youtube_dl/extractor/facebook.py

   1 import json
   2 import re
   3 import socket
   4
   5 from .common import InfoExtractor
   6 from ..utils import (
   7     compat_http_client,
   8     compat_str,
   9     compat_urllib_error,
  10     compat_urllib_parse,
  11     compat_urllib_request,
  12
  13     ExtractorError,
  14 )
  15
  16
  17 class FacebookIE(InfoExtractor):
  18     """Information Extractor for Facebook"""
  19
  20     _VALID_URL = r'''(?x)
  21         (?:https?://)?(?:\w+\.)?facebook\.com/
  22         (?:[^#?]*\#!/)?
  23         (?:video/video\.php|photo\.php|video/embed)\?(?:.*?)
  24         (?:v|video_id)=(?P<id>[0-9]+)
  25         (?:.*)'''
  26     _LOGIN_URL = 'https://www.facebook.com/login.php?next=http%3A%2F%2Ffacebook.com%2Fhome.php&login_attempt=1'
  27     _CHECKPOINT_URL = 'https://www.facebook.com/checkpoint/?next=http%3A%2F%2Ffacebook.com%2Fhome.php&_fb_noscript=1'
  28     _NETRC_MACHINE = 'facebook'
  29     IE_NAME = u'facebook'
  30     _TEST = {
  31         u'url': u'https://www.facebook.com/photo.php?v=120708114770723',
  32         u'file': u'120708114770723.mp4',
  33         u'md5': u'48975a41ccc4b7a581abd68651c1a5a8',
  34         u'info_dict': {
  35             u"duration": 279,
  36             u"title": u"PEOPLE ARE AWESOME 2013"
  37         }
  38     }
  39
  40     def report_login(self):
  41         """Report attempt to log in."""
  42         self.to_screen(u'Logging in')
  43
  44     def _login(self):
  45         (useremail, password) = self._get_login_info()
  46         if useremail is None:
  47             return
  48
  49         login_page_req = compat_urllib_request.Request(self._LOGIN_URL)
  50         login_page_req.add_header('Cookie', 'locale=en_US')
  51         self.report_login()
  52         login_page = self._download_webpage(login_page_req, None, note=False,
  53             errnote=u'Unable to download login page')
  54         lsd = self._search_regex(r'"lsd":"(\w*?)"', login_page, u'lsd')
  55         lgnrnd = self._search_regex(r'name="lgnrnd" value="([^"]*?)"', login_page, u'lgnrnd')
  56
  57         login_form = {
  58             'email': useremail,
  59             'pass': password,
  60             'lsd': lsd,
  61             'lgnrnd': lgnrnd,
  62             'next': 'http://facebook.com/home.php',
  63             'default_persistent': '0',
  64             'legacy_return': '1',
  65             'timezone': '-60',
  66             'trynum': '1',
  67             }
  68         request = compat_urllib_request.Request(self._LOGIN_URL, compat_urllib_parse.urlencode(login_form))
  69         request.add_header('Content-Type', 'application/x-www-form-urlencoded')
  70         try:
  71             login_results = compat_urllib_request.urlopen(request).read()
  72             if re.search(r'<form(.*)name="login"(.*)</form>', login_results) is not None:
  73                 self._downloader.report_warning(u'unable to log in: bad username/password, or exceded login rate limit (~3/min). Check credentials or wait.')
  74                 return
  75
  76             check_form = {
  77                 'fb_dtsg': self._search_regex(r'"fb_dtsg":"(.*?)"', login_results, u'fb_dtsg'),
  78                 'nh': self._search_regex(r'name="nh" value="(\w*?)"', login_results, u'nh'),
  79                 'name_action_selected': 'dont_save',
  80                 'submit[Continue]': self._search_regex(r'<input value="(.*?)" name="submit\[Continue\]"', login_results, u'continue'),
  81             }
  82             check_req = compat_urllib_request.Request(self._CHECKPOINT_URL, compat_urllib_parse.urlencode(check_form))
  83             check_req.add_header('Content-Type', 'application/x-www-form-urlencoded')
  84             check_response = compat_urllib_request.urlopen(check_req).read()
  85             if re.search(r'id="checkpointSubmitButton"', check_response) is not None:
  86                 self._downloader.report_warning(u'Unable to confirm login, you have to login in your brower and authorize the login.')
  87         except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
  88             self._downloader.report_warning(u'unable to log in: %s' % compat_str(err))
  89             return
  90
  91     def _real_initialize(self):
  92         self._login()
  93
  94     def _real_extract(self, url):
  95         mobj = re.match(self._VALID_URL, url)
  96         if mobj is None:
  97             raise ExtractorError(u'Invalid URL: %s' % url)
  98         video_id = mobj.group('id')
  99
 100         url = 'https://www.facebook.com/video/video.php?v=%s' % video_id
 101         webpage = self._download_webpage(url, video_id)
 102
 103         BEFORE = '{swf.addParam(param[0], param[1]);});\n'
 104         AFTER = '.forEach(function(variable) {swf.addVariable(variable[0], variable[1]);});'
 105         m = re.search(re.escape(BEFORE) + '(.*?)' + re.escape(AFTER), webpage)
 106         if not m:
 107             m_msg = re.search(r'class="[^"]*uiInterstitialContent[^"]*"><div>(.*?)</div>', webpage)
 108             if m_msg is not None:
 109                 raise ExtractorError(
 110                     u'The video is not available, Facebook said: "%s"' % m_msg.group(1),
 111                     expected=True)
 112             else:
 113                 raise ExtractorError(u'Cannot parse data')
 114         data = dict(json.loads(m.group(1)))
 115         params_raw = compat_urllib_parse.unquote(data['params'])
 116         params = json.loads(params_raw)
 117         video_data = params['video_data'][0]
 118         video_url = video_data.get('hd_src')
 119         if not video_url:
 120             video_url = video_data['sd_src']
 121         if not video_url:
 122             raise ExtractorError(u'Cannot find video URL')
 123         video_duration = int(video_data['video_duration'])
 124         thumbnail = video_data['thumbnail_src']
 125
 126         video_title = self._html_search_regex(
 127             r'<h2 class="uiHeaderTitle">([^<]*)</h2>', webpage, u'title')
 128
 129         info = {
 130             'id': video_id,
 131             'title': video_title,
 132             'url': video_url,
 133             'ext': 'mp4',
 134             'duration': video_duration,
 135             'thumbnail': thumbnail,
 136         }
 137         return [info]