[Findit] Improve output format and cherry-pick bugs fix.

tools/findit/git_repository_parser.py

 # Copyright (c) 2014 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 base64
 import xml.dom.minidom as minidom
 from xml.parsers.expat import ExpatError
 
 import crash_utils
 from repository_parser_interface import ParserInterface
 
 FILE_CHANGE_TYPE_MAP = {
     'add': 'A',
+    'copy': 'C',
     'delete': 'D',
-    'modify': 'M'
+    'modify': 'M',
+    'rename': 'R'
 }
 
 
+def _ConvertToFileChangeType(file_action):
+  # TODO(stgao): verify impact on code that checks the file change type.
+  return file_action[0].upper()
+
+
 class GitParser(ParserInterface):
   """Parser for Git repository in googlesource.
 
   Attributes:
     parsed_deps: A map from component path to its repository name, regression,
                  etc.
     url_parts_map: A map from url type to its url parts. This parts are added
                    the base url to form different urls.
   """
 
@@ skipping 67 matching lines @@
 
       # Go through changed files, they are in li.
       lis = ul.getElementsByTagName('li')
       for li in lis:
         # Retrieve path and action of the changed file
         file_path = li.getElementsByTagName('a')[0].firstChild.nodeValue
         file_change_type = li.getElementsByTagName('span')[
             0].getAttribute('class')
 
         # Normalize file action so that it is same as SVN parser.
-        file_change_type = FILE_CHANGE_TYPE_MAP[file_change_type]
+        file_change_type = _ConvertToFileChangeType(file_change_type)
 
         # Add the changed file to the map.
         if file_path not in file_to_revision_map:
           file_to_revision_map[file_path] = []
         file_to_revision_map[file_path].append((githash, file_change_type))
 
       # Add this revision object to the map.
       revision_map[githash] = revision
 
     # Parse one revision for the start range, because googlesource does not
@@ skipping 61 matching lines @@
     revision['author'] = json_revision['author']['name']
     revision['message'] = json_revision['message']
     revision['url'] = url
 
     # Iterate through the changed files.
     for diff in json_revision['tree_diff']:
       file_path = diff['new_path']
       file_change_type = diff['type']
 
       # Normalize file action so that it fits with svn_repository_parser.
-      file_change_type = FILE_CHANGE_TYPE_MAP[file_change_type]
+      file_change_type = _ConvertToFileChangeType(file_change_type)
 
       # Add the file to the map.
       if file_path not in file_to_revision_map:
         file_to_revision_map[file_path] = []
       file_to_revision_map[file_path].append((githash, file_change_type))
 
     # Add this CL to the map.
     revision_map[githash] = revision
 
     return
 
   def ParseLineDiff(self, path, component, file_change_type, githash):
     changed_line_numbers = []
     changed_line_contents = []
     base_url = self.component_to_url_map[component]['repository']
     backup_url = (base_url + self.url_parts_map['revision_url']) % githash
 
     # If the file is added (not modified), treat it as if it is not changed.
-    if file_change_type == 'A':
+    if file_change_type in ('A', 'R'):

[aarya, 2014/08/26 20:34:18]

you forgot 'C'

[stgao, 2014/08/26 20:50:35]

Done.

+      # TODO(stgao): Maybe return whole file change for Add, Rename, and Copy?
       return (backup_url, changed_line_numbers, changed_line_contents)
 
     # Retrieves the diff data from URL, and if it fails, return emptry lines.
     url = (base_url + self.url_parts_map['diff_url']) % (githash, path)
     data = crash_utils.GetDataFromURL(url + '?format=text')
     if not data:
       return (backup_url, changed_line_numbers, changed_line_contents)
 
     # Decode the returned object to line diff info
     diff = base64.b64decode(data).splitlines()
@@ skipping 54 matching lines @@
         revision_url = base_url + revision_url_parts
         # TODO(jeun): Add a way to get content from JSON object.
         content = None
 
         (revision_info, _) = self.ParseChangelog(component, revision, revision)
         message = revision_info[revision]['message']
         return (content, revision, author, revision_url, message)
 
     # Return none if the region does not exist.
     return None