Adding protobuf headers that will be used by upcoming continuous speech recognition classes (Speech…

content/content.gyp

 # Copyright (c) 2012 The Chromium Authors. All rights reserved.
 # Use of this source code is governed by a BSD-style license that can be
 # found in the LICENSE file.
 
 {
   'variables': {
     'chromium_code': 1,  # Use higher warning level.
     'directxsdk_exists': '<!(python <(DEPTH)/build/dir_exists.py ../third_party/directxsdk)',
   },
   'includes': [
     '../build/win_precompile.gypi',
+    'browser/speech/proto/speech_proto.gypi',

[jam, 2012/05/15 21:17:01]

why do you need to include this in both content.gyp and content_browser.gypi?

[Primiano Tucci (use gerrit), 2012/05/16 08:43:39]

The file is included only in content.gyp. The speech_proto.gypi defines a gyp
target section, and it is my understanding that I cannot move this include
directly into content_browser.gypi because content_browser.gypi is included
within a target itself.
The other line inside content_browser.gypi is just a dependency on speech_proto,
which requires the corresponding target to be previously defined.

[hans, 2012/05/16 10:42:47]

Maybe speech_proto.gypi should actually be speec_proto.gyp? Then we could add a
dependency on it in chrome_browser.gypi. I'm not an expert on this gyp stuff,
though.

[Primiano Tucci (use gerrit), 2012/05/16 13:34:19]

Nice! Seems to work requiring no edits to content.gyp.

     'content_shell.gypi',
     'content_tests.gypi',
   ],
   'target_defaults': {
     'defines': ['CONTENT_IMPLEMENTATION'],
   },
   'conditions': [
    # In component mode, we build all of content as a single DLL.
    # However, in the static mode, we need to build content as multiple
    # targets in order to prevent dependencies from getting introduced
@@ skipping 189 matching lines @@
        },
        {
          'target_name': 'content_jni_headers',
          'type': 'none',
          'includes': [ 'content_jni.gypi' ],
        },
      ],
    }],  # OS == "android"
   ],
 }