[Findit] Re-org code.

appengine/findit/common/http_client_appengine.py

 # Copyright 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 json
 import logging
 
 from google.appengine.api import urlfetch
 
 from common import auth_util
-from common.retry_http_client import RetryHttpClient
+from libs.http.retry_http_client import RetryHttpClient
 
 
 # TODO(katesonia): Move this to config.
 _INTERNAL_HOSTS_TO_SCOPES = {
     'https://chrome-internal.googlesource.com/': (
         'https://www.googleapis.com/auth/gerritcodereview'),
     'https://codereview.chromium.org/': (
         'https://www.googleapis.com/auth/userinfo.email'),
 }
 
 
 class HttpClientAppengine(RetryHttpClient):  # pragma: no cover

[chanli, 2016/11/10 23:04:04]

Why does this http_client_appengine stay with Findit? Predator doesn't need it? 

[wrengr, 2016/11/11 00:06:45]

If you're referring to the "findit" string that occurs in the path
"appengine/findit/blah", that's a misnomer. That occurrence of the string should
(a) really be "cl_culprit_finder" as per the renaming, and (b) will be removed
in a subsequent CL when Predator and Findit(waterfall) are moved to be
top-level.

If you're referring to the stuff in "appengine/findit/libs", part of the problem
is how the proposed directory changes affect shared libraries. That is: certain
libraries (e.g., lib.http.*) are shared by both Findit(waterfall) and Predator.
For both projects to have access to the code, we use symlinks; however, the code
does have to physically reside somewhere. In stgao's current proposal that
physical location is inside Findit(waterfall); i.e., Predator's "lib" will be a
symlink to Findit's "lib", whereas Findit's "lib" will be actual files. In the
most recent meeting, me and mbarbella brought this up as being suboptimal since
it suggests that the lib code is "really part of Findit" rather than making it
clear that it is shared/independent code. Though I think this is safe to address
in a separate CL.

[stgao, 2016/11/14 23:30:51]

We will work on the file moving incrementally, not all in this CL.
This CL is to set up the new code structure and add some example of filing
moving to illustrate the purpose of the new directories.

   """A http client for running on appengine."""
 
   def __init__(self, follow_redirects=True, *args, **kwargs):
     super(HttpClientAppengine, self).__init__(*args, **kwargs)
     self.follow_redirects = follow_redirects
 
   def _ExpandAuthorizationHeaders(self, headers, scope):
     headers['Authorization'] = 'Bearer ' + auth_util.GetAuthToken(scope)
 
   def _ShouldLogError(self, status_code):
@@ skipping 31 matching lines @@
     return result.status_code, result.content
 
   def _Get(self, url, timeout, headers):
     return self._SendRequest(url, urlfetch.GET, None, timeout, headers)
 
   def _Post(self, url, data, timeout, headers):
     return self._SendRequest(url, urlfetch.POST, data, timeout, headers)
 
   def _Put(self, url, data, timeout, headers):
     return self._SendRequest(url, urlfetch.PUT, data, timeout, headers)