_ Git - youtube-dl/blob - youtube_dl/extractor/facebook.py

   1 from __future__ import unicode_literals
   2
   3 import json
   4 import re
   5 import socket
   6
   7 from .common import InfoExtractor
   8 from ..utils import (
   9     compat_http_client,
  10     compat_str,
  11     compat_urllib_error,
  12     compat_urllib_parse,
  13     compat_urllib_request,
  14     urlencode_postdata,
  15
  16     ExtractorError,
  17 )
  18
  19
  20 class FacebookIE(InfoExtractor):
  21     """Information Extractor for Facebook"""
  22
  23     _VALID_URL = r'''(?x)
  24         (?:https?://)?(?:\w+\.)?facebook\.com/
  25         (?:[^#?]*\#!/)?
  26         (?:video/video\.php|photo\.php|video/embed)\?(?:.*?)
  27         (?:v|video_id)=(?P<id>[0-9]+)
  28         (?:.*)'''
  29     _LOGIN_URL = 'https://www.facebook.com/login.php?next=http%3A%2F%2Ffacebook.com%2Fhome.php&login_attempt=1'
  30     _CHECKPOINT_URL = 'https://www.facebook.com/checkpoint/?next=http%3A%2F%2Ffacebook.com%2Fhome.php&_fb_noscript=1'
  31     _NETRC_MACHINE = 'facebook'
  32     IE_NAME = 'facebook'
  33     _TEST = {
  34         'url': 'https://www.facebook.com/photo.php?v=120708114770723',
  35         'md5': '48975a41ccc4b7a581abd68651c1a5a8',
  36         'info_dict': {
  37             'id': '120708114770723',
  38             'ext': 'mp4',
  39             'duration': 279,
  40             'title': 'PEOPLE ARE AWESOME 2013'
  41         }
  42     }
  43
  44     def report_login(self):
  45         """Report attempt to log in."""
  46         self.to_screen('Logging in')
  47
  48     def _login(self):
  49         (useremail, password) = self._get_login_info()
  50         if useremail is None:
  51             return
  52
  53         login_page_req = compat_urllib_request.Request(self._LOGIN_URL)
  54         login_page_req.add_header('Cookie', 'locale=en_US')
  55         login_page = self._download_webpage(login_page_req, None,
  56             note='Downloading login page',
  57             errnote='Unable to download login page')
  58         lsd = self._search_regex(
  59             r'<input type="hidden" name="lsd" value="([^"]*)"',
  60             login_page, 'lsd')
  61         lgnrnd = self._search_regex(r'name="lgnrnd" value="([^"]*?)"', login_page, 'lgnrnd')
  62
  63         login_form = {
  64             'email': useremail,
  65             'pass': password,
  66             'lsd': lsd,
  67             'lgnrnd': lgnrnd,
  68             'next': 'http://facebook.com/home.php',
  69             'default_persistent': '0',
  70             'legacy_return': '1',
  71             'timezone': '-60',
  72             'trynum': '1',
  73             }
  74         request = compat_urllib_request.Request(self._LOGIN_URL, urlencode_postdata(login_form))
  75         request.add_header('Content-Type', 'application/x-www-form-urlencoded')
  76         try:
  77             login_results = self._download_webpage(request, None,
  78                 note='Logging in', errnote='unable to fetch login page')
  79             if re.search(r'<form(.*)name="login"(.*)</form>', login_results) is not None:
  80                 self._downloader.report_warning('unable to log in: bad username/password, or exceded login rate limit (~3/min). Check credentials or wait.')
  81                 return
  82
  83             check_form = {
  84                 'fb_dtsg': self._search_regex(r'name="fb_dtsg" value="(.+?)"', login_results, 'fb_dtsg'),
  85                 'nh': self._search_regex(r'name="nh" value="(\w*?)"', login_results, 'nh'),
  86                 'name_action_selected': 'dont_save',
  87                 'submit[Continue]': self._search_regex(r'<button[^>]+value="(.*?)"[^>]+name="submit\[Continue\]"', login_results, 'continue'),
  88             }
  89             check_req = compat_urllib_request.Request(self._CHECKPOINT_URL, urlencode_postdata(check_form))
  90             check_req.add_header('Content-Type', 'application/x-www-form-urlencoded')
  91             check_response = self._download_webpage(check_req, None,
  92                 note='Confirming login')
  93             if re.search(r'id="checkpointSubmitButton"', check_response) is not None:
  94                 self._downloader.report_warning('Unable to confirm login, you have to login in your brower and authorize the login.')
  95         except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
  96             self._downloader.report_warning('unable to log in: %s' % compat_str(err))
  97             return
  98
  99     def _real_initialize(self):
 100         self._login()
 101
 102     def _real_extract(self, url):
 103         mobj = re.match(self._VALID_URL, url)
 104         if mobj is None:
 105             raise ExtractorError('Invalid URL: %s' % url)
 106         video_id = mobj.group('id')
 107
 108         url = 'https://www.facebook.com/video/video.php?v=%s' % video_id
 109         webpage = self._download_webpage(url, video_id)
 110
 111         BEFORE = '{swf.addParam(param[0], param[1]);});\n'
 112         AFTER = '.forEach(function(variable) {swf.addVariable(variable[0], variable[1]);});'
 113         m = re.search(re.escape(BEFORE) + '(.*?)' + re.escape(AFTER), webpage)
 114         if not m:
 115             m_msg = re.search(r'class="[^"]*uiInterstitialContent[^"]*"><div>(.*?)</div>', webpage)
 116             if m_msg is not None:
 117                 raise ExtractorError(
 118                     'The video is not available, Facebook said: "%s"' % m_msg.group(1),
 119                     expected=True)
 120             else:
 121                 raise ExtractorError('Cannot parse data')
 122         data = dict(json.loads(m.group(1)))
 123         params_raw = compat_urllib_parse.unquote(data['params'])
 124         params = json.loads(params_raw)
 125         video_data = params['video_data'][0]
 126         video_url = video_data.get('hd_src')
 127         if not video_url:
 128             video_url = video_data['sd_src']
 129         if not video_url:
 130             raise ExtractorError('Cannot find video URL')
 131         video_duration = int(video_data['video_duration'])
 132         thumbnail = video_data['thumbnail_src']
 133
 134         video_title = self._html_search_regex(
 135             r'<h2 class="uiHeaderTitle">([^<]*)</h2>', webpage, 'title')
 136
 137         info = {
 138             'id': video_id,
 139             'title': video_title,
 140             'url': video_url,
 141             'ext': 'mp4',
 142             'duration': video_duration,
 143             'thumbnail': thumbnail,
 144         }
 145         return [info]