3 @copyright: (c) 2005 by Szoftver Messias Bt.
6 Objects of the MozillaEmulator class can emulate a browser that is capable of:
9 - configurable user agent string
11 - multipart POST (send files)
12 - receive content into file
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:
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.
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
35 _moduleLogger = logging.getLogger("browser_emu")
36 socket.setdefaulttimeout(10)
39 class MozillaEmulator(object):
41 def __init__(self, trycount = 1):
42 """Create a new MozillaEmulator object.
44 @param trycount: The download() method will retry the operation if it fails. You can specify -1 for infinite retrying.
45 A value of 0 means no retrying. A value of 1 means one retry. etc."""
47 self.trycount = trycount
48 self._cookies = cookielib.LWPCookieJar()
49 self._loadedFromCookies = False
51 def load_cookies(self, path):
52 assert not self._loadedFromCookies, "Load cookies only once"
56 self._cookies.filename = path
59 except cookielib.LoadError:
60 _moduleLogger.exception("Bad cookie file")
62 _moduleLogger.exception("No cookie file")
64 _moduleLogger.exception("Unknown error with cookies")
66 self._loadedFromCookies = True
68 return self._loadedFromCookies
70 def save_cookies(self):
71 if self._loadedFromCookies:
74 def clear_cookies(self):
75 if self._loadedFromCookies:
78 def download(self, url,
79 postdata = None, extraheaders = None, forbidRedirect = False,
80 trycount = None, only_head = False,
82 """Download an URL with GET or POST methods.
84 @param postdata: It can be a string that will be POST-ed to the URL.
85 When None is given, the method will be GET instead.
86 @param extraheaders: You can add/modify HTTP headers with a dict here.
87 @param forbidRedirect: Set this flag if you do not want to handle
88 HTTP 301 and 302 redirects.
89 @param trycount: Specify the maximum number of retries here.
90 0 means no retry on error. Using -1 means infinite retring.
91 None means the default value (that is self.trycount).
92 @param only_head: Create the openerdirector and return it. In other
93 words, this will not retrieve any content except HTTP headers.
95 @return: The raw HTML page data
97 _moduleLogger.info("Performing download of %s" % url)
99 if extraheaders is None:
102 trycount = self.trycount
107 req, u = self._build_opener(url, postdata, extraheaders, forbidRedirect)
108 openerdirector = u.open(req)
110 _moduleLogger.info("%r - %r" % (req.get_method(), url))
111 _moduleLogger.info("%r - %r" % (openerdirector.code, openerdirector.msg))
112 _moduleLogger.info("%r" % (openerdirector.headers))
113 self._cookies.extract_cookies(openerdirector, req)
115 return openerdirector
117 return self._read(openerdirector, trycount)
118 except urllib2.URLError:
120 if (-1 < trycount) and (trycount < cnt):
124 _moduleLogger.info("MozillaEmulator: urllib2.URLError, retryting %d" % cnt)
126 def _build_opener(self, url, postdata = None, extraheaders = None, forbidRedirect = False):
127 if extraheaders is None:
131 'Accept': 'text/xml,application/xml,application/xhtml+xml,text/html;q=0.9,text/plain;q=0.8,image/png',
132 'Accept-Language': 'en,en-us;q=0.5',
133 'Accept-Charset': 'ISO-8859-1,utf-8;q=0.7,*;q=0.7',
135 for key, value in extraheaders.iteritems():
136 txheaders[key] = value
137 req = urllib2.Request(url, postdata, txheaders)
138 self._cookies.add_cookie_header(req)
140 redirector = HTTPNoRedirector()
141 #_moduleLogger.info("Redirection disabled")
143 redirector = urllib2.HTTPRedirectHandler()
144 #_moduleLogger.info("Redirection enabled")
146 http_handler = urllib2.HTTPHandler(debuglevel=self.debug)
147 https_handler = urllib2.HTTPSHandler(debuglevel=self.debug)
149 u = urllib2.build_opener(
152 urllib2.HTTPCookieProcessor(self._cookies),
157 'Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.7.8) Gecko/20050511 Firefox/1.0.4'
159 if not postdata is None:
160 req.add_data(postdata)
163 def _read(self, openerdirector, trycount):
166 chunk = openerdirector.read()
168 #while chunk and cnt < trycount:
171 # chunk = openerdirector.read()
172 # chunks.append(chunk)
174 data = "".join(chunks)
176 if "Content-Length" in openerdirector.info():
177 assert len(data) == int(openerdirector.info()["Content-Length"]), "The packet header promised %s of data but only was able to read %s of data" % (
178 openerdirector.info()["Content-Length"],
185 class HTTPNoRedirector(urllib2.HTTPRedirectHandler):
186 """This is a custom http redirect handler that FORBIDS redirection."""
188 def http_error_302(self, req, fp, code, msg, headers):
189 e = urllib2.HTTPError(req.get_full_url(), code, msg, headers, fp)
190 if e.code in (301, 302):
191 if 'location' in headers:
192 newurl = headers.getheaders('location')[0]
193 elif 'uri' in headers:
194 newurl = headers.getheaders('uri')[0]
196 _moduleLogger.info("New url: %s" % e.newurl)