Add a tool to fix (some) errors reported by gn gen --check.

tools/gn_check_autofix.py

Index: tools/gn_check_autofix.py
diff --git a/tools/gn_check_autofix.py b/tools/gn_check_autofix.py
new file mode 100644
index 0000000000000000000000000000000000000000..f06504bbd4b007d0633ed51680d4cfb7ee06d02a
--- /dev/null
+++ b/tools/gn_check_autofix.py
@@ -0,0 +1,79 @@
+#!/usr/bin/env python
+
+# Copyright (c) 2016 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.
+
+import os

[kjellander_webrtc, 2016/12/07 12:31:38]

Add module documentation that at least mentions that this won't change existing
deps entries but rather add a deps += at the end of the target.

+import pprint
+import re
+import subprocess
+import sys
+
+from collections import defaultdict
+

[kjellander_webrtc, 2016/12/07 12:31:38]

nit: Two blank lines between top-level definitions, one blank line between
method definitions.
https://google.github.io/styleguide/pyguide.html?showone=Blank_Lines#Blank_Lines

+def Run(cmd):
+  sub = subprocess.Popen(cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
+  return sub.communicate()
+
+def fix_missing(filename, missing):

[kjellander_webrtc, 2016/12/07 12:31:38]

Rename to FixMissing, according to
http://dev.chromium.org/chromium-os/python-style-guidelines
(one of two exceptions from the Google Python Style guide that Chromium uses):
"Naming: Functions and Method names use CapWords() style, not
lower_with_under()"

+  with open(filename) as f:
+    contents = f.readlines()
+  target_re = re.compile(r'(?P<indentation_level>\s*)\w*\("(?P<target_name>\w*)"\) {$')

[kjellander_webrtc, 2016/12/07 12:31:38]

Wrap at column 80 

+  blocks = ""

[kjellander_webrtc, 2016/12/07 12:31:38]

Use single-quotes consistently for strings. 
From http://google.github.io/styleguide/pyguide.html?showone=Strings#Strings:
"Be consistent with your choice of string quote character within a file. Pick '
or " and stick with it. "

+  seen_deps = False
+  last_indentation_level = None
+  for line in contents:
+    match = target_re.match(line)
+    if match:
+      last_target = match.group('target_name')
+      if last_target in missing:
+        last_indentation_level = match.group('indentation_level')
+    elif last_indentation_level is not None:
+      match = re.match(last_indentation_level + '}$', line)
+      if match:
+        deps = ((['deps += [\n'] if seen_deps else ['deps = [']) +
+                map(lambda x: '  "' + x + '",\n', list(missing[last_target])) +
+                [']\n'])
+        deps = map(lambda x: last_indentation_level + '  '  + x, deps)
+        deps = ''.join(deps)
+        blocks += deps
+        seen_deps = False
+        last_indentation_level = None
+      elif line.strip().startswith('deps'):
+        seen_deps = True
+    blocks += line
+  with open(filename, 'w') as f:
+    f.write(blocks)
+  Run(['gn', 'format', filename])
+
+missing = defaultdict(lambda: defaultdict(set))

[kjellander_webrtc, 2016/12/07 12:31:38]

Use a main() function instead:
http://google.github.io/styleguide/pyguide.html?showone=Main#Main

+
+errors = Run(['gn', 'gen', 'out/gn', '--check'])[0]
+errors = errors.split('ERROR at //')[1:]
+for error in errors:
+  error = error.splitlines()
+  target_msg = 'The target:'
+  if target_msg not in error:
+    # print '\n'.join(error)
+    continue
+  index = error.index(target_msg) + 1
+  path, target = error[index].strip().split(':')
+  path = os.path.join(path[2:], 'BUILD.gn')
+  dep = error[index+2].strip()
+  if 'rtc_base' in dep and not 'rtc_base_approved' in dep:

[kjellander_webrtc, 2016/12/07 12:31:38]

Let's remove this as we decided to drop the PRESUBMIT check due to the ongoing
refactoring work for base/

+    print '\n'.join(error)
+    continue
+  if ':' not in dep:
+    print '\n'.join(error)
+    continue
+  if target == "audio_decoder_unittests_bundle_data":

[kjellander_webrtc, 2016/12/07 12:31:38]

Is this needed? I suggest keeping hardcoded assumptions out of this script. Some
manual work is to be expected even after running this script anyway.

+    print path, target, dep
+  missing[path][target].add(dep)
+
+for path, missing_deps in missing.items():
+  fix_missing(path, missing_deps)