Changed parser to identify review urls following Reviewed-on:

infra/tools/antibody/git_commit_parser.py

 # Copyright 2015 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.
 
 import datetime
 import dateutil.parser
 import pytz
 import re
 import subprocess
 
@@ skipping 61 matching lines @@
       tbr = [x.strip().split('@')[0] for x in tbr.split(',')]
     else:
       tbr = ['NOBODY']
   return tbr
 
 
 # TODO(keelerh): scan all review urls in a commit and compare the diffs to
 # identify the correct one
 def get_review_url(git_line):
   review_url = None
-  if re.match(r'^Review:.+$', git_line):
-    review_url = git_line[8:]
-  elif re.match(r'^Review URL:.+$', git_line):
+  if re.match(r'^Review: .+$', git_line):
+    review_url = git_line[7:]
+  elif re.match(r'^Review URL: .+$', git_line):
+    review_url = git_line[11:]
+  elif re.match(r'^Code review URL: .+$', git_line):
+    review_url = git_line[16:]
+  elif re.match(r'^Reviewed-on: .+$', git_line):
     review_url = git_line[12:]

[pgervais, 2015/07/23 18:50:55]

This is consistent but the number in git_line should match the number of
characters in the regexps. 
Example:
'^Review: .+$' essentially matches 'Review: ' at the beginning of a line
(because there is ^ at the beginning), with anything afterward. 'Review: ' is 8
characters long, so line 83 should read
review_url = git_line[8:]

Same for the other lines (just add one everywhere).

-  elif re.match(r'^Code review URL:.+$', git_line):
-    review_url = git_line[17:]
   return review_url
 
 
 def get_features_for_git_commit(git_commit):
   """Retrieves the git commit features
 
   Arg:
     git_commit(dict): a commit message parsed into a dictionary
 
   Return:
@@ skipping 103 matching lines @@
         WHERE type='tbr'
         GROUP BY git_commit_hash) tbr_count
       ON commit_people.git_commit_hash = tbr_count.git_commit_hash
       INNER JOIN git_commit
       ON commit_people.git_commit_hash = git_commit.hash
       WHERE tbr_count.c <> 0
       AND git_commit.review_url IS NOT NULL
       AND commit_people.type='author'""")
   commits_with_review_urls = cc.fetchall()
   return [x[0] for x in commits_with_review_urls]