and query for an output every 10 ms.
- Add an option to decode video frame into an Android surface object. Create
shared with video renderer EGL context and external texture on
video decoder thread.
- Support external texture rendering in Android renderer.
- Support TextureVideoFrame in Java and use it to pass texture from video decoder
to renderer.
- Fix HW encoder and decoder detection code to avoid query codec capabilities
from sw codecs.
BUG=
R=tkchin@webrtc.org
Review URL: https://webrtc-codereview.appspot.com/18299004
git-svn-id: http://webrtc.googlecode.com/svn/trunk@7185 4adac7df-926f-26a2-2b94-8c16560cd09d
1. the options set in the first CreateOffer call should not affect the result of a second CreateOffer call, if SetLocalDescription is not called after the first CreateOffer. So the member var options_ of MediaStreamSignaling is removed to make each CreateOffer independent.
Instead, MediaSession is responsible to make sure that an m-line in the current local description is never removed from the newly created offer.
2. OfferToReceiveAudio used to default to true, i.e. always having m=audio line even if no audio track. This is changed so that by default m=audio is only added if there are any audio tracks.
BUG=2108
R=pthatcher@webrtc.org
Committed: https://code.google.com/p/webrtc/source/detail?r=7068
Review URL: https://webrtc-codereview.appspot.com/16309004
git-svn-id: http://webrtc.googlecode.com/svn/trunk@7124 4adac7df-926f-26a2-2b94-8c16560cd09d
1. the options set in the first CreateOffer call should not affect the result of a second CreateOffer call, if SetLocalDescription is not called after the first CreateOffer. So the member var options_ of MediaStreamSignaling is removed to make each CreateOffer independent.
Instead, MediaSession is responsible to make sure that an m-line in the current local description is never removed from the newly created offer.
2. OfferToReceiveAudio used to default to true, i.e. always having m=audio line even if no audio track. This is changed so that by default m=audio is only added if there are any audio tracks.
BUG=2108
R=pthatcher@webrtc.org
Review URL: https://webrtc-codereview.appspot.com/16309004
git-svn-id: http://webrtc.googlecode.com/svn/trunk@7068 4adac7df-926f-26a2-2b94-8c16560cd09d
The bot that had the problem was using an old version of STL, so relanding.
> Revert 6863 "Refactor StatsCollector and associated types."
>
> Breaks chrome compilation on Mac:
>
> /Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX10.6.sdk/usr/include/c++/4.2.1/bits/vector.tcc:252:8:
> error: no matching constructor for initialization of
> 'webrtc::StatsReport'
> _Tp __x_copy = __x;
> ^ ~~~
> /Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX10.6.sdk/usr/include/c++/4.2.1/bits/stl_vector.h:608:4:
> note: in instantiation of member function
> 'std::vector<webrtc::StatsReport, std::allocator<webrtc::StatsReport>
> >::_M_insert_aux' requested here
> _M_insert_aux(end(), __x);
> ^
> ../../content/renderer/media/mock_peer_connection_impl.cc:282:11:
> note: in instantiation of member function
> 'std::vector<webrtc::StatsReport, std::allocator<webrtc::StatsReport>
> >::push_back' requested here
> reports.push_back(report1);
> ^
> ../../third_party/libjingle/source/talk/app/webrtc/statstypes.h:49:3:
> note: candidate constructor not viable: requires 0 arguments, but 1
> was provided
> StatsReport() : timestamp(0) {}
>
>
>
> > Refactor StatsCollector and associated types.
> > * Due to the type changes, I'm going to update the OnCompleted event in two phases to sync with Chrome. This is the first phase.
> > * Reports are now managed in a set, not a map, since it's enough to store the id in one place.
> > * Report ids are now const.
> > * Copying of data has been greatly reduced.
> > * This change includes preparation work for making GetStats fully async.
> >
> > This is a reland of r6778 which was reverted due to fyi bots failing.
> > I found and fixed the issue which was that in a couple of places I needed to replace a report instead of finding+updating an existing one.
> >
> > R=xians@webrtc.org
> >
> > Review URL: https://webrtc-codereview.appspot.com/15119004
>
> TBR=tommi@webrtc.org
>
> Review URL: https://webrtc-codereview.appspot.com/21169004TBR=niklas.enbom@webrtc.org
Review URL: https://webrtc-codereview.appspot.com/22099004
git-svn-id: http://webrtc.googlecode.com/svn/trunk@6908 4adac7df-926f-26a2-2b94-8c16560cd09d
Breaks chrome compilation on Mac:
/Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX10.6.sdk/usr/include/c++/4.2.1/bits/vector.tcc:252:8:
error: no matching constructor for initialization of
'webrtc::StatsReport'
_Tp __x_copy = __x;
^ ~~~
/Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX10.6.sdk/usr/include/c++/4.2.1/bits/stl_vector.h:608:4:
note: in instantiation of member function
'std::vector<webrtc::StatsReport, std::allocator<webrtc::StatsReport>
>::_M_insert_aux' requested here
_M_insert_aux(end(), __x);
^
../../content/renderer/media/mock_peer_connection_impl.cc:282:11:
note: in instantiation of member function
'std::vector<webrtc::StatsReport, std::allocator<webrtc::StatsReport>
>::push_back' requested here
reports.push_back(report1);
^
../../third_party/libjingle/source/talk/app/webrtc/statstypes.h:49:3:
note: candidate constructor not viable: requires 0 arguments, but 1
was provided
StatsReport() : timestamp(0) {}
> Refactor StatsCollector and associated types.
> * Due to the type changes, I'm going to update the OnCompleted event in two phases to sync with Chrome. This is the first phase.
> * Reports are now managed in a set, not a map, since it's enough to store the id in one place.
> * Report ids are now const.
> * Copying of data has been greatly reduced.
> * This change includes preparation work for making GetStats fully async.
>
> This is a reland of r6778 which was reverted due to fyi bots failing.
> I found and fixed the issue which was that in a couple of places I needed to replace a report instead of finding+updating an existing one.
>
> R=xians@webrtc.org
>
> Review URL: https://webrtc-codereview.appspot.com/15119004TBR=tommi@webrtc.org
Review URL: https://webrtc-codereview.appspot.com/21169004
git-svn-id: http://webrtc.googlecode.com/svn/trunk@6897 4adac7df-926f-26a2-2b94-8c16560cd09d
Some constraints, like kEchoCancellation, kMediaStreamAudioDucking are supported in Chrome but not in Libjingle, if the users set it in mandatory, LocalAudioSource::Initialize() will fail the getUserMedia call.
This patch fixes the problem by fully initializing the LocalAudioSource even though some constraints are not supported in libjingle.
BUT=crbug/398080
TEST=manual test:
var constraints = {audio: { mandatory: { googEchoCancellation: true } }};
getUserMedia(constraints, gotStream, gotStreamFailed);
verify you get a gotStream callback
R=henrika@webrtc.org, tommi@webrtc.org
Review URL: https://webrtc-codereview.appspot.com/21049004
git-svn-id: http://webrtc.googlecode.com/svn/trunk@6885 4adac7df-926f-26a2-2b94-8c16560cd09d
* Due to the type changes, I'm going to update the OnCompleted event in two phases to sync with Chrome. This is the first phase.
* Reports are now managed in a set, not a map, since it's enough to store the id in one place.
* Report ids are now const.
* Copying of data has been greatly reduced.
* This change includes preparation work for making GetStats fully async.
This is a reland of r6778 which was reverted due to fyi bots failing.
I found and fixed the issue which was that in a couple of places I needed to replace a report instead of finding+updating an existing one.
R=xians@webrtc.org
Review URL: https://webrtc-codereview.appspot.com/15119004
git-svn-id: http://webrtc.googlecode.com/svn/trunk@6863 4adac7df-926f-26a2-2b94-8c16560cd09d
* Due to the type changes, I'm going to update the OnCompleted event in two phases to sync with Chrome. This is the first phase.
* Reports are now managed in a set, not a map, since it's enough to store the id in one place.
* Report ids are now const.
* Copying of data has been greatly reduced.
* This change includes preparation work for making GetStats fully async.
(This is a reland of the original attempt in r6747)
R=xians@webrtc.org
Review URL: https://webrtc-codereview.appspot.com/18919004
git-svn-id: http://webrtc.googlecode.com/svn/trunk@6778 4adac7df-926f-26a2-2b94-8c16560cd09d
