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