Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/facebook.py

   1 import json
   2 import netrc
   3 import re
   4 import socket
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     compat_http_client,
   9     compat_str,
  10     compat_urllib_error,
  11     compat_urllib_parse,
  12     compat_urllib_request,
  13
  14     ExtractorError,
  15 )
  16
  17
  18 class FacebookIE(InfoExtractor):
  19     """Information Extractor for Facebook"""
  20
  21     _VALID_URL = r'^(?:https?://)?(?:\w+\.)?facebook\.com/(?:video/video|photo)\.php\?(?:.*?)v=(?P<ID>\d+)(?:.*)'
  22     _LOGIN_URL = 'https://login.facebook.com/login.php?m&next=http%3A%2F%2Fm.facebook.com%2Fhome.php&'
  23     _NETRC_MACHINE = 'facebook'
  24     IE_NAME = u'facebook'
  25
  26     def report_login(self):
  27         """Report attempt to log in."""
  28         self.to_screen(u'Logging in')
  29
  30     def _real_initialize(self):
  31         if self._downloader is None:
  32             return
  33
  34         useremail = None
  35         password = None
  36         downloader_params = self._downloader.params
  37
  38         # Attempt to use provided username and password or .netrc data
  39         if downloader_params.get('username', None) is not None:
  40             useremail = downloader_params['username']
  41             password = downloader_params['password']
  42         elif downloader_params.get('usenetrc', False):
  43             try:
  44                 info = netrc.netrc().authenticators(self._NETRC_MACHINE)
  45                 if info is not None:
  46                     useremail = info[0]
  47                     password = info[2]
  48                 else:
  49                     raise netrc.NetrcParseError('No authenticators for %s' % self._NETRC_MACHINE)
  50             except (IOError, netrc.NetrcParseError) as err:
  51                 self._downloader.report_warning(u'parsing .netrc: %s' % compat_str(err))
  52                 return
  53
  54         if useremail is None:
  55             return
  56
  57         # Log in
  58         login_form = {
  59             'email': useremail,
  60             'pass': password,
  61             'login': 'Log+In'
  62             }
  63         request = compat_urllib_request.Request(self._LOGIN_URL, compat_urllib_parse.urlencode(login_form))
  64         try:
  65             self.report_login()
  66             login_results = compat_urllib_request.urlopen(request).read()
  67             if re.search(r'<form(.*)name="login"(.*)</form>', login_results) is not None:
  68                 self._downloader.report_warning(u'unable to log in: bad username/password, or exceded login rate limit (~3/min). Check credentials or wait.')
  69                 return
  70         except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
  71             self._downloader.report_warning(u'unable to log in: %s' % compat_str(err))
  72             return
  73
  74     def _real_extract(self, url):
  75         mobj = re.match(self._VALID_URL, url)
  76         if mobj is None:
  77             raise ExtractorError(u'Invalid URL: %s' % url)
  78         video_id = mobj.group('ID')
  79
  80         url = 'https://www.facebook.com/video/video.php?v=%s' % video_id
  81         webpage = self._download_webpage(url, video_id)
  82
  83         BEFORE = '{swf.addParam(param[0], param[1]);});\n'
  84         AFTER = '.forEach(function(variable) {swf.addVariable(variable[0], variable[1]);});'
  85         m = re.search(re.escape(BEFORE) + '(.*?)' + re.escape(AFTER), webpage)
  86         if not m:
  87             raise ExtractorError(u'Cannot parse data')
  88         data = dict(json.loads(m.group(1)))
  89         params_raw = compat_urllib_parse.unquote(data['params'])
  90         params = json.loads(params_raw)
  91         video_data = params['video_data'][0]
  92         video_url = video_data.get('hd_src')
  93         if not video_url:
  94             video_url = video_data['sd_src']
  95         if not video_url:
  96             raise ExtractorError(u'Cannot find video URL')
  97         video_duration = int(video_data['video_duration'])
  98         thumbnail = video_data['thumbnail_src']
  99
 100         video_title = self._html_search_regex('<h2 class="uiHeaderTitle">([^<]+)</h2>',
 101             webpage, u'title')
 102
 103         info = {
 104             'id': video_id,
 105             'title': video_title,
 106             'url': video_url,
 107             'ext': 'mp4',
 108             'duration': video_duration,
 109             'thumbnail': thumbnail,
 110         }
 111         return [info]