vcs: Add custom response header 'X-RhodeCode-Backend' to indicate VCS responses and which backend is in use.

rhodecode/lib/middleware/simplesvn.py

@@ -1,130 +1,136 b''
 # -*- coding: utf-8 -*-
 
 # Copyright (C) 2010-2016  RhodeCode GmbH
 #
 # This program is free software: you can redistribute it and/or modify
 # it under the terms of the GNU Affero General Public License, version 3
 # (only), as published by the Free Software Foundation.
 #
 # This program is distributed in the hope that it will be useful,
 # but WITHOUT ANY WARRANTY; without even the implied warranty of
 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 # GNU General Public License for more details.
 #
 # You should have received a copy of the GNU Affero General Public License
 # along with this program.  If not, see <http://www.gnu.org/licenses/>.
 #
 # This program is dual-licensed. If you wish to learn more about the
 # RhodeCode Enterprise Edition, including its added features, Support services,
 # and proprietary license terms, please see https://rhodecode.com/licenses/
 
 from urlparse import urljoin
 
 import requests
 
 import rhodecode
 from rhodecode.lib.middleware import simplevcs
 from rhodecode.lib.utils import is_valid_repo
 
 
 class SimpleSvnApp(object):
     IGNORED_HEADERS = [
         'connection', 'keep-alive', 'content-encoding',
         'transfer-encoding', 'content-length']
 
     def __init__(self, config):
         self.config = config
 
     def __call__(self, environ, start_response):
         request_headers = self._get_request_headers(environ)
 
         data = environ['wsgi.input']
         # johbo: Avoid that we end up with sending the request in chunked
         # transfer encoding (mainly on Gunicorn). If we know the content
         # length, then we should transfer the payload in one request.
         if environ['REQUEST_METHOD'] == 'MKCOL' or 'CONTENT_LENGTH' in environ:
             data = data.read()
 
         response = requests.request(
             environ['REQUEST_METHOD'], self._get_url(environ['PATH_INFO']),
             data=data, headers=request_headers)
 
         response_headers = self._get_response_headers(response.headers)
         start_response(
             '{} {}'.format(response.status_code, response.reason),
             response_headers)
         return response.iter_content(chunk_size=1024)
 
     def _get_url(self, path):
         return urljoin(
             self.config.get('subversion_http_server_url', ''), path)
 
     def _get_request_headers(self, environ):
         headers = {}
 
         for key in environ:
             if not key.startswith('HTTP_'):
                 continue
             new_key = key.split('_')
             new_key = [k.capitalize() for k in new_key[1:]]
             new_key = '-'.join(new_key)
             headers[new_key] = environ[key]
 
         if 'CONTENT_TYPE' in environ:
             headers['Content-Type'] = environ['CONTENT_TYPE']
 
         if 'CONTENT_LENGTH' in environ:
             headers['Content-Length'] = environ['CONTENT_LENGTH']
 
         return headers
 
     def _get_response_headers(self, headers):
-        return [
+        headers = [
             (h, headers[h])
             for h in headers
             if h.lower() not in self.IGNORED_HEADERS
         ]
 
+        # Add custom response header to indicate that this is a VCS response
+        # and which backend is used.
+        headers.append(('X-RhodeCode-Backend', 'svn'))
+
+        return headers
+
 
 class SimpleSvn(simplevcs.SimpleVCS):
 
     SCM = 'svn'
     READ_ONLY_COMMANDS = ('OPTIONS', 'PROPFIND', 'GET', 'REPORT')
 
     def _get_repository_name(self, environ):
         """
         Gets repository name out of PATH_INFO header
 
         :param environ: environ where PATH_INFO is stored
         """
         path = environ['PATH_INFO'].split('!')
         repo_name = path[0].strip('/')
 
         # SVN includes the whole path in it's requests, including
         # subdirectories inside the repo. Therefore we have to search for
         # the repo root directory.
         if not is_valid_repo(repo_name, self.basepath, self.SCM):
             current_path = ''
             for component in repo_name.split('/'):
                 current_path += component
                 if is_valid_repo(current_path, self.basepath, self.SCM):
                     return current_path
                 current_path += '/'
 
         return repo_name
 
     def _get_action(self, environ):
         return (
             'pull'
             if environ['REQUEST_METHOD'] in self.READ_ONLY_COMMANDS
             else 'push')
 
     def _create_wsgi_app(self, repo_path, repo_name, config):
         return SimpleSvnApp(config)
 
     def _create_config(self, extras, repo_name):
         server_url = rhodecode.CONFIG.get(
             'rhodecode_subversion_http_server_url', '')
         extras['subversion_http_server_url'] = (
             server_url or 'http://localhost/')
         return extras

rhodecode/lib/middleware/utils/scm_app_http.py

@@ -1,135 +1,140 b''
 # -*- coding: utf-8 -*-
 
 # Copyright (C) 2014-2016  RhodeCode GmbH
 #
 # This program is free software: you can redistribute it and/or modify
 # it under the terms of the GNU Affero General Public License, version 3
 # (only), as published by the Free Software Foundation.
 #
 # This program is distributed in the hope that it will be useful,
 # but WITHOUT ANY WARRANTY; without even the implied warranty of
 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 # GNU General Public License for more details.
 #
 # You should have received a copy of the GNU Affero General Public License
 # along with this program.  If not, see <http://www.gnu.org/licenses/>.
 #
 # This program is dual-licensed. If you wish to learn more about the
 # RhodeCode Enterprise Edition, including its added features, Support services,
 # and proprietary license terms, please see https://rhodecode.com/licenses/
 
 """
 Implementation of the scm_app interface using raw HTTP communication.
 """
 
 import base64
 import logging
 import urlparse
 import wsgiref.util
 
 import msgpack
 import requests
 import webob.request
 
 import rhodecode
 
 
 log = logging.getLogger(__name__)
 
 
 def create_git_wsgi_app(repo_path, repo_name, config):
     url = _vcs_streaming_url() + 'git/'
-    return VcsHttpProxy(url, repo_path, repo_name, config)
+    return VcsHttpProxy(url, repo_path, repo_name, config, 'git')
 
 
 def create_hg_wsgi_app(repo_path, repo_name, config):
     url = _vcs_streaming_url() + 'hg/'
-    return VcsHttpProxy(url, repo_path, repo_name, config)
+    return VcsHttpProxy(url, repo_path, repo_name, config, 'hg')
 
 
 def _vcs_streaming_url():
     template = 'http://{}/stream/'
     return template.format(rhodecode.CONFIG['vcs.server'])
 
 
 # TODO: johbo: Avoid the global.
 session = requests.Session()
 # Requests speedup, avoid reading .netrc and similar
 session.trust_env = False
 
 
 class VcsHttpProxy(object):
     """
     A WSGI application which proxies vcs requests.
 
     The goal is to shuffle the data around without touching it. The only
     exception is the extra data from the config object which we send to the
     server as well.
     """
 
-    def __init__(self, url, repo_path, repo_name, config):
+    def __init__(self, url, repo_path, repo_name, config, backend):
         """
         :param str url: The URL of the VCSServer to call.
         """
         self._url = url
         self._repo_name = repo_name
         self._repo_path = repo_path
         self._config = config
+        self._backend = backend
         log.debug(
             "Creating VcsHttpProxy for repo %s, url %s",
             repo_name, url)
 
     def __call__(self, environ, start_response):
         status = '200 OK'
 
         config = msgpack.packb(self._config)
         request = webob.request.Request(environ)
         request_headers = request.headers
         request_headers.update({
             # TODO: johbo: Remove this, rely on URL path only
             'X-RC-Repo-Name': self._repo_name,
             'X-RC-Repo-Path': self._repo_path,
             'X-RC-Path-Info': environ['PATH_INFO'],
             # TODO: johbo: Avoid encoding and put this into payload?
             'X-RC-Repo-Config': base64.b64encode(config),
         })
 
         data = environ['wsgi.input'].read()
         method = environ['REQUEST_METHOD']
 
         # Preserve the query string
         url = self._url
         url = urlparse.urljoin(url, self._repo_name)
         if environ.get('QUERY_STRING'):
             url += '?' + environ['QUERY_STRING']
 
         response = session.request(
             method, url,
             data=data,
             headers=request_headers,
             stream=True)
 
         # Preserve the headers of the response, except hop_by_hop ones
         response_headers = [
             (h, v) for h, v in response.headers.items()
             if not wsgiref.util.is_hop_by_hop(h)
         ]
 
+        # Add custom response header to indicate that this is a VCS response
+        # and which backend is used.
+        response_headers.append(('X-RhodeCode-Backend', self._backend))
+
         # TODO: johbo: Better way to get the status including text?
         status = str(response.status_code)
         start_response(status, response_headers)
         return _maybe_stream(response)
 
 
 def _maybe_stream(response):
     """
     Try to generate chunks from the response if it is chunked.
     """
     if _is_chunked(response):
         return response.raw.read_chunked()
     else:
         return [response.content]
 
 
 def _is_chunked(response):
     return response.headers.get('Transfer-Encoding', '') == 'chunked'