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     _TEST = {
  26         u'url': u'https://www.facebook.com/photo.php?v=120708114770723',
  27         u'file': u'120708114770723.mp4',
  28         u'md5': u'48975a41ccc4b7a581abd68651c1a5a8',
  29         u'info_dict': {
  30             u"duration": 279,
  31             u"title": u"PEOPLE ARE AWESOME 2013"
  32         }
  33     }
  34
  35     def report_login(self):
  36         """Report attempt to log in."""
  37         self.to_screen(u'Logging in')
  38
  39     def _real_initialize(self):
  40         if self._downloader is None:
  41             return
  42
  43         useremail = None
  44         password = None
  45         downloader_params = self._downloader.params
  46
  47         # Attempt to use provided username and password or .netrc data
  48         if downloader_params.get('username', None) is not None:
  49             useremail = downloader_params['username']
  50             password = downloader_params['password']
  51         elif downloader_params.get('usenetrc', False):
  52             try:
  53                 info = netrc.netrc().authenticators(self._NETRC_MACHINE)
  54                 if info is not None:
  55                     useremail = info[0]
  56                     password = info[2]
  57                 else:
  58                     raise netrc.NetrcParseError('No authenticators for %s' % self._NETRC_MACHINE)
  59             except (IOError, netrc.NetrcParseError) as err:
  60                 self._downloader.report_warning(u'parsing .netrc: %s' % compat_str(err))
  61                 return
  62
  63         if useremail is None:
  64             return
  65
  66         # Log in
  67         login_form = {
  68             'email': useremail,
  69             'pass': password,
  70             'login': 'Log+In'
  71             }
  72         request = compat_urllib_request.Request(self._LOGIN_URL, compat_urllib_parse.urlencode(login_form))
  73         try:
  74             self.report_login()
  75             login_results = compat_urllib_request.urlopen(request).read()
  76             if re.search(r'<form(.*)name="login"(.*)</form>', login_results) is not None:
  77                 self._downloader.report_warning(u'unable to log in: bad username/password, or exceded login rate limit (~3/min). Check credentials or wait.')
  78                 return
  79         except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
  80             self._downloader.report_warning(u'unable to log in: %s' % compat_str(err))
  81             return
  82
  83     def _real_extract(self, url):
  84         mobj = re.match(self._VALID_URL, url)
  85         if mobj is None:
  86             raise ExtractorError(u'Invalid URL: %s' % url)
  87         video_id = mobj.group('ID')
  88
  89         url = 'https://www.facebook.com/video/video.php?v=%s' % video_id
  90         webpage = self._download_webpage(url, video_id)
  91
  92         BEFORE = '{swf.addParam(param[0], param[1]);});\n'
  93         AFTER = '.forEach(function(variable) {swf.addVariable(variable[0], variable[1]);});'
  94         m = re.search(re.escape(BEFORE) + '(.*?)' + re.escape(AFTER), webpage)
  95         if not m:
  96             raise ExtractorError(u'Cannot parse data')
  97         data = dict(json.loads(m.group(1)))
  98         params_raw = compat_urllib_parse.unquote(data['params'])
  99         params = json.loads(params_raw)
 100         video_data = params['video_data'][0]
 101         video_url = video_data.get('hd_src')
 102         if not video_url:
 103             video_url = video_data['sd_src']
 104         if not video_url:
 105             raise ExtractorError(u'Cannot find video URL')
 106         video_duration = int(video_data['video_duration'])
 107         thumbnail = video_data['thumbnail_src']
 108
 109         video_title = self._html_search_regex('<h2 class="uiHeaderTitle">([^<]+)</h2>',
 110             webpage, u'title')
 111
 112         info = {
 113             'id': video_id,
 114             'title': video_title,
 115             'url': video_url,
 116             'ext': 'mp4',
 117             'duration': video_duration,
 118             'thumbnail': thumbnail,
 119         }
 120         return [info]