Add video_loopback to gn.

webrtc/BUILD.gn

 # Copyright (c) 2014 The WebRTC project authors. All Rights Reserved.
 #
 # Use of this source code is governed by a BSD-style license
 # that can be found in the LICENSE file in the root of the source
 # tree. An additional intellectual property rights grant can be found
 # in the file PATENTS.  All contributing project authors may
 # be found in the AUTHORS file in the root of the source tree.
 
 # TODO(kjellander): Rebase this to webrtc/build/common.gypi changes after r6330.
 
@@ skipping 278 matching lines @@
   executable("webrtc_tests") {
     testonly = true
     deps = [
       ":webrtc",
       "modules/video_capture:video_capture_internal_impl",
       "test",
     ]
   }
 }
 
+if (!build_with_chromium) {
+  executable("video_loopback") {

[kjellander_webrtc, 2016/08/10 14:46:50]

Put this new target inside the existing if (!build_with_chromium) condition
above instead of a new one.

[stefan-webrtc, 2016/08/11 10:38:35]

Done.

+    testonly = true
+    sources = [
+      "test/mac/run_test.mm",
+      "test/run_test.cc",
+      "test/run_test.h",
+      "video/video_loopback.cc",
+    ]
+
+    if (is_mac) {
+      sources -= [ "test/run_test.cc" ]

[kjellander_webrtc, 2016/08/10 14:46:49]

GN disencourages use of negative source entries. Can you just flip it around so
that:

if (is_mac) {
  sources += [ "test/mac/run_test.mm" ]
} else {
  sources += [ "test/run_test.cc" ]
}

and then don't include "test/mac/run_test.mm", in the original sources entry
above.
Please make similar GYP changes too, to avoid diverging.

[stefan-webrtc, 2016/08/11 10:38:35]

Done.

+    }
+    deps = [
+      ":video_quality_test",
+      ":webrtc",

[kjellander_webrtc, 2016/08/10 14:46:50]

Is there a way to not depend on the monolith target "webrtc" ? I'm trying to get
rid of that long-term and it can usually be replaced by other targets.

[stefan-webrtc, 2016/08/11 10:38:35]

Changed to call:call

+      "modules/video_capture:video_capture_internal_impl",

[kjellander_webrtc, 2016/08/10 14:46:49]

Please add the added dependencies to GYP as well to avoid diverging (I guess
they're implicitly pulled in).

[stefan-webrtc, 2016/08/11 10:38:35]

Done.

+      "system_wrappers:metrics_default",
+      "test:field_trial",
+      "test:test_common",
+      "test:test_renderer",
+      "//testing/gmock",
+      "//testing/gtest",
+      "//third_party/gflags",
+    ]
+    if (is_clang && !is_nacl) {
+      # Suppress warnings from Chrome's Clang plugins.
+      # See http://code.google.com/p/webrtc/issues/detail?id=163 for details.
+      configs -= [ "//build/config/clang:find_bad_constructs" ]

[kjellander_webrtc, 2016/08/10 14:46:50]

Could you try fixing these warnings instead of having to suppress them? It's OK
to keep it if it's a lot of work but sometimes it's just a few code changes.

[stefan-webrtc, 2016/08/11 10:38:35]

I prefer not to in this case since it means I'll have to make interface changes
and add explicit constructors to all our send and receive stream configs etc.
It's a bit error prone, and this CL shouldn't really be touching APIs.

[kjellander_webrtc, 2016/08/15 11:27:38]

Fair enough.

+    }
+  }
+}
+
 source_set("webrtc_common") {
   sources = [
     "audio_sink.h",
     "common_types.cc",
     "common_types.h",
     "config.cc",
     "config.h",
     "engine_configurations.h",
     "typedefs.h",
   ]
@@ skipping 457 matching lines @@
     if (is_ios || (is_mac && target_cpu != "x86")) {
       defines = [ "CARBON_DEPRECATED=YES" ]
     }
     if (is_clang) {
       # Suppress warnings from the Chromium Clang plugin.
       # See http://code.google.com/p/webrtc/issues/detail?id=163 for details.
       configs -= [ "//build/config/clang:find_bad_constructs" ]
     }
   }
 }