diff options
author | RaNaN <Mast3rRaNaN@hotmail.de> | 2010-10-06 14:09:26 +0200 |
---|---|---|
committer | RaNaN <Mast3rRaNaN@hotmail.de> | 2010-10-06 14:09:26 +0200 |
commit | 1edbe8364951623aacf7c6477ce651394f513eec (patch) | |
tree | 12597980896bf8e1ce6e6bfd8b48c62a8352f69f /module/plugins/hoster/FreakshareCom.py | |
parent | closed #148 (diff) | |
download | pyload-1edbe8364951623aacf7c6477ce651394f513eec.tar.xz |
freakshare.com, netload fix
Diffstat (limited to 'module/plugins/hoster/FreakshareCom.py')
-rw-r--r-- | module/plugins/hoster/FreakshareCom.py | 125 |
1 files changed, 125 insertions, 0 deletions
diff --git a/module/plugins/hoster/FreakshareCom.py b/module/plugins/hoster/FreakshareCom.py new file mode 100644 index 000000000..524081002 --- /dev/null +++ b/module/plugins/hoster/FreakshareCom.py @@ -0,0 +1,125 @@ +#!/usr/bin/env python +# -*- coding: utf-8 -*- + +import re +from module.plugins.Hoster import Hoster +from module.plugins.ReCaptcha import ReCaptcha + +class FreakshareCom(Hoster): + __name__ = "FreakshareCom" + __type__ = "hoster" + __pattern__ = r"http://(?:www\.)?freakshare\.(net|com)/files/\S*?/" + __version__ = "0.32" + __description__ = """Freakshare.com Download Hoster""" + __author_name__ = ("sitacuisses","spoob","mkaay") + __author_mail__ = ("sitacuisses@yahoo.de","spoob@pyload.org","mkaay@mkaay.de") + + def setup(self): + self.html = None + self.wantReconnect = False + self.multiDL = False + self.req_opts = [] + + def process(self, pyfile): + pyfile.url = pyfile.url.replace("freakshare.net/","freakshare.com/") + + self.prepare() + self.get_file_url() + + self.download(self.pyfile.url, post=self.req_opts) + + + def prepare(self): + pyfile = self.pyfile + + self.wantReconnect = False + + self.download_html() + + if not self.file_exists(): + self.offline() + + self.setWait( self.get_waiting_time() ) + + pyfile.name = self.get_file_name() + + self.wait() + + return True + + def download_html(self): + url = self.pyfile.url.replace("freakshare.net/","freakshare.com/") + self.html = self.load(url, cookies=True) + + def get_file_url(self): + """ returns the absolute downloadable filepath + """ + if self.html is None: + self.download_html() + if not self.wantReconnect: + self.req_opts = self.get_download_options() # get the Post options for the Request + #file_url = self.pyfile.url + #return file_url + else: + self.offline() + + def get_file_name(self): + if self.html is None: + self.download_html() + if not self.wantReconnect: + file_name = re.search(r"<h1\sclass=\"box_heading\"\sstyle=\"text-align:center\;\">([^ ]+)", self.html).group(1) + return file_name + else: + return self.pyfile.url + + def get_waiting_time(self): + if self.html is None: + self.download_html() + + if "Der Traffic f\xc3\xbcr heute ist verbraucht!" in self.html or "Your Traffic is used up for today" in self.html: + self.wantReconnect = True + return 24*3600 + + timestring = re.search('\s*var\stime\s=\s(\d*?)\.\d*;', self.html).group(1) + if timestring: + sec = int(timestring) + 1 #add 1 sec as tenths of seconds are cut off + else: + sec = 0 + return sec + + def file_exists(self): + """ returns True or False + """ + if self.html is None: + self.download_html() + if re.search(r"Sorry, this Download doesnt exist anymore", self.html) is not None: + return False + else: + return True + + def get_download_options(self): + re_envelope = re.search(r".*?value=\"Free\sDownload\".*?\n*?(.*?<.*?>\n*)*?\n*\s*?</form>", self.html).group(0) #get the whole request + to_sort = re.findall(r"<input\stype=\"hidden\"\svalue=\"(.*?)\"\sname=\"(.*?)\"\s\/>", re_envelope) + request_options = [] + + for item in to_sort: #Name value pairs are output reversed from regex, so we reorder them + request_options.append((item[1], item[0])) + + herewego = self.load(self.pyfile.url, None, request_options, cookies=True) # the actual download-Page + + to_sort = re.findall(r"<input\stype=\".*?\"\svalue=\"(\S*?)\".*?name=\"(\S*?)\"\s.*?\/>", herewego) + request_options = [] + + for item in to_sort: #Same as above + request_options.append((item[1], item[0])) + + challenge = re.search(r"http://api\.recaptcha\.net/challenge\?k=([0-9A-Za-z]+)", herewego) + + if challenge: + re_captcha = ReCaptcha(self) + challenge, result = re_captcha.challenge(challenge.group(1)) + + request_options.append(("recaptcha_challenge_field", challenge)) + request_options.append(("recaptcha_response_field", result)) + + return request_options
\ No newline at end of file |