git.maemo.org Git - theonering/blob - src/browser_emu.py

   1 """
   2 @author:          Laszlo Nagy
   3 @copyright:   (c) 2005 by Szoftver Messias Bt.
   4 @licence:        BSD style
   5
   6 Objects of the MozillaEmulator class can emulate a browser that is capable of:
   7
   8         - cookie management
   9         - configurable user agent string
  10         - GET and POST
  11         - multipart POST (send files)
  12         - receive content into file
  13
  14 I have seen many requests on the python mailing list about how to emulate a browser. I'm using this class for years now, without any problems. This is how you can use it:
  15
  16         1. Use firefox
  17         2. Install and open the livehttpheaders plugin
  18         3. Use the website manually with firefox
  19         4. Check the GET and POST requests in the livehttpheaders capture window
  20         5. Create an instance of the above class and send the same GET and POST requests to the server.
  21
  22 Optional steps:
  23
  24         - You can change user agent string in the build_opened method
  25         - The "encode_multipart_formdata" function can be used alone to create POST data from a list of field values and files
  26 """
  27
  28 import urllib2
  29 import cookielib
  30 import logging
  31
  32 import socket
  33
  34
  35 socket.setdefaulttimeout(10)
  36
  37
  38 class MozillaEmulator(object):
  39
  40         def __init__(self, trycount = 1):
  41                 """Create a new MozillaEmulator object.
  42
  43                 @param trycount: The download() method will retry the operation if it fails. You can specify -1 for infinite retrying.
  44                          A value of 0 means no retrying. A value of 1 means one retry. etc."""
  45                 self.cookies = cookielib.LWPCookieJar()
  46                 self.debug = False
  47                 self.trycount = trycount
  48
  49         def build_opener(self, url, postdata = None, extraheaders = None, forbid_redirect = False):
  50                 if extraheaders is None:
  51                         extraheaders = {}
  52
  53                 txheaders = {
  54                         'Accept': 'text/xml,application/xml,application/xhtml+xml,text/html;q=0.9,text/plain;q=0.8,image/png',
  55                         'Accept-Language': 'en,en-us;q=0.5',
  56                         'Accept-Charset': 'ISO-8859-1,utf-8;q=0.7,*;q=0.7',
  57                 }
  58                 for key, value in extraheaders.iteritems():
  59                         txheaders[key] = value
  60                 req = urllib2.Request(url, postdata, txheaders)
  61                 self.cookies.add_cookie_header(req)
  62                 if forbid_redirect:
  63                         redirector = HTTPNoRedirector()
  64                 else:
  65                         redirector = urllib2.HTTPRedirectHandler()
  66
  67                 http_handler = urllib2.HTTPHandler(debuglevel=self.debug)
  68                 https_handler = urllib2.HTTPSHandler(debuglevel=self.debug)
  69
  70                 u = urllib2.build_opener(
  71                         http_handler,
  72                         https_handler,
  73                         urllib2.HTTPCookieProcessor(self.cookies),
  74                         redirector
  75                 )
  76                 u.addheaders = [(
  77                         'User-Agent',
  78                         'Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.7.8) Gecko/20050511 Firefox/1.0.4'
  79                 )]
  80                 if not postdata is None:
  81                         req.add_data(postdata)
  82                 return (req, u)
  83
  84         def download(self, url,
  85                         postdata = None, extraheaders = None, forbid_redirect = False,
  86                         trycount = None, only_head = False,
  87                 ):
  88                 """Download an URL with GET or POST methods.
  89
  90                 @param postdata: It can be a string that will be POST-ed to the URL.
  91                         When None is given, the method will be GET instead.
  92                 @param extraheaders: You can add/modify HTTP headers with a dict here.
  93                 @param forbid_redirect: Set this flag if you do not want to handle
  94                         HTTP 301 and 302 redirects.
  95                 @param trycount: Specify the maximum number of retries here.
  96                         0 means no retry on error. Using -1 means infinite retring.
  97                         None means the default value (that is self.trycount).
  98                 @param only_head: Create the openerdirector and return it. In other
  99                         words, this will not retrieve any content except HTTP headers.
 100
 101                 @return: The raw HTML page data
 102                 """
 103                 logging.warning("Performing download of %s" % url)
 104
 105                 if extraheaders is None:
 106                         extraheaders = {}
 107                 if trycount is None:
 108                         trycount = self.trycount
 109                 cnt = 0
 110
 111                 while True:
 112                         try:
 113                                 req, u = self.build_opener(url, postdata, extraheaders, forbid_redirect)
 114                                 openerdirector = u.open(req)
 115                                 if self.debug:
 116                                         print req.get_method(), url
 117                                         print openerdirector.code, openerdirector.msg
 118                                         print openerdirector.headers
 119                                 self.cookies.extract_cookies(openerdirector, req)
 120                                 if only_head:
 121                                         return openerdirector
 122
 123                                 return self._read(openerdirector, trycount)
 124                         except urllib2.URLError:
 125                                 cnt += 1
 126                                 if (-1 < trycount) and (trycount < cnt):
 127                                         raise
 128
 129                         # Retry :-)
 130                         if self.debug:
 131                                 print "MozillaEmulator: urllib2.URLError, retryting ", cnt
 132
 133         def _read(self, openerdirector, trycount):
 134                 chunks = []
 135
 136                 chunk = openerdirector.read()
 137                 chunks.append(chunk)
 138                 #while chunk and cnt < trycount:
 139                 #       time.sleep(1)
 140                 #       cnt += 1
 141                 #       chunk = openerdirector.read()
 142                 #       chunks.append(chunk)
 143
 144                 data = "".join(chunks)
 145
 146                 if "Content-Length" in openerdirector.info():
 147                         assert len(data) == int(openerdirector.info()["Content-Length"]), "The packet header promised %s of data but only was able to read %s of data" % (
 148                                 openerdirector.info()["Content-Length"],
 149                                 len(data),
 150                         )
 151
 152                 return data
 153
 154
 155 class HTTPNoRedirector(urllib2.HTTPRedirectHandler):
 156         """This is a custom http redirect handler that FORBIDS redirection."""
 157
 158         def http_error_302(self, req, fp, code, msg, headers):
 159                 e = urllib2.HTTPError(req.get_full_url(), code, msg, headers, fp)
 160                 if e.code in (301, 302):
 161                         if 'location' in headers:
 162                                 newurl = headers.getheaders('location')[0]
 163                         elif 'uri' in headers:
 164                                 newurl = headers.getheaders('uri')[0]
 165                         e.newurl = newurl
 166                 raise e