Logo Search packages:      
Sourcecode: mozc version File versions  Download package


#!/usr/bin/env python
# Copyright 2007 Google Inc.
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#     http://www.apache.org/licenses/LICENSE-2.0
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# See the License for the specific language governing permissions and
# limitations under the License.

"""Tool for uploading diffs from a version control system to the codereview app.

Usage summary: upload.py [options] [-- diff_options]

Diff options are passed to the diff command of the underlying system.

Supported version control systems:

It is important for Git/Mercurial users to specify a tree/node/branch to diff
against by using the '--rev' option.
# This code is derived from appcfg.py in the App Engine SDK (open source),
# and from ASPN recipe #146306.

import cookielib
import getpass
import logging
import md5
import mimetypes
import optparse
import os
import re
import socket
import subprocess
import sys
import urllib
import urllib2
import urlparse

  import readline
except ImportError:

# The logging verbosity:
#  0: Errors only.
#  1: Status messages.
#  2: Info logs.
#  3: Debug logs.
verbosity = 1

# Max size of patch or base file.
MAX_UPLOAD_SIZE = 900 * 1024

def GetEmail(prompt):
  """Prompts the user for their email address and returns it.

  The last used email address is saved to a file and offered up as a suggestion
  to the user. If the user presses enter without typing in anything the last
  used email address is used. If the user enters a new address, it is saved
  for next time we prompt.

  last_email_file_name = os.path.expanduser("~/.last_codereview_email_address")
  last_email = ""
  if os.path.exists(last_email_file_name):
      last_email_file = open(last_email_file_name, "r")
      last_email = last_email_file.readline().strip("\n")
      prompt += " [%s]" % last_email
    except IOError, e:
  email = raw_input(prompt + ": ").strip()
  if email:
      last_email_file = open(last_email_file_name, "w")
    except IOError, e:
    email = last_email
  return email

def StatusUpdate(msg):
  """Print a status message to stdout.

  If 'verbosity' is greater than 0, print the message.

    msg: The string to print.
  if verbosity > 0:
    print msg

def ErrorExit(msg):
  """Print an error message to stderr and exit."""
  print >>sys.stderr, msg

class ClientLoginError(urllib2.HTTPError):
  """Raised to indicate there was an error authenticating with ClientLogin."""

  def __init__(self, url, code, msg, headers, args):
    urllib2.HTTPError.__init__(self, url, code, msg, headers, None)
    self.args = args
    self.reason = args["Error"]

class AbstractRpcServer(object):
  """Provides a common interface for a simple RPC server."""

00127   def __init__(self, host, auth_function, host_override=None, extra_headers={},
    """Creates a new HttpRpcServer.

      host: The host to send requests to.
      auth_function: A function that takes no arguments and returns an
        (email, password) tuple when called. Will be called if authentication
        is required.
      host_override: The host header to send to the server (defaults to host).
      extra_headers: A dict of extra headers to append to every request.
      save_cookies: If True, save the authentication cookies to local disk.
        If False, use an in-memory cookiejar instead.  Subclasses must
        implement this functionality.  Defaults to False.
    self.host = host
    self.host_override = host_override
    self.auth_function = auth_function
    self.authenticated = False
    self.extra_headers = extra_headers
    self.save_cookies = save_cookies
    self.opener = self._GetOpener()
    if self.host_override:
      logging.info("Server: %s; Host: %s", self.host, self.host_override)
      logging.info("Server: %s", self.host)

00154   def _GetOpener(self):
    """Returns an OpenerDirector for making HTTP requests.

      A urllib2.OpenerDirector object.
    raise NotImplementedError()

00162   def _CreateRequest(self, url, data=None):
    """Creates a new urllib request."""
    logging.debug("Creating request for: '%s' with payload:\n%s", url, data)
    req = urllib2.Request(url, data=data)
    if self.host_override:
      req.add_header("Host", self.host_override)
    for key, value in self.extra_headers.iteritems():
      req.add_header(key, value)
    return req

00172   def _GetAuthToken(self, email, password):
    """Uses ClientLogin to authenticate the user, returning an auth token.

      email:    The user's email address
      password: The user's password

      ClientLoginError: If there was an error authenticating with ClientLogin.
      HTTPError: If there was some other form of HTTP error.

      The authentication token returned by ClientLogin.
    account_type = "GOOGLE"
    if self.host.endswith(".google.com"):
      # Needed for use inside Google.
      account_type = "HOSTED"
    req = self._CreateRequest(
            "Email": email,
            "Passwd": password,
            "service": "ah",
            "source": "rietveld-codereview-upload",
            "accountType": account_type,
      response = self.opener.open(req)
      response_body = response.read()
      response_dict = dict(x.split("=")
                           for x in response_body.split("\n") if x)
      return response_dict["Auth"]
    except urllib2.HTTPError, e:
      if e.code == 403:
        body = e.read()
        response_dict = dict(x.split("=", 1) for x in body.split("\n") if x)
        raise ClientLoginError(req.get_full_url(), e.code, e.msg,
                               e.headers, response_dict)

00215   def _GetAuthCookie(self, auth_token):
    """Fetches authentication cookies for an authentication token.

      auth_token: The authentication token returned by ClientLogin.

      HTTPError: If there was an error fetching the authentication cookies.
    # This is a dummy value to allow us to identify when we're successful.
    continue_location = "http://localhost/"
    args = {"continue": continue_location, "auth": auth_token}
    req = self._CreateRequest("http://%s/_ah/login?%s" %
                              (self.host, urllib.urlencode(args)))
      response = self.opener.open(req)
    except urllib2.HTTPError, e:
      response = e
    if (response.code != 302 or
        response.info()["location"] != continue_location):
      raise urllib2.HTTPError(req.get_full_url(), response.code, response.msg,
                              response.headers, response.fp)
    self.authenticated = True

00239   def _Authenticate(self):
    """Authenticates the user.

    The authentication process works as follows:
     1) We get a username and password from the user
     2) We use ClientLogin to obtain an AUTH token for the user
        (see http://code.google.com/apis/accounts/AuthForInstalledApps.html).
     3) We pass the auth token to /_ah/login on the server to obtain an
        authentication cookie. If login was successful, it tries to redirect
        us to the URL we provided.

    If we attempt to access the upload API without first obtaining an
    authentication cookie, it returns a 401 response and directs us to
    authenticate ourselves with ClientLogin.
    for i in range(3):
      credentials = self.auth_function()
        auth_token = self._GetAuthToken(credentials[0], credentials[1])
      except ClientLoginError, e:
        if e.reason == "BadAuthentication":
          print >>sys.stderr, "Invalid username or password."
        if e.reason == "CaptchaRequired":
          print >>sys.stderr, (
              "Please go to\n"
              "and verify you are a human.  Then try again.")
        if e.reason == "NotVerified":
          print >>sys.stderr, "Account not verified."
        if e.reason == "TermsNotAgreed":
          print >>sys.stderr, "User has not agreed to TOS."
        if e.reason == "AccountDeleted":
          print >>sys.stderr, "The user account has been deleted."
        if e.reason == "AccountDisabled":
          print >>sys.stderr, "The user account has been disabled."
        if e.reason == "ServiceDisabled":
          print >>sys.stderr, ("The user's access to the service has been "
        if e.reason == "ServiceUnavailable":
          print >>sys.stderr, "The service is not available; try again later."

00291   def Send(self, request_path, payload=None,
    """Sends an RPC and returns the response.

      request_path: The path to send the request to, eg /api/appversion/create.
      payload: The body of the request, or None to send an empty request.
      content_type: The Content-Type header to use.
      timeout: timeout in seconds; default None i.e. no timeout.
        (Note: for large requests on OS X, the timeout doesn't work right.)
      kwargs: Any keyword arguments are converted into query string parameters.

      The response body, as a string.
    # TODO: Don't require authentication.  Let the server say
    # whether it is necessary.
    if not self.authenticated:

    old_timeout = socket.getdefaulttimeout()
      tries = 0
      while True:
        tries += 1
        args = dict(kwargs)
        url = "http://%s%s" % (self.host, request_path)
        if args:
          url += "?" + urllib.urlencode(args)
        req = self._CreateRequest(url=url, data=payload)
        req.add_header("Content-Type", content_type)
          f = self.opener.open(req)
          response = f.read()
          return response
        except urllib2.HTTPError, e:
          if tries > 3:
          elif e.code == 401:
##           elif e.code >= 500 and e.code < 600:
##             # Server Error - try again.
##             continue

class HttpRpcServer(AbstractRpcServer):
  """Provides a simplified RPC-style interface for HTTP requests."""

00347   def _Authenticate(self):
    """Save the cookie jar after authentication."""
    super(HttpRpcServer, self)._Authenticate()
    if self.save_cookies:
      StatusUpdate("Saving authentication cookies to %s" % self.cookie_file)

00354   def _GetOpener(self):
    """Returns an OpenerDirector that supports cookies and ignores redirects.

      A urllib2.OpenerDirector object.
    opener = urllib2.OpenerDirector()
    if self.save_cookies:
      self.cookie_file = os.path.expanduser("~/.codereview_upload_cookies")
      self.cookie_jar = cookielib.MozillaCookieJar(self.cookie_file)
      if os.path.exists(self.cookie_file):
          self.authenticated = True
          StatusUpdate("Loaded authentication cookies from %s" %
        except (cookielib.LoadError, IOError):
          # Failed to load cookies - just ignore them.
        # Create an empty cookie file with mode 600
        fd = os.open(self.cookie_file, os.O_CREAT, 0600)
      # Always chmod the cookie file
      os.chmod(self.cookie_file, 0600)
      # Don't save cookies across runs of update.py.
      self.cookie_jar = cookielib.CookieJar()
    return opener

parser = optparse.OptionParser(usage="%prog [options] [-- diff_options]")
parser.add_option("-y", "--assume_yes", action="store_true",
                  dest="assume_yes", default=False,
                  help="Assume that the answer to yes/no questions is 'yes'.")
# Logging
group = parser.add_option_group("Logging options")
group.add_option("-q", "--quiet", action="store_const", const=0,
                 dest="verbose", help="Print errors only.")
group.add_option("-v", "--verbose", action="store_const", const=2,
                 dest="verbose", default=1,
                 help="Print info level logs (default).")
group.add_option("--noisy", action="store_const", const=3,
                 dest="verbose", help="Print all logs.")
# Review server
group = parser.add_option_group("Review server options")
group.add_option("-s", "--server", action="store", dest="server",
                 help=("The server to upload to. The format is host[:port]. "
                       "Defaults to 'codereview.appspot.com'."))
group.add_option("-e", "--email", action="store", dest="email",
                 metavar="EMAIL", default=None,
                 help="The username to use. Will prompt if omitted.")
group.add_option("-H", "--host", action="store", dest="host",
                 metavar="HOST", default=None,
                 help="Overrides the Host header sent with all RPCs.")
group.add_option("--no_cookies", action="store_false",
                 dest="save_cookies", default=True,
                 help="Do not save authentication cookies to local disk.")
# Issue
group = parser.add_option_group("Issue options")
group.add_option("-d", "--description", action="store", dest="description",
                 metavar="DESCRIPTION", default=None,
                 help="Optional description when creating an issue.")
group.add_option("-f", "--description_file", action="store",
                 dest="description_file", metavar="DESCRIPTION_FILE",
                 help="Optional path of a file that contains "
                      "the description when creating an issue.")
group.add_option("-r", "--reviewers", action="store", dest="reviewers",
                 metavar="REVIEWERS", default=None,
                 help="Add reviewers (comma separated email addresses).")
group.add_option("--cc", action="store", dest="cc",
                 metavar="CC", default=None,
                 help="Add CC (comma separated email addresses).")
# Upload options
group = parser.add_option_group("Patch options")
group.add_option("-m", "--message", action="store", dest="message",
                 metavar="MESSAGE", default=None,
                 help="A message to identify the patch. "
                      "Will prompt if omitted.")
group.add_option("-i", "--issue", type="int", action="store",
                 metavar="ISSUE", default=None,
                 help="Issue number to which to add. Defaults to new issue.")
group.add_option("--download_base", action="store_true",
                 dest="download_base", default=False,
                 help="Base files will be downloaded by the server "
                 "(side-by-side diffs may not work on files with CRs).")
group.add_option("--rev", action="store", dest="revision",
                 metavar="REV", default=None,
                 help="Branch/tree/revision to diff against (used by DVCS).")
group.add_option("--send_mail", action="store_true",
                 dest="send_mail", default=False,
                 help="Send notification email to reviewers.")

def GetRpcServer(options):
  """Returns an instance of an AbstractRpcServer.

    A new AbstractRpcServer, on which RPC calls can be made.

  rpc_server_class = HttpRpcServer

  def GetUserCredentials():
    """Prompts the user for a username and password."""
    email = options.email
    if email is None:
      email = GetEmail("Email (login for uploading to %s)" % options.server)
    password = getpass.getpass("Password for %s: " % email)
    return (email, password)

  # If this is the dev_appserver, use fake authentication.
  host = (options.host or options.server).lower()
  if host == "localhost" or host.startswith("localhost:"):
    email = options.email
    if email is None:
      email = "test@example.com"
      logging.info("Using debug user %s.  Override with --email" % email)
    server = rpc_server_class(
        lambda: (email, "password"),
                       'dev_appserver_login="%s:False"' % email},
    # Don't try to talk to ClientLogin.
    server.authenticated = True
    return server

  return rpc_server_class(options.server, GetUserCredentials,

def EncodeMultipartFormData(fields, files):
  """Encode form fields for multipart/form-data.

    fields: A sequence of (name, value) elements for regular form fields.
    files: A sequence of (name, filename, value) elements for data to be
           uploaded as files.
    (content_type, body) ready for httplib.HTTP instance.

  BOUNDARY = '-M-A-G-I-C---B-O-U-N-D-A-R-Y-'
  CRLF = '\r\n'
  lines = []
  for (key, value) in fields:
    lines.append('--' + BOUNDARY)
    lines.append('Content-Disposition: form-data; name="%s"' % key)
  for (key, filename, value) in files:
    lines.append('--' + BOUNDARY)
    lines.append('Content-Disposition: form-data; name="%s"; filename="%s"' %
             (key, filename))
    lines.append('Content-Type: %s' % GetContentType(filename))
  lines.append('--' + BOUNDARY + '--')
  body = CRLF.join(lines)
  content_type = 'multipart/form-data; boundary=%s' % BOUNDARY
  return content_type, body

def GetContentType(filename):
  """Helper to guess the content-type from the filename."""
  return mimetypes.guess_type(filename)[0] or 'application/octet-stream'

# Use a shell for subcommands on Windows to get a PATH search.
use_shell = sys.platform.startswith("win")

def RunShellWithReturnCode(command, print_output=False,
  """Executes a command and returns the output from stdout and the return code.

    command: Command to execute.
    print_output: If True, the output is printed to stdout.
                  If False, both stdout and stderr are ignored.
    universal_newlines: Use universal_newlines flag (default: True).

    Tuple (output, return code)
  logging.info("Running %s", command)
  p = subprocess.Popen(command, stdout=subprocess.PIPE, stderr=subprocess.PIPE,
                       shell=use_shell, universal_newlines=universal_newlines)
  if print_output:
    output_array = []
    while True:
      line = p.stdout.readline()
      if not line:
      print line.strip("\n")
    output = "".join(output_array)
    output = p.stdout.read()
  errout = p.stderr.read()
  if print_output and errout:
    print >>sys.stderr, errout
  return output, p.returncode

def RunShell(command, silent_ok=False, universal_newlines=True,
  data, retcode = RunShellWithReturnCode(command, print_output,
  if retcode:
    ErrorExit("Got error status from %s:\n%s" % (command, data))
  if not silent_ok and not data:
    ErrorExit("No output from %s" % command)
  return data

class VersionControlSystem(object):
  """Abstract base class providing an interface to the VCS."""

00591   def __init__(self, options):

      options: Command line options.
    self.options = options

00599   def GenerateDiff(self, args):
    """Return the current diff as a string.

      args: Extra arguments to pass to the diff command.
    raise NotImplementedError(
        "abstract method -- subclass %s must override" % self.__class__)

00608   def GetUnknownFiles(self):
    """Return a list of files unknown to the VCS."""
    raise NotImplementedError(
        "abstract method -- subclass %s must override" % self.__class__)

00613   def CheckForUnknownFiles(self):
    """Show an "are you sure?" prompt if there are unknown files."""
    unknown_files = self.GetUnknownFiles()
    if unknown_files:
      print "The following files are not added to version control:"
      for line in unknown_files:
        print line
      prompt = "Are you sure to continue?(y/N) "
      answer = raw_input(prompt).strip()
      if answer != "y":
        ErrorExit("User aborted")

00625   def GetBaseFile(self, filename):
    """Get the content of the upstream version of a file.

      A tuple (base_content, new_content, is_binary, status)
        base_content: The contents of the base file.
        new_content: For text files, this is empty.  For binary files, this is
          the contents of the new file, since the diff output won't contain
          information to reconstruct the current file.
        is_binary: True iff the file is binary.
        status: The status of the file.

    raise NotImplementedError(
        "abstract method -- subclass %s must override" % self.__class__)

00642   def GetBaseFiles(self, diff):
    """Helper that calls GetBase file for each file in the patch.

      A dictionary that maps from filename to GetBaseFile's tuple.  Filenames
      are retrieved based on lines that start with "Index:" or
      "Property changes on:".
    files = {}
    for line in diff.splitlines(True):
      if line.startswith('Index:') or line.startswith('Property changes on:'):
        unused, filename = line.split(':', 1)
        # On Windows if a file has property changes its filename uses '\'
        # instead of '/'.
        filename = filename.strip().replace('\\', '/')
        files[filename] = self.GetBaseFile(filename)
    return files

00661   def UploadBaseFiles(self, issue, rpc_server, patch_list, patchset, options,
    """Uploads the base files (and if necessary, the current ones as well)."""

    def UploadFile(filename, file_id, content, is_binary, status, is_base):
      """Uploads a file to the server."""
      file_too_large = False
      if is_base:
        type = "base"
        type = "current"
      if len(content) > MAX_UPLOAD_SIZE:
        print ("Not uploading the %s file for %s because it's too large." %
               (type, filename))
        file_too_large = True
        content = ""
      checksum = md5.new(content).hexdigest()
      if options.verbose > 0 and not file_too_large:
        print "Uploading %s file for %s" % (type, filename)
      url = "/%d/upload_content/%d/%d" % (int(issue), int(patchset), file_id)
      form_fields = [("filename", filename),
                     ("status", status),
                     ("checksum", checksum),
                     ("is_binary", str(is_binary)),
                     ("is_current", str(not is_base)),
      if file_too_large:
        form_fields.append(("file_too_large", "1"))
      if options.email:
        form_fields.append(("user", options.email))
      ctype, body = EncodeMultipartFormData(form_fields,
                                            [("data", filename, content)])
      response_body = rpc_server.Send(url, body,
      if not response_body.startswith("OK"):
        StatusUpdate("  --> %s" % response_body)

    patches = dict()
    [patches.setdefault(v, k) for k, v in patch_list]
    for filename in patches.keys():
      base_content, new_content, is_binary, status = files[filename]
      file_id_str = patches.get(filename)
      if file_id_str.find("nobase") != -1:
        base_content = None
        file_id_str = file_id_str[file_id_str.rfind("_") + 1:]
      file_id = int(file_id_str)
      if base_content != None:
        UploadFile(filename, file_id, base_content, is_binary, status, True)
      if new_content != None:
        UploadFile(filename, file_id, new_content, is_binary, status, False)

00713   def IsImage(self, filename):
    """Returns true if the filename has an image extension."""
    mimetype =  mimetypes.guess_type(filename)[0]
    if not mimetype:
      return False
    return mimetype.startswith("image/")

class SubversionVCS(VersionControlSystem):
  """Implementation of the VersionControlSystem interface for Subversion."""

00724   def __init__(self, options):
    super(SubversionVCS, self).__init__(options)
    if self.options.revision:
      match = re.match(r"(\d+)(:(\d+))?", self.options.revision)
      if not match:
        ErrorExit("Invalid Subversion revision %s." % self.options.revision)
      self.rev_start = match.group(1)
      self.rev_end = match.group(3)
      self.rev_start = self.rev_end = None
    # Cache output from "svn list -r REVNO dirname".
    # Keys: dirname, Values: 2-tuple (ouput for start rev and end rev).
    self.svnls_cache = {}
    # SVN base URL is required to fetch files deleted in an older revision.
    # Result is cached to not guess it over and over again in GetBaseFile().
    required = self.options.download_base or self.options.revision is not None
    self.svn_base = self._GuessBase(required)

00742   def GuessBase(self, required):
    """Wrapper for _GuessBase."""
    return self.svn_base

00746   def _GuessBase(self, required):
    """Returns the SVN base URL.

      required: If true, exits if the url can't be guessed, otherwise None is
    info = RunShell(["svn", "info"])
    for line in info.splitlines():
      words = line.split()
      if len(words) == 2 and words[0] == "URL:":
        url = words[1]
        scheme, netloc, path, params, query, fragment = urlparse.urlparse(url)
        username, netloc = urllib.splituser(netloc)
        if username:
          logging.info("Removed username from base URL")
        if netloc.endswith("svn.python.org"):
          if netloc == "svn.python.org":
            if path.startswith("/projects/"):
              path = path[9:]
          elif netloc != "pythondev@svn.python.org":
            ErrorExit("Unrecognized Python URL: %s" % url)
          base = "http://svn.python.org/view/*checkout*%s/" % path
          logging.info("Guessed Python base = %s", base)
        elif netloc.endswith("svn.collab.net"):
          if path.startswith("/repos/"):
            path = path[6:]
          base = "http://svn.collab.net/viewvc/*checkout*%s/" % path
          logging.info("Guessed CollabNet base = %s", base)
        elif netloc.endswith(".googlecode.com"):
          path = path + "/"
          base = urlparse.urlunparse(("http", netloc, path, params,
                                      query, fragment))
          logging.info("Guessed Google Code base = %s", base)
          path = path + "/"
          base = urlparse.urlunparse((scheme, netloc, path, params,
                                      query, fragment))
          logging.info("Guessed base = %s", base)
        return base
    if required:
      ErrorExit("Can't find URL in output from svn info")
    return None

00790   def GenerateDiff(self, args):
    cmd = ["svn", "diff"]
    if self.options.revision:
      cmd += ["-r", self.options.revision]
    data = RunShell(cmd)
    count = 0
    for line in data.splitlines():
      if line.startswith("Index:") or line.startswith("Property changes on:"):
        count += 1
    if not count:
      ErrorExit("No valid patches found in output from svn diff")
    return data

00805   def _CollapseKeywords(self, content, keyword_str):
    """Collapses SVN keywords."""
    # svn cat translates keywords but svn diff doesn't. As a result of this
    # behavior patching.PatchChunks() fails with a chunk mismatch error.
    # This part was originally written by the Review Board development team
    # who had the same problem (http://reviews.review-board.org/r/276/).
    # Mapping of keywords to known aliases
    svn_keywords = {
      # Standard keywords
      'Date':                ['Date', 'LastChangedDate'],
      'Revision':            ['Revision', 'LastChangedRevision', 'Rev'],
      'Author':              ['Author', 'LastChangedBy'],
      'HeadURL':             ['HeadURL', 'URL'],
      'Id':                  ['Id'],

      # Aliases
      'LastChangedDate':     ['LastChangedDate', 'Date'],
      'LastChangedRevision': ['LastChangedRevision', 'Rev', 'Revision'],
      'LastChangedBy':       ['LastChangedBy', 'Author'],
      'URL':                 ['URL', 'HeadURL'],

    def repl(m):
       if m.group(2):
         return "$%s::%s$" % (m.group(1), " " * len(m.group(3)))
       return "$%s$" % m.group(1)
    keywords = [keyword
                for name in keyword_str.split(" ")
                for keyword in svn_keywords.get(name, [])]
    return re.sub(r"\$(%s):(:?)([^\$]+)\$" % '|'.join(keywords), repl, content)

00836   def GetUnknownFiles(self):
    status = RunShell(["svn", "status", "--ignore-externals"], silent_ok=True)
    unknown_files = []
    for line in status.split("\n"):
      if line and line[0] == "?":
    return unknown_files

00844   def ReadFile(self, filename):
    """Returns the contents of a file."""
    file = open(filename, 'rb')
    result = ""
      result = file.read()
    return result

00854   def GetStatus(self, filename):
    """Returns the status of a file."""
    if not self.options.revision:
      status = RunShell(["svn", "status", "--ignore-externals", filename])
      if not status:
        ErrorExit("svn status returned no output for %s" % filename)
      status_lines = status.splitlines()
      # If file is in a cl, the output will begin with
      # "\n--- Changelist 'cl_name':\n".  See
      # http://svn.collab.net/repos/svn/trunk/notes/changelist-design.txt
      if (len(status_lines) == 3 and
          not status_lines[0] and
          status_lines[1].startswith("--- Changelist")):
        status = status_lines[2]
        status = status_lines[0]
    # If we have a revision to diff against we need to run "svn list"
    # for the old and the new revision and compare the results to get
    # the correct status for a file.
      dirname, relfilename = os.path.split(filename)
      if dirname not in self.svnls_cache:
        cmd = ["svn", "list", "-r", self.rev_start, dirname or "."]
        out, returncode = RunShellWithReturnCode(cmd)
        if returncode:
          ErrorExit("Failed to get status for %s." % filename)
        old_files = out.splitlines()
        args = ["svn", "list"]
        if self.rev_end:
          args += ["-r", self.rev_end]
        cmd = args + [dirname or "."]
        out, returncode = RunShellWithReturnCode(cmd)
        if returncode:
          ErrorExit("Failed to run command %s" % cmd)
        self.svnls_cache[dirname] = (old_files, out.splitlines())
      old_files, new_files = self.svnls_cache[dirname]
      if relfilename in old_files and relfilename not in new_files:
        status = "D   "
      elif relfilename in old_files and relfilename in new_files:
        status = "M   "
        status = "A   "
    return status

00898   def GetBaseFile(self, filename):
    status = self.GetStatus(filename)
    base_content = None
    new_content = None

    # If a file is copied its status will be "A  +", which signifies
    # "addition-with-history".  See "svn st" for more information.  We need to
    # upload the original file or else diff parsing will fail if the file was
    # edited.
    if status[0] == "A" and status[3] != "+":
      # We'll need to upload the new content if we're adding a binary file
      # since diff's output won't contain it.
      mimetype = RunShell(["svn", "propget", "svn:mime-type", filename],
      base_content = ""
      is_binary = mimetype and not mimetype.startswith("text/")
      if is_binary and self.IsImage(filename):
        new_content = self.ReadFile(filename)
    elif (status[0] in ("M", "D", "R") or
          (status[0] == "A" and status[3] == "+") or  # Copied file.
          (status[0] == " " and status[1] == "M")):  # Property change.
      args = []
      if self.options.revision:
        url = "%s/%s@%s" % (self.svn_base, filename, self.rev_start)
        # Don't change filename, it's needed later.
        url = filename
        args += ["-r", "BASE"]
      cmd = ["svn"] + args + ["propget", "svn:mime-type", url]
      mimetype, returncode = RunShellWithReturnCode(cmd)
      if returncode:
        # File does not exist in the requested revision.
        # Reset mimetype, it contains an error message.
        mimetype = ""
      get_base = False
      is_binary = mimetype and not mimetype.startswith("text/")
      if status[0] == " ":
        # Empty base content just to force an upload.
        base_content = ""
      elif is_binary:
        if self.IsImage(filename):
          get_base = True
          if status[0] == "M":
            if not self.rev_end:
              new_content = self.ReadFile(filename)
              url = "%s/%s@%s" % (self.svn_base, filename, self.rev_end)
              new_content = RunShell(["svn", "cat", url],
                                     universal_newlines=True, silent_ok=True)
          base_content = ""
        get_base = True

      if get_base:
        if is_binary:
          universal_newlines = False
          universal_newlines = True
        if self.rev_start:
          # "svn cat -r REV delete_file.txt" doesn't work. cat requires
          # the full URL with "@REV" appended instead of using "-r" option.
          url = "%s/%s@%s" % (self.svn_base, filename, self.rev_start)
          base_content = RunShell(["svn", "cat", url],
          base_content = RunShell(["svn", "cat", filename],
        if not is_binary:
          args = []
          if self.rev_start:
            url = "%s/%s@%s" % (self.svn_base, filename, self.rev_start)
            url = filename
            args += ["-r", "BASE"]
          cmd = ["svn"] + args + ["propget", "svn:keywords", url]
          keywords, returncode = RunShellWithReturnCode(cmd)
          if keywords and not returncode:
            base_content = self._CollapseKeywords(base_content, keywords)
      StatusUpdate("svn status returned unexpected output: %s" % status)
    return base_content, new_content, is_binary, status[0:5]

class GitVCS(VersionControlSystem):
  """Implementation of the VersionControlSystem interface for Git."""

00988   def __init__(self, options):
    super(GitVCS, self).__init__(options)
    # Map of filename -> hash of base file.
    self.base_hashes = {}

00993   def GenerateDiff(self, extra_args):
    # This is more complicated than svn's GenerateDiff because we must convert
    # the diff output to include an svn-style "Index:" line as well as record
    # the hashes of the base files, so we can upload them along with our diff.
    if self.options.revision:
      extra_args = [self.options.revision] + extra_args
    gitdiff = RunShell(["git", "diff", "--full-index"] + extra_args)
    svndiff = []
    filecount = 0
    filename = None
    for line in gitdiff.splitlines():
      match = re.match(r"diff --git a/(.*) b/.*$", line)
      if match:
        filecount += 1
        filename = match.group(1)
        svndiff.append("Index: %s\n" % filename)
        # The "index" line in a git diff looks like this (long hashes elided):
        #   index 82c0d44..b2cee3f 100755
        # We want to save the left hash, as that identifies the base file.
        match = re.match(r"index (\w+)\.\.", line)
        if match:
          self.base_hashes[filename] = match.group(1)
      svndiff.append(line + "\n")
    if not filecount:
      ErrorExit("No valid patches found in output from git diff")
    return "".join(svndiff)

01021   def GetUnknownFiles(self):
    status = RunShell(["git", "ls-files", "--exclude-standard", "--others"],
    return status.splitlines()

01026   def GetBaseFile(self, filename):
    hash = self.base_hashes[filename]
    base_content = None
    new_content = None
    is_binary = False
    if hash == "0" * 40:  # All-zero hash indicates no base file.
      status = "A"
      base_content = ""
      status = "M"
      base_content, returncode = RunShellWithReturnCode(["git", "show", hash])
      if returncode:
        ErrorExit("Got error status from 'git show %s'" % hash)
    return (base_content, new_content, is_binary, status)

class MercurialVCS(VersionControlSystem):
  """Implementation of the VersionControlSystem interface for Mercurial."""

  def __init__(self, options, repo_dir):
    super(MercurialVCS, self).__init__(options)
    # Absolute path to repository (we can be in a subdir)
    self.repo_dir = os.path.normpath(repo_dir)
    # Compute the subdir
    cwd = os.path.normpath(os.getcwd())
    assert cwd.startswith(self.repo_dir)
    self.subdir = cwd[len(self.repo_dir):].lstrip(r"\/")
    if self.options.revision:
      self.base_rev = self.options.revision
      self.base_rev = RunShell(["hg", "parent", "-q"]).split(':')[1].strip()

01058   def _GetRelPath(self, filename):
    """Get relative path of a file according to the current directory,
    given its logical path in the repo."""
    assert filename.startswith(self.subdir), filename
    return filename[len(self.subdir):].lstrip(r"\/")

01064   def GenerateDiff(self, extra_args):
    # If no file specified, restrict to the current subdir
    extra_args = extra_args or ["."]
    cmd = ["hg", "diff", "--git", "-r", self.base_rev] + extra_args
    data = RunShell(cmd, silent_ok=True)
    svndiff = []
    filecount = 0
    for line in data.splitlines():
      m = re.match("diff --git a/(\S+) b/(\S+)", line)
      if m:
        # Modify line to make it look like as it comes from svn diff.
        # With this modification no changes on the server side are required
        # to make upload.py work with Mercurial repos.
        # NOTE: for proper handling of moved/copied files, we have to use
        # the second filename.
        filename = m.group(2)
        svndiff.append("Index: %s" % filename)
        svndiff.append("=" * 67)
        filecount += 1
    if not filecount:
      ErrorExit("No valid patches found in output from hg diff")
    return "\n".join(svndiff) + "\n"

01090   def GetUnknownFiles(self):
    """Return a list of files unknown to the VCS."""
    args = []
    status = RunShell(["hg", "status", "--rev", self.base_rev, "-u", "."],
    unknown_files = []
    for line in status.splitlines():
      st, fn = line.split(" ", 1)
      if st == "?":
    return unknown_files

01102   def GetBaseFile(self, filename):
    # "hg status" and "hg cat" both take a path relative to the current subdir
    # rather than to the repo root, but "hg diff" has given us the full path
    # to the repo root.
    base_content = ""
    new_content = None
    is_binary = False
    oldrelpath = relpath = self._GetRelPath(filename)
    # "hg status -C" returns two lines for moved/copied files, one otherwise
    out = RunShell(["hg", "status", "-C", "--rev", self.base_rev, relpath])
    out = out.splitlines()
    # HACK: strip error message about missing file/directory if it isn't in
    # the working copy
    if out[0].startswith('%s: ' % relpath):
      out = out[1:]
    if len(out) > 1:
      # Moved/copied => considered as modified, use old filename to
      # retrieve base contents
      oldrelpath = out[1].strip()
      status = "M"
      status, _ = out[0].split(' ', 1)
    if status != "A":
      base_content = RunShell(["hg", "cat", "-r", self.base_rev, oldrelpath],
      is_binary = "\0" in base_content  # Mercurial's heuristic
    if status != "R":
      new_content = open(relpath, "rb").read()
      is_binary = is_binary or "\0" in new_content
    if is_binary and base_content:
      # Fetch again without converting newlines
      base_content = RunShell(["hg", "cat", "-r", self.base_rev, oldrelpath],
        silent_ok=True, universal_newlines=False)
    if not is_binary or not self.IsImage(relpath):
      new_content = None
    return base_content, new_content, is_binary, status

# NOTE: The SplitPatch function is duplicated in engine.py, keep them in sync.
def SplitPatch(data):
  """Splits a patch into separate pieces for each file.

    data: A string containing the output of svn diff.

    A list of 2-tuple (filename, text) where text is the svn diff output
      pertaining to filename.
  patches = []
  filename = None
  diff = []
  for line in data.splitlines(True):
    new_filename = None
    if line.startswith('Index:'):
      unused, new_filename = line.split(':', 1)
      new_filename = new_filename.strip()
    elif line.startswith('Property changes on:'):
      unused, temp_filename = line.split(':', 1)
      # When a file is modified, paths use '/' between directories, however
      # when a property is modified '\' is used on Windows.  Make them the same
      # otherwise the file shows up twice.
      temp_filename = temp_filename.strip().replace('\\', '/')
      if temp_filename != filename:
        # File has property changes but no modifications, create a new diff.
        new_filename = temp_filename
    if new_filename:
      if filename and diff:
        patches.append((filename, ''.join(diff)))
      filename = new_filename
      diff = [line]
    if diff is not None:
  if filename and diff:
    patches.append((filename, ''.join(diff)))
  return patches

def UploadSeparatePatches(issue, rpc_server, patchset, data, options):
  """Uploads a separate patch for each file in the diff output.

  Returns a list of [patch_key, filename] for each file.
  patches = SplitPatch(data)
  rv = []
  for patch in patches:
    if len(patch[1]) > MAX_UPLOAD_SIZE:
      print ("Not uploading the patch for " + patch[0] +
             " because the file is too large.")
    form_fields = [("filename", patch[0])]
    if not options.download_base:
      form_fields.append(("content_upload", "1"))
    files = [("data", "data.diff", patch[1])]
    ctype, body = EncodeMultipartFormData(form_fields, files)
    url = "/%d/upload_patch/%d" % (int(issue), int(patchset))
    print "Uploading patch for " + patch[0]
    response_body = rpc_server.Send(url, body, content_type=ctype)
    lines = response_body.splitlines()
    if not lines or lines[0] != "OK":
      StatusUpdate("  --> %s" % response_body)
    rv.append([lines[1], patch[0]])
  return rv

def GuessVCS(options):
  """Helper to guess the version control system.

  This examines the current directory, guesses which VersionControlSystem
  we're using, and returns an instance of the appropriate class.  Exit with an
  error if we can't figure it out.

    A VersionControlSystem instance. Exits if the VCS can't be guessed.
  # Mercurial has a command to get the base directory of a repository
  # Try running it, but don't die if we don't have hg installed.
  # NOTE: we try Mercurial first as it can sit on top of an SVN working copy.
    out, returncode = RunShellWithReturnCode(["hg", "root"])
    if returncode == 0:
      return MercurialVCS(options, out.strip())
  except OSError, (errno, message):
    if errno != 2:  # ENOENT -- they don't have hg installed.

  # Subversion has a .svn in all working directories.
  if os.path.isdir('.svn'):
    logging.info("Guessed VCS = Subversion")
    return SubversionVCS(options)

  # Git has a command to test if you're in a git tree.
  # Try running it, but don't die if we don't have git installed.
    out, returncode = RunShellWithReturnCode(["git", "rev-parse",
    if returncode == 0:
      return GitVCS(options)
  except OSError, (errno, message):
    if errno != 2:  # ENOENT -- they don't have git installed.

  ErrorExit(("Could not guess version control system. "
             "Are you in a working copy directory?"))

def RealMain(argv, data=None):
  """The real main function.

    argv: Command line arguments.
    data: Diff contents. If None (default) the diff is generated by
      the VersionControlSystem implementation returned by GuessVCS().

    A 2-tuple (issue id, patchset id).
    The patchset id is None if the base files are not uploaded by this
    script (applies only to SVN checkouts).
  logging.basicConfig(format=("%(asctime).19s %(levelname)s %(filename)s:"
                              "%(lineno)s %(message)s "))
  os.environ['LC_ALL'] = 'C'
  options, args = parser.parse_args(argv[1:])
  global verbosity
  verbosity = options.verbose
  if verbosity >= 3:
  elif verbosity >= 2:
  vcs = GuessVCS(options)
  if isinstance(vcs, SubversionVCS):
    # base field is only allowed for Subversion.
    # Note: Fetching base files may become deprecated in future releases.
    base = vcs.GuessBase(options.download_base)
    base = None
  if not base and options.download_base:
    options.download_base = True
    logging.info("Enabled upload of base file")
  if not options.assume_yes:
  if data is None:
    data = vcs.GenerateDiff(args)
  files = vcs.GetBaseFiles(data)
  if verbosity >= 1:
    print "Upload server:", options.server, "(change with -s/--server)"
  if options.issue:
    prompt = "Message describing this patch set: "
    prompt = "New issue subject: "
  message = options.message or raw_input(prompt).strip()
  if not message:
    ErrorExit("A non-empty message is required")
  rpc_server = GetRpcServer(options)
  form_fields = [("subject", message)]
  if base:
    form_fields.append(("base", base))
  if options.issue:
    form_fields.append(("issue", str(options.issue)))
  if options.email:
    form_fields.append(("user", options.email))
  if options.reviewers:
    for reviewer in options.reviewers.split(','):
      if "@" in reviewer and not reviewer.split("@")[1].count(".") == 1:
        ErrorExit("Invalid email address: %s" % reviewer)
    form_fields.append(("reviewers", options.reviewers))
  if options.cc:
    for cc in options.cc.split(','):
      if "@" in cc and not cc.split("@")[1].count(".") == 1:
        ErrorExit("Invalid email address: %s" % cc)
    form_fields.append(("cc", options.cc))
  description = options.description
  if options.description_file:
    if options.description:
      ErrorExit("Can't specify description and description_file")
    file = open(options.description_file, 'r')
    description = file.read()
  if description:
    form_fields.append(("description", description))
  # Send a hash of all the base file so the server can determine if a copy
  # already exists in an earlier patchset.
  base_hashes = ""
  for file, info in files.iteritems():
    if not info[0] is None:
      checksum = md5.new(info[0]).hexdigest()
      if base_hashes:
        base_hashes += "|"
      base_hashes += checksum + ":" + file
  form_fields.append(("base_hashes", base_hashes))
  # If we're uploading base files, don't send the email before the uploads, so
  # that it contains the file status.
  if options.send_mail and options.download_base:
    form_fields.append(("send_mail", "1"))
  if not options.download_base:
    form_fields.append(("content_upload", "1"))
  if len(data) > MAX_UPLOAD_SIZE:
    print "Patch is large, so uploading file patches separately."
    uploaded_diff_file = []
    form_fields.append(("separate_patches", "1"))
    uploaded_diff_file = [("data", "data.diff", data)]
  ctype, body = EncodeMultipartFormData(form_fields, uploaded_diff_file)
  response_body = rpc_server.Send("/upload", body, content_type=ctype)
  patchset = None
  if not options.download_base or not uploaded_diff_file:
    lines = response_body.splitlines()
    if len(lines) >= 2:
      msg = lines[0]
      patchset = lines[1].strip()
      patches = [x.split(" ", 1) for x in lines[2:]]
      msg = response_body
    msg = response_body
  if not response_body.startswith("Issue created.") and \
  not response_body.startswith("Issue updated."):
  issue = msg[msg.rfind("/")+1:]

  if not uploaded_diff_file:
    result = UploadSeparatePatches(issue, rpc_server, patchset, data, options)
    if not options.download_base:
      patches = result

  if not options.download_base:
    vcs.UploadBaseFiles(issue, rpc_server, patches, patchset, options, files)
    if options.send_mail:
      rpc_server.Send("/" + issue + "/mail", payload="")
  return issue, patchset

def main():
  except KeyboardInterrupt:

if __name__ == "__main__":

Generated by  Doxygen 1.6.0   Back to index