mechanize/_mozillacookiejar.py

   1 """Mozilla / Netscape cookie loading / saving.
   2
   3 Copyright 2002-2006 John J Lee <jjl@pobox.com>
   4 Copyright 1997-1999 Gisle Aas (original libwww-perl code)
   5
   6 This code is free software; you can redistribute it and/or modify it
   7 under the terms of the BSD or ZPL 2.1 licenses (see the file
   8 COPYING.txt included with the distribution).
   9
  10 """
  11
  12 import re, time, logging
  13
  14 from _clientcookie import reraise_unmasked_exceptions, FileCookieJar, Cookie, \
  15      MISSING_FILENAME_TEXT, LoadError
  16 debug = logging.getLogger("ClientCookie").debug
  17
  18
  19 class MozillaCookieJar(FileCookieJar):
  20     """
  21
  22     WARNING: you may want to backup your browser's cookies file if you use
  23     this class to save cookies.  I *think* it works, but there have been
  24     bugs in the past!
  25
  26     This class differs from CookieJar only in the format it uses to save and
  27     load cookies to and from a file.  This class uses the Mozilla/Netscape
  28     `cookies.txt' format.  lynx uses this file format, too.
  29
  30     Don't expect cookies saved while the browser is running to be noticed by
  31     the browser (in fact, Mozilla on unix will overwrite your saved cookies if
  32     you change them on disk while it's running; on Windows, you probably can't
  33     save at all while the browser is running).
  34
  35     Note that the Mozilla/Netscape format will downgrade RFC2965 cookies to
  36     Netscape cookies on saving.
  37
  38     In particular, the cookie version and port number information is lost,
  39     together with information about whether or not Path, Port and Discard were
  40     specified by the Set-Cookie2 (or Set-Cookie) header, and whether or not the
  41     domain as set in the HTTP header started with a dot (yes, I'm aware some
  42     domains in Netscape files start with a dot and some don't -- trust me, you
  43     really don't want to know any more about this).
  44
  45     Note that though Mozilla and Netscape use the same format, they use
  46     slightly different headers.  The class saves cookies using the Netscape
  47     header by default (Mozilla can cope with that).
  48
  49     """
  50     magic_re = "#( Netscape)? HTTP Cookie File"
  51     header = """\
  52     # Netscape HTTP Cookie File
  53     # http://www.netscape.com/newsref/std/cookie_spec.html
  54     # This is a generated file!  Do not edit.
  55
  56 """
  57
  58     def _really_load(self, f, filename, ignore_discard, ignore_expires):
  59         now = time.time()
  60
  61         magic = f.readline()
  62         if not re.search(self.magic_re, magic):
  63             f.close()
  64             raise LoadError(
  65                 "%s does not look like a Netscape format cookies file" %
  66                 filename)
  67
  68         try:
  69             while 1:
  70                 line = f.readline()
  71                 if line == "": break
  72
  73                 # last field may be absent, so keep any trailing tab
  74                 if line.endswith("\n"): line = line[:-1]
  75
  76                 # skip comments and blank lines XXX what is $ for?
  77                 if (line.strip().startswith("#") or
  78                     line.strip().startswith("$") or
  79                     line.strip() == ""):
  80                     continue
  81
  82                 domain, domain_specified, path, secure, expires, name, value = \
  83                         line.split("\t")
  84                 secure = (secure == "TRUE")
  85                 domain_specified = (domain_specified == "TRUE")
  86                 if name == "":
  87                     name = value
  88                     value = None
  89
  90                 initial_dot = domain.startswith(".")
  91                 assert domain_specified == initial_dot
  92
  93                 discard = False
  94                 if expires == "":
  95                     expires = None
  96                     discard = True
  97
  98                 # assume path_specified is false
  99                 c = Cookie(0, name, value,
 100                            None, False,
 101                            domain, domain_specified, initial_dot,
 102                            path, False,
 103                            secure,
 104                            expires,
 105                            discard,
 106                            None,
 107                            None,
 108                            {})
 109                 if not ignore_discard and c.discard:
 110                     continue
 111                 if not ignore_expires and c.is_expired(now):
 112                     continue
 113                 self.set_cookie(c)
 114
 115         except:
 116             reraise_unmasked_exceptions((IOError,))
 117             raise LoadError("invalid Netscape format file %s: %s" %
 118                           (filename, line))
 119
 120     def save(self, filename=None, ignore_discard=False, ignore_expires=False):
 121         if filename is None:
 122             if self.filename is not None: filename = self.filename
 123             else: raise ValueError(MISSING_FILENAME_TEXT)
 124
 125         f = open(filename, "w")
 126         try:
 127             debug("Saving Netscape cookies.txt file")
 128             f.write(self.header)
 129             now = time.time()
 130             for cookie in self:
 131                 if not ignore_discard and cookie.discard:
 132                     debug("   Not saving %s: marked for discard", cookie.name)
 133                     continue
 134                 if not ignore_expires and cookie.is_expired(now):
 135                     debug("   Not saving %s: expired", cookie.name)
 136                     continue
 137                 if cookie.secure: secure = "TRUE"
 138                 else: secure = "FALSE"
 139                 if cookie.domain.startswith("."): initial_dot = "TRUE"
 140                 else: initial_dot = "FALSE"
 141                 if cookie.expires is not None:
 142                     expires = str(cookie.expires)
 143                 else:
 144                     expires = ""
 145                 if cookie.value is None:
 146                     # cookies.txt regards 'Set-Cookie: foo' as a cookie
 147                     # with no name, whereas cookielib regards it as a
 148                     # cookie with no value.
 149                     name = ""
 150                     value = cookie.name
 151                 else:
 152                     name = cookie.name
 153                     value = cookie.value
 154                 f.write(
 155                     "\t".join([cookie.domain, initial_dot, cookie.path,
 156                                secure, expires, name, value])+
 157                     "\n")
 158         finally:
 159             f.close()