Chromium Code Reviews
|
|
Chromium Code Reviews
Issue
2944643002:
Support building WebRTC without audio and video for IOS (Closed)
|
Created:
3 years, 6 months ago by Zhi Huang Modified:
3 years, 5 months ago Reviewers:
Zach Stein, tommi, Chuck, magjed_webrtc, chlady, Taylor Brandstetter, tkchin_webrtc, pthatcher1 CC:
webrtc-reviews_webrtc.org, tterriberry_mozilla.com Target Ref:
refs/heads/master Project:
webrtc Visibility:
Public. |
DescriptionSupport building WebRTC without audio and video for IOS.
Reorganized the targets in webrtc/sdk/BUILD.gn so that the applications which use
WebRTC DataChannel only can depend on the "peerconnection_factory_no_media"
instead of "rtc_sdk_objc" to reduce the binary size.
Provided a no-media implementation of RTCPeerConnectionFactory using the macro
"HAVE_NO_MEDIA".
BUG=webrtc:7613
Review-Url: https://codereview.webrtc.org/2944643002
Cr-Commit-Position: refs/heads/master@{#18819}
Committed: https://chromium.googlesource.com/external/webrtc/+/a4c113afe1378f00b791971b0ce28f4a06f312ac
Patch Set 1 : Hacky build #Patch Set 2 : Moduar WebRTC for Objc. #Patch Set 3 : Cleaner solution. #
Total comments: 1
Patch Set 4 : Reponse to comments. #
Total comments: 9
Patch Set 5 : Simpler solution. #Patch Set 6 : Rebase. Modified the BUILD. #Patch Set 7 : Add comments to BUILD.gn #
Total comments: 15
Patch Set 8 : Format #
Messages
Total messages: 54 (38 generated)
Patchset #1 (id:1) has been deleted
Description was changed from ========== Hacky build BUG=None ========== to ========== Build Datachannel only Objc. BUG=None ==========
Patchset #1 (id:20001) has been deleted
Patchset #1 (id:40001) has been deleted
zhihuang@webrtc.org changed reviewers: + deadbeef@webrtc.org, zstein@webrtc.org
FYI.
Patchset #2 (id:80001) has been deleted
Patchset #1 (id:60001) has been deleted
Patchset #3 (id:140001) has been deleted
Patchset #3 (id:160001) has been deleted
zhihuang@webrtc.org changed reviewers: + magjed@webrtc.org, pthatcher@webrtc.org, tkchin@webrtc.org
Hi everyone,
This is a CL to support building DataChannelOnly WebRTC with Objc wrapper. The
idea here is to have a base target (objc_peerconnection_base) and some A/V
modules.
I know the compile-time polymorphism is not ideal so I try to make that part as
small as possible. It seems that at least we need two implementations for the
methods that create mediastream, videotrack, or videosource. because the there
is a deps chain:
objc_peerconnection:RTCMediaStream // Need null impl
objc_PeerConnection:RTCVideoTrack // Need null impl
objc_PeerConnection:RTCVideoSource
objc_video:ObjcVideoTrackSource // Shouldn’t depend on this
... // more video stuff
We need to do something for those places we draw the line.
Not sure if this approach is good enough to unblock other projects.
Feel free to leave comments.
Thanks.
https://codereview.webrtc.org/2944643002/diff/180001/webrtc/sdk/objc/Framewor... File webrtc/sdk/objc/Framework/Classes/PeerConnection/RTCPeerConnectionFactory+Media.mm (right): https://codereview.webrtc.org/2944643002/diff/180001/webrtc/sdk/objc/Framewor... webrtc/sdk/objc/Framework/Classes/PeerConnection/RTCPeerConnectionFactory+Media.mm:73: #if defined(HAVE_RTC_AUDIO_) || defined(HAVE_RTC_VIDEO) Would we need this with just audio?
https://codereview.webrtc.org/2944643002/diff/200001/webrtc/sdk/objc/Framewor... File webrtc/sdk/objc/Framework/Classes/PeerConnection/RTCPeerConnection+Media.mm (right): https://codereview.webrtc.org/2944643002/diff/200001/webrtc/sdk/objc/Framewor... webrtc/sdk/objc/Framework/Classes/PeerConnection/RTCPeerConnection+Media.mm:15: why does this need to be in its own file? this file is also supposed to implement the Media category, but it's only used to implement C++ methods? https://codereview.webrtc.org/2944643002/diff/200001/webrtc/sdk/objc/Framewor... File webrtc/sdk/objc/Framework/Classes/PeerConnection/RTCPeerConnectionFactory+Media.mm (right): https://codereview.webrtc.org/2944643002/diff/200001/webrtc/sdk/objc/Framewor... webrtc/sdk/objc/Framework/Classes/PeerConnection/RTCPeerConnectionFactory+Media.mm:31: rtc::scoped_refptr<webrtc::PeerConnectionFactoryInterface> CreatePeerConnectionFactory( why is this method declared only if !defined(HAVE_RTC_AUDIO) && !defined(HAVE_RTC_VIDEO)? Shouldn't this also be an if / else block inside the init method? https://codereview.webrtc.org/2944643002/diff/200001/webrtc/sdk/objc/Framewor... webrtc/sdk/objc/Framework/Classes/PeerConnection/RTCPeerConnectionFactory+Media.mm:54: @implementation RTCPeerConnectionFactory (Media) same question: why does this need to be in its own file? https://codereview.webrtc.org/2944643002/diff/200001/webrtc/sdk/objc/Framewor... File webrtc/sdk/objc/Framework/Classes/PeerConnection/RTCPeerConnectionFactory+Private.h (right): https://codereview.webrtc.org/2944643002/diff/200001/webrtc/sdk/objc/Framewor... webrtc/sdk/objc/Framework/Classes/PeerConnection/RTCPeerConnectionFactory+Private.h:19: std::unique_ptr<rtc::Thread> _networkThread; style: don't declare ivars in @interface decls If you need them shared, declare them as properties instead https://codereview.webrtc.org/2944643002/diff/200001/webrtc/sdk/objc/Framewor... File webrtc/sdk/objc/Framework/Classes/PeerConnection/RTCPeerConnectionFactory+Video.mm (right): https://codereview.webrtc.org/2944643002/diff/200001/webrtc/sdk/objc/Framewor... webrtc/sdk/objc/Framework/Classes/PeerConnection/RTCPeerConnectionFactory+Video.mm:28: @implementation RTCPeerConnectionFactory (Video) ditto: why separate file? https://codereview.webrtc.org/2944643002/diff/200001/webrtc/sdk/objc/Framewor... webrtc/sdk/objc/Framework/Classes/PeerConnection/RTCPeerConnectionFactory+Video.mm:54: return [[RTCVideoTrack alloc] initWithFactory:self source:source trackId:trackId]; for the generic case, why do we need to special case it? Since we have a thin wrapper, it should just return whatever the C++ impl returns https://codereview.webrtc.org/2944643002/diff/200001/webrtc/sdk/objc/Framewor... File webrtc/sdk/objc/Framework/Classes/PeerConnection/RTCPeerConnectionFactory.mm (left): https://codereview.webrtc.org/2944643002/diff/200001/webrtc/sdk/objc/Framewor... webrtc/sdk/objc/Framework/Classes/PeerConnection/RTCPeerConnectionFactory.mm:37: - (instancetype)init { designated initializer *must* be implemented in the this file categories are not good places to put designated initializers https://codereview.webrtc.org/2944643002/diff/200001/webrtc/sdk/objc/Framewor... File webrtc/sdk/objc/Framework/Headers/WebRTC/RTCPeerConnectionFactory.h (right): https://codereview.webrtc.org/2944643002/diff/200001/webrtc/sdk/objc/Framewor... webrtc/sdk/objc/Framework/Headers/WebRTC/RTCPeerConnectionFactory.h:61: - (instancetype)init; do not place designated ctors in category
Patchset #5 (id:220001) has been deleted
Patchset #5 (id:240001) has been deleted
Description was changed from ========== Build Datachannel only Objc. BUG=None ========== to ========== Support building WebRTC without audio and video for IOS BUG=webrtc:7613 ==========
PTAL. This is a pretty simple change. Only two 2 files and < 200 lines of delta.
The CQ bit was checked by zhihuang@webrtc.org to run a CQ dry run
Dry run: CQ is trying da patch. Follow status at: https://chromium-cq-status.appspot.com/v2/patch-status/codereview.webrtc.org/...
Description was changed from ========== Support building WebRTC without audio and video for IOS BUG=webrtc:7613 ========== to ========== Support building WebRTC without audio and video for IOS Refactored the target structures. BUG=webrtc:7613 ==========
Description was changed from ========== Support building WebRTC without audio and video for IOS Refactored the target structures. BUG=webrtc:7613 ========== to ========== Support building WebRTC without audio and video for IOS Reorganized the targets in webrtc/sdk/BUILD.gn. BUG=webrtc:7613 ==========
The CQ bit was unchecked by commit-bot@chromium.org
Dry run: This issue passed the CQ dry run.
If the goal is to reduce the binary size for "data-only" applications with as simple a change as possible, while we figure out the long term "slim/modular" plan, this change (patchset 6) seems reasonable. Though before landing it, the BUILD.gn could use some comments. It wouldn't be clear to a casual reader why "objc_videotracksource" is split out of "objc_video", or why "objc_peerconnectionfactory_no_media" still brings in most of the audio/video-specific sources. https://codereview.webrtc.org/2944643002/diff/200001/webrtc/sdk/objc/Framewor... File webrtc/sdk/objc/Framework/Classes/PeerConnection/RTCPeerConnectionFactory.mm (left): https://codereview.webrtc.org/2944643002/diff/200001/webrtc/sdk/objc/Framewor... webrtc/sdk/objc/Framework/Classes/PeerConnection/RTCPeerConnectionFactory.mm:37: - (instancetype)init { On 2017/06/20 21:51:24, tkchin_webrtc wrote: > designated initializer *must* be implemented in the this file > > categories are not good places to put designated initializers If I understand designated initializers correctly, the right way to do this would be to have a designated initializer that takes modules as input and calls CreateModularPeerConnectionFactory, and have the other initializers call that.
On 2017/06/25 17:41:56, Taylor Brandstetter wrote: > If the goal is to reduce the binary size for "data-only" applications with as > simple a change as possible, while we figure out the long term "slim/modular" > plan, this change (patchset 6) seems reasonable. > > Though before landing it, the BUILD.gn could use some comments. It wouldn't be > clear to a casual reader why "objc_videotracksource" is split out of > "objc_video", or why "objc_peerconnectionfactory_no_media" still brings in most > of the audio/video-specific sources. > > https://codereview.webrtc.org/2944643002/diff/200001/webrtc/sdk/objc/Framewor... > File > webrtc/sdk/objc/Framework/Classes/PeerConnection/RTCPeerConnectionFactory.mm > (left): > > https://codereview.webrtc.org/2944643002/diff/200001/webrtc/sdk/objc/Framewor... > webrtc/sdk/objc/Framework/Classes/PeerConnection/RTCPeerConnectionFactory.mm:37: > - (instancetype)init { > On 2017/06/20 21:51:24, tkchin_webrtc wrote: > > designated initializer *must* be implemented in the this file > > > > categories are not good places to put designated initializers > > If I understand designated initializers correctly, the right way to do this > would be to have a designated initializer that takes modules as input and calls > CreateModularPeerConnectionFactory, and have the other initializers call that. After talking to Zeke, we agreed that we should try to mapping the native C++ structure in Objc layer. Since the audiotrack.cc, videotrack.cc etc. are included in native pc:peerconnection, I did the same thing for objc wrapper. BTW, I added some comments to make it clear. PTAL.
On 2017/06/26 19:26:43, Zhi Huang wrote: > On 2017/06/25 17:41:56, Taylor Brandstetter wrote: > > If the goal is to reduce the binary size for "data-only" applications with as > > simple a change as possible, while we figure out the long term "slim/modular" > > plan, this change (patchset 6) seems reasonable. > > > > Though before landing it, the BUILD.gn could use some comments. It wouldn't be > > clear to a casual reader why "objc_videotracksource" is split out of > > "objc_video", or why "objc_peerconnectionfactory_no_media" still brings in > most > > of the audio/video-specific sources. > > > > > https://codereview.webrtc.org/2944643002/diff/200001/webrtc/sdk/objc/Framewor... > > File > > webrtc/sdk/objc/Framework/Classes/PeerConnection/RTCPeerConnectionFactory.mm > > (left): > > > > > https://codereview.webrtc.org/2944643002/diff/200001/webrtc/sdk/objc/Framewor... > > > webrtc/sdk/objc/Framework/Classes/PeerConnection/RTCPeerConnectionFactory.mm:37: > > - (instancetype)init { > > On 2017/06/20 21:51:24, tkchin_webrtc wrote: > > > designated initializer *must* be implemented in the this file > > > > > > categories are not good places to put designated initializers > > > > If I understand designated initializers correctly, the right way to do this > > would be to have a designated initializer that takes modules as input and > calls > > CreateModularPeerConnectionFactory, and have the other initializers call that. > > After talking to Zeke, we agreed that we should try to mapping the native C++ > structure in Objc layer. > Since the audiotrack.cc, videotrack.cc etc. are included in native > pc:peerconnection, I did the same thing for objc wrapper. > > BTW, I added some comments to make it clear. PTAL. Hello Magus, Could you please take a look since Zeke is OOO until July 9th? Thanks!
tkchin@webrtc.org changed reviewers: + chlady@webrtc.org, haysc@webrtc.org
Chris / Chuck, can one of you take a look as well? The goal here is to keep ObjC a thin wrapper around the C++, whereby we rely (almost) entirely on C++ behavior. The only gotcha is stuff that is ObjC-only, but that should be of secondary importance in the goal to reduce binary size since the ObjC-only media bits are small compared to the C++ bits.
deadbeef@webrtc.org changed reviewers: + tommi@webrtc.org
Both the webrtc/sdk/objc owners are OOO now... So should we land with TBR? Or get review from a higher level owner (+tommi)?
looks good generally, I'd like to format things a little better. git cl format should take care of that for the most part but for the gn file I'd like to make the comments a little more readable so I commented on those. Once that's done, then this lgtm. https://codereview.webrtc.org/2944643002/diff/300001/webrtc/sdk/BUILD.gn File webrtc/sdk/BUILD.gn (right): https://codereview.webrtc.org/2944643002/diff/300001/webrtc/sdk/BUILD.gn#newc... webrtc/sdk/BUILD.gn:107: # build the PeerConnectionFactory without audio/video support. As the comment is currently written, I had some problems parsing it. What about: # Keep videotracksource related code in a separate target so that we can build # PeerConnectionFactory without audio/video support when that's called for. Also, I think it would be nice if we could keep comment lines limited to 80 columns to make them more readable in code reviews. https://codereview.webrtc.org/2944643002/diff/300001/webrtc/sdk/BUILD.gn#newc... webrtc/sdk/BUILD.gn:131: # Suppress warnings from the Chromium Clang plugin (bugs.webrtc.org/163). nit: keep shorter than 80 characters https://codereview.webrtc.org/2944643002/diff/300001/webrtc/sdk/BUILD.gn#newc... webrtc/sdk/BUILD.gn:273: # Suppress warnings from the Chromium Clang plugin (bugs.webrtc.org/163). <80 https://codereview.webrtc.org/2944643002/diff/300001/webrtc/sdk/BUILD.gn#newc... webrtc/sdk/BUILD.gn:309: # Suppress warnings from the Chromium Clang plugin (bugs.webrtc.org/163). <80 https://codereview.webrtc.org/2944643002/diff/300001/webrtc/sdk/BUILD.gn#newc... webrtc/sdk/BUILD.gn:327: # Build the PeerConnectionFactory without audio/video support. This target depends on the <80 (here is a good example of how it's easier to read the comment if we wrap at 80) https://codereview.webrtc.org/2944643002/diff/300001/webrtc/sdk/BUILD.gn#newc... webrtc/sdk/BUILD.gn:342: # Suppress warnings from the Chromium Clang plugin (bugs.webrtc.org/163). ditto https://codereview.webrtc.org/2944643002/diff/300001/webrtc/sdk/objc/Framewor... File webrtc/sdk/objc/Framework/Classes/PeerConnection/RTCPeerConnectionFactory.mm (right): https://codereview.webrtc.org/2944643002/diff/300001/webrtc/sdk/objc/Framewor... webrtc/sdk/objc/Framework/Classes/PeerConnection/RTCPeerConnectionFactory.mm:28: // The no-media version PeerConnectionFactory doesn't depend on media related C++ target. please run |git cl format| https://codereview.webrtc.org/2944643002/diff/300001/webrtc/sdk/objc/Framewor... webrtc/sdk/objc/Framework/Classes/PeerConnection/RTCPeerConnectionFactory.mm:59: nullptr /*default_adm*/, can we use // comment style? nullptr, // default_adm (etc)
The CQ bit was checked by zhihuang@webrtc.org to run a CQ dry run
Dry run: CQ is trying da patch. Follow status at: https://chromium-cq-status.appspot.com/v2/patch-status/codereview.webrtc.org/...
The CQ bit was unchecked by commit-bot@chromium.org
Dry run: This issue passed the CQ dry run.
Patchset #8 (id:320001) has been deleted
The CQ bit was checked by zhihuang@webrtc.org
The patchset sent to the CQ was uploaded after l-g-t-m from tommi@webrtc.org Link to the patchset: https://codereview.webrtc.org/2944643002/#ps340001 (title: "Format")
The CQ bit was unchecked by zhihuang@webrtc.org
Description was changed from ========== Support building WebRTC without audio and video for IOS Reorganized the targets in webrtc/sdk/BUILD.gn. BUG=webrtc:7613 ========== to ========== Support building WebRTC without audio and video for IOS Reorganized the targets in webrtc/sdk/BUILD.gn so that the applications which use WebRTC DataChannel only can depend on the peerconnection_factory_no_media instead of rtc_sdk_objc to reduce the binary size. Provided a no-media implementation of RTCPeerConnectionFactory using the macro. BUG=webrtc:7613 ==========
Description was changed from ========== Support building WebRTC without audio and video for IOS Reorganized the targets in webrtc/sdk/BUILD.gn so that the applications which use WebRTC DataChannel only can depend on the peerconnection_factory_no_media instead of rtc_sdk_objc to reduce the binary size. Provided a no-media implementation of RTCPeerConnectionFactory using the macro. BUG=webrtc:7613 ========== to ========== Support building WebRTC without audio and video for IOS Reorganized the targets in webrtc/sdk/BUILD.gn so that the applications which use WebRTC DataChannel only can depend on the "peerconnection_factory_no_media" instead of "rtc_sdk_objc" to reduce the binary size. Provided a no-media implementation of RTCPeerConnectionFactory using the macro. BUG=webrtc:7613 ==========
Description was changed from ========== Support building WebRTC without audio and video for IOS Reorganized the targets in webrtc/sdk/BUILD.gn so that the applications which use WebRTC DataChannel only can depend on the "peerconnection_factory_no_media" instead of "rtc_sdk_objc" to reduce the binary size. Provided a no-media implementation of RTCPeerConnectionFactory using the macro. BUG=webrtc:7613 ========== to ========== Support building WebRTC without audio and video for IOS Reorganized the targets in webrtc/sdk/BUILD.gn so that the applications which use WebRTC DataChannel only can depend on the "peerconnection_factory_no_media" instead of "rtc_sdk_objc" to reduce the binary size. Provided a no-media implementation of RTCPeerConnectionFactory using the macro "HAVE_NO_MEDIA". BUG=webrtc:7613 ==========
Description was changed from ========== Support building WebRTC without audio and video for IOS Reorganized the targets in webrtc/sdk/BUILD.gn so that the applications which use WebRTC DataChannel only can depend on the "peerconnection_factory_no_media" instead of "rtc_sdk_objc" to reduce the binary size. Provided a no-media implementation of RTCPeerConnectionFactory using the macro "HAVE_NO_MEDIA". BUG=webrtc:7613 ========== to ========== Support building WebRTC without audio and video for IOS. Reorganized the targets in webrtc/sdk/BUILD.gn so that the applications which use WebRTC DataChannel only can depend on the "peerconnection_factory_no_media" instead of "rtc_sdk_objc" to reduce the binary size. Provided a no-media implementation of RTCPeerConnectionFactory using the macro "HAVE_NO_MEDIA". BUG=webrtc:7613 ==========
The CQ bit was checked by zhihuang@webrtc.org
CQ is trying da patch. Follow status at: https://chromium-cq-status.appspot.com/v2/patch-status/codereview.webrtc.org/...
Thanks for the quick review Tommi! https://codereview.webrtc.org/2944643002/diff/300001/webrtc/sdk/BUILD.gn File webrtc/sdk/BUILD.gn (right): https://codereview.webrtc.org/2944643002/diff/300001/webrtc/sdk/BUILD.gn#newc... webrtc/sdk/BUILD.gn:107: # build the PeerConnectionFactory without audio/video support. On 2017/06/28 15:06:44, tommi (wëbrtc) wrote: > As the comment is currently written, I had some problems parsing it. What > about: > > # Keep videotracksource related code in a separate target so that we can build > # PeerConnectionFactory without audio/video support when that's called for. > > Also, I think it would be nice if we could keep comment lines limited to 80 > columns to make them more readable in code reviews. Your suggestion is better. Done. https://codereview.webrtc.org/2944643002/diff/300001/webrtc/sdk/BUILD.gn#newc... webrtc/sdk/BUILD.gn:131: # Suppress warnings from the Chromium Clang plugin (bugs.webrtc.org/163). On 2017/06/28 15:06:45, tommi (wëbrtc) wrote: > nit: keep shorter than 80 characters Done. https://codereview.webrtc.org/2944643002/diff/300001/webrtc/sdk/BUILD.gn#newc... webrtc/sdk/BUILD.gn:273: # Suppress warnings from the Chromium Clang plugin (bugs.webrtc.org/163). On 2017/06/28 15:06:44, tommi (wëbrtc) wrote: > <80 Done. https://codereview.webrtc.org/2944643002/diff/300001/webrtc/sdk/BUILD.gn#newc... webrtc/sdk/BUILD.gn:309: # Suppress warnings from the Chromium Clang plugin (bugs.webrtc.org/163). On 2017/06/28 15:06:45, tommi (wëbrtc) wrote: > <80 Done. https://codereview.webrtc.org/2944643002/diff/300001/webrtc/sdk/BUILD.gn#newc... webrtc/sdk/BUILD.gn:342: # Suppress warnings from the Chromium Clang plugin (bugs.webrtc.org/163). On 2017/06/28 15:06:44, tommi (wëbrtc) wrote: > ditto Done. https://codereview.webrtc.org/2944643002/diff/300001/webrtc/sdk/objc/Framewor... File webrtc/sdk/objc/Framework/Classes/PeerConnection/RTCPeerConnectionFactory.mm (right): https://codereview.webrtc.org/2944643002/diff/300001/webrtc/sdk/objc/Framewor... webrtc/sdk/objc/Framework/Classes/PeerConnection/RTCPeerConnectionFactory.mm:28: // The no-media version PeerConnectionFactory doesn't depend on media related C++ target. On 2017/06/28 15:06:45, tommi (wëbrtc) wrote: > please run |git cl format| This is what I got after git cl format. It seems that the style guide are different for objc file. The column limitation is 100. But I can manually make it 80 since this is c++ code. https://codereview.webrtc.org/2944643002/diff/300001/webrtc/sdk/objc/Framewor... webrtc/sdk/objc/Framework/Classes/PeerConnection/RTCPeerConnectionFactory.mm:59: nullptr /*default_adm*/, On 2017/06/28 15:06:45, tommi (wëbrtc) wrote: > can we use // comment style? > > nullptr, // default_adm > (etc) Yeah, sure. Done.
The CQ bit was unchecked by commit-bot@chromium.org
Try jobs failed on following builders: linux_tsan2 on master.tryserver.webrtc (JOB_FAILED, http://build.chromium.org/p/tryserver.webrtc/builders/linux_tsan2/builds/23756)
The CQ bit was checked by zhihuang@webrtc.org
CQ is trying da patch. Follow status at: https://chromium-cq-status.appspot.com/v2/patch-status/codereview.webrtc.org/...
CQ is committing da patch.
Bot data: {"patchset_id": 340001, "attempt_start_ts": 1498682474039680,
"parent_rev": "9588682dfe13afef51cdafb2abc30e720e382eca", "commit_rev":
"a4c113afe1378f00b791971b0ce28f4a06f312ac"}
Message was sent while issue was closed.
Description was changed from ========== Support building WebRTC without audio and video for IOS. Reorganized the targets in webrtc/sdk/BUILD.gn so that the applications which use WebRTC DataChannel only can depend on the "peerconnection_factory_no_media" instead of "rtc_sdk_objc" to reduce the binary size. Provided a no-media implementation of RTCPeerConnectionFactory using the macro "HAVE_NO_MEDIA". BUG=webrtc:7613 ========== to ========== Support building WebRTC without audio and video for IOS. Reorganized the targets in webrtc/sdk/BUILD.gn so that the applications which use WebRTC DataChannel only can depend on the "peerconnection_factory_no_media" instead of "rtc_sdk_objc" to reduce the binary size. Provided a no-media implementation of RTCPeerConnectionFactory using the macro "HAVE_NO_MEDIA". BUG=webrtc:7613 Review-Url: https://codereview.webrtc.org/2944643002 Cr-Commit-Position: refs/heads/master@{#18819} Committed: https://chromium.googlesource.com/external/webrtc/+/a4c113afe1378f00b791971b0... ==========
Message was sent while issue was closed.
Committed patchset #8 (id:340001) as https://chromium.googlesource.com/external/webrtc/+/a4c113afe1378f00b791971b0... |
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
