[Findit] Support sending cookies for http requests.

tools/findit/crash_utils.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 cgi
 import ConfigParser
 import json
 import logging
 import os
 import time
 import urllib2
 
+from common import utils
 from result import Result
 
 
 INFINITY = float('inf')
 
 
 def ParseURLsFromConfig(file_name):
   """Parses URLS from the config file.
 
   The file should be in python config format, where svn section is in the
@@ skipping 52 matching lines @@
     type_to_url = url_map_for_repository[component_path]
 
     # Add all URLs to this map.
     for option in config.options(section):
       url = config.get(section, option)
       type_to_url[option] = url
 
   return repository_type_to_url_map
 
 
 def NormalizePathLinux(path, parsed_deps):

[Martin Barbella, 2014/08/15 23:00:02]

Is this being called directly at the moment? There should probably be something
like NormalizePath which is used that calls whatever the platform specific
version of the function to do this is. Ideally, this should be implemented in a
way where it doesn't matter what platform we're running on.

   """Normalizes linux path.
 
   Args:
     path: A string representing a path.
     parsed_deps: A map from component path to its component name, repository,
                  etc.
 
   Returns:
     A tuple containing a component this path is in (e.g blink, skia, etc)
     and a path in that component's repository.
   """
   # First normalize the path by retreiving the absolute path.
   normalized_path = os.path.abspath(path)
 
   # Iterate through all component paths in the parsed DEPS, in the decreasing
   # order of the length of the file path.
   for component_path in sorted(parsed_deps,
                                key=(lambda path: -len(path))):
     # New_path is the component path with 'src/' removed.
     new_path = component_path
     if new_path.startswith('src/') and new_path != 'src/':
       new_path = new_path[len('src/'):]
 
     # If this path is the part of file path, this file must be from this
     # component.
     if new_path in normalized_path:
 
       # Currently does not support googlecode.
       if 'googlecode' in parsed_deps[component_path]['repository']:
         return (None, '', '')

[Martin Barbella, 2014/08/15 23:00:02]

Do the callees of this handle this case properly? It might be best to raise an
exception.

 
       # Normalize the path by stripping everything off the component's relative
       # path.
       normalized_path = normalized_path.split(new_path,1)[1]
 
       # Add 'src/' or 'Source/' at the front of the normalized path, depending
       # on what prefix the component path uses. For example, blink uses
       # 'Source' but chromium uses 'src/', and blink component path is
       # 'src/third_party/WebKit/Source', so add 'Source/' in front of the
       # normalized path.
@@ skipping 63 matching lines @@
 
   Args:
     url: URL to get data from.
     retries: Number of times to retry connection.
     sleep_time: Time in seconds to wait before retrying connection.
     timeout: Time in seconds to wait before time out.
 
   Returns:
     None if the data retrieval fails, or the raw data.
   """
-  data = None
-  for i in range(retries):
+  count = 0
+  while True:
+    count += 1
     # Retrieves data from URL.
     try:
-      data = urllib2.urlopen(url, timeout=timeout)
-
-      # If retrieval is successful, return the data.
-      if data:
-        return data.read()
-
-    # If retrieval fails, try after sleep_time second.
-    except urllib2.URLError:
-      time.sleep(sleep_time)
-      continue
+      _, data = utils.GetHttpClient().Get(url)
+      return data
     except IOError:
-      time.sleep(sleep_time)
-      continue
+      if count < retries:
+        # If retrieval fails, try after sleep_time second.
+        time.sleep(sleep_time)
+      else:
+        break
 
   # Return None if it fails to read data from URL 'retries' times.
   return None
 
 
 def FindMinLineDistance(crashed_line_list, changed_line_numbers):
   """Calculates how far the changed line is from one of the crashes.
 
   Finds the minimum distance between the lines that the file crashed on
   and the lines that the file changed. For example, if the file crashed on
@@ skipping 202 matching lines @@
     A list of result objects.
   """
   result_list = []
 
   for blame in blame_list:
     suspected_cl = blame.revision
     revision_url = blame.url
     component_name = blame.component_name
     author = blame.author
     reason = (
         'The CL changes line %s of file %s from stack %d.' %

[Martin Barbella, 2014/08/15 23:00:02]

If we are planning on using this output directly anywhere, we don't want to say
something like "from stack 2". This should be human-readable.

         (blame.line_number, blame.file, blame.stack_frame_index))
     # Blame object does not have review url and reviewers.
     review_url = None
     reviewers = None
     line_content = blame.content
 
     result = Result(suspected_cl, revision_url, component_name, author, reason,
                     review_url, reviewers, line_content)
     result_list.append(result)
 
   return result_list
 
 
 def ResultListToJSON(result_list):
   """Converts result list to JSON format.
 
   Args:
     result_list: A list of result objects
 
   Returns:
     A string, JSON format of the result_list.
 
   """
   return json.dumps([result.ToDictionary() for result in result_list])