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 ..compat import (
   9     compat_http_client,
  10     compat_urllib_error,
  11     compat_urllib_parse_unquote,
  12 )
  13 from ..utils import (
  14     error_to_compat_str,
  15     ExtractorError,
  16     limit_length,
  17     sanitized_Request,
  18     urlencode_postdata,
  19     get_element_by_id,
  20     clean_html,
  21 )
  22
  23
  24 class FacebookIE(InfoExtractor):
  25     _VALID_URL = r'''(?x)
  26         https?://(?:\w+\.)?facebook\.com/
  27         (?:[^#]*?\#!/)?
  28         (?:
  29             (?:video/video\.php|photo\.php|video\.php|video/embed)\?(?:.*?)
  30             (?:v|video_id)=|
  31             [^/]+/videos/(?:[^/]+/)?
  32         )
  33         (?P<id>[0-9]+)
  34         (?:.*)'''
  35     _LOGIN_URL = 'https://www.facebook.com/login.php?next=http%3A%2F%2Ffacebook.com%2Fhome.php&login_attempt=1'
  36     _CHECKPOINT_URL = 'https://www.facebook.com/checkpoint/?next=http%3A%2F%2Ffacebook.com%2Fhome.php&_fb_noscript=1'
  37     _NETRC_MACHINE = 'facebook'
  38     IE_NAME = 'facebook'
  39     _TESTS = [{
  40         'url': 'https://www.facebook.com/video.php?v=637842556329505&fref=nf',
  41         'md5': '6a40d33c0eccbb1af76cf0485a052659',
  42         'info_dict': {
  43             'id': '637842556329505',
  44             'ext': 'mp4',
  45             'title': 're:Did you know Kei Nishikori is the first Asian man to ever reach a Grand Slam',
  46             'uploader': 'Tennis on Facebook',
  47         }
  48     }, {
  49         'note': 'Video without discernible title',
  50         'url': 'https://www.facebook.com/video.php?v=274175099429670',
  51         'info_dict': {
  52             'id': '274175099429670',
  53             'ext': 'mp4',
  54             'title': 'Facebook video #274175099429670',
  55             'uploader': 'Asif Nawab Butt',
  56         },
  57         'expected_warnings': [
  58             'title'
  59         ]
  60     }, {
  61         'url': 'https://www.facebook.com/video.php?v=10204634152394104',
  62         'only_matching': True,
  63     }, {
  64         'url': 'https://www.facebook.com/amogood/videos/1618742068337349/?fref=nf',
  65         'only_matching': True,
  66     }, {
  67         'url': 'https://www.facebook.com/ChristyClarkForBC/videos/vb.22819070941/10153870694020942/?type=2&theater',
  68         'only_matching': True,
  69     }]
  70
  71     def _login(self):
  72         (useremail, password) = self._get_login_info()
  73         if useremail is None:
  74             return
  75
  76         login_page_req = sanitized_Request(self._LOGIN_URL)
  77         self._set_cookie('facebook.com', 'locale', 'en_US')
  78         login_page = self._download_webpage(login_page_req, None,
  79                                             note='Downloading login page',
  80                                             errnote='Unable to download login page')
  81         lsd = self._search_regex(
  82             r'<input type="hidden" name="lsd" value="([^"]*)"',
  83             login_page, 'lsd')
  84         lgnrnd = self._search_regex(r'name="lgnrnd" value="([^"]*?)"', login_page, 'lgnrnd')
  85
  86         login_form = {
  87             'email': useremail,
  88             'pass': password,
  89             'lsd': lsd,
  90             'lgnrnd': lgnrnd,
  91             'next': 'http://facebook.com/home.php',
  92             'default_persistent': '0',
  93             'legacy_return': '1',
  94             'timezone': '-60',
  95             'trynum': '1',
  96         }
  97         request = sanitized_Request(self._LOGIN_URL, urlencode_postdata(login_form))
  98         request.add_header('Content-Type', 'application/x-www-form-urlencoded')
  99         try:
 100             login_results = self._download_webpage(request, None,
 101                                                    note='Logging in', errnote='unable to fetch login page')
 102             if re.search(r'<form(.*)name="login"(.*)</form>', login_results) is not None:
 103                 error = self._html_search_regex(
 104                     r'(?s)<div[^>]+class=(["\']).*?login_error_box.*?\1[^>]*><div[^>]*>.*?</div><div[^>]*>(?P<error>.+?)</div>',
 105                     login_results, 'login error', default=None, group='error')
 106                 if error:
 107                     raise ExtractorError('Unable to login: %s' % error, expected=True)
 108                 self._downloader.report_warning('unable to log in: bad username/password, or exceded login rate limit (~3/min). Check credentials or wait.')
 109                 return
 110
 111             fb_dtsg = self._search_regex(
 112                 r'name="fb_dtsg" value="(.+?)"', login_results, 'fb_dtsg', default=None)
 113             h = self._search_regex(
 114                 r'name="h"\s+(?:\w+="[^"]+"\s+)*?value="([^"]+)"', login_results, 'h', default=None)
 115
 116             if not fb_dtsg or not h:
 117                 return
 118
 119             check_form = {
 120                 'fb_dtsg': fb_dtsg,
 121                 'h': h,
 122                 'name_action_selected': 'dont_save',
 123             }
 124             check_req = sanitized_Request(self._CHECKPOINT_URL, urlencode_postdata(check_form))
 125             check_req.add_header('Content-Type', 'application/x-www-form-urlencoded')
 126             check_response = self._download_webpage(check_req, None,
 127                                                     note='Confirming login')
 128             if re.search(r'id="checkpointSubmitButton"', check_response) is not None:
 129                 self._downloader.report_warning('Unable to confirm login, you have to login in your brower and authorize the login.')
 130         except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
 131             self._downloader.report_warning('unable to log in: %s' % error_to_compat_str(err))
 132             return
 133
 134     def _real_initialize(self):
 135         self._login()
 136
 137     def _real_extract(self, url):
 138         video_id = self._match_id(url)
 139         url = 'https://www.facebook.com/video/video.php?v=%s' % video_id
 140         webpage = self._download_webpage(url, video_id)
 141
 142         BEFORE = '{swf.addParam(param[0], param[1]);});\n'
 143         AFTER = '.forEach(function(variable) {swf.addVariable(variable[0], variable[1]);});'
 144         m = re.search(re.escape(BEFORE) + '(.*?)' + re.escape(AFTER), webpage)
 145         if not m:
 146             m_msg = re.search(r'class="[^"]*uiInterstitialContent[^"]*"><div>(.*?)</div>', webpage)
 147             if m_msg is not None:
 148                 raise ExtractorError(
 149                     'The video is not available, Facebook said: "%s"' % m_msg.group(1),
 150                     expected=True)
 151             else:
 152                 raise ExtractorError('Cannot parse data')
 153         data = dict(json.loads(m.group(1)))
 154         params_raw = compat_urllib_parse_unquote(data['params'])
 155         params = json.loads(params_raw)
 156
 157         formats = []
 158         for format_id, f in params['video_data'].items():
 159             if not f or not isinstance(f, list):
 160                 continue
 161             for quality in ('sd', 'hd'):
 162                 for src_type in ('src', 'src_no_ratelimit'):
 163                     src = f[0].get('%s_%s' % (quality, src_type))
 164                     if src:
 165                         formats.append({
 166                             'format_id': '%s_%s_%s' % (format_id, quality, src_type),
 167                             'url': src,
 168                             'preference': -10 if format_id == 'progressive' else 0,
 169                         })
 170         if not formats:
 171             raise ExtractorError('Cannot find video formats')
 172
 173         video_title = self._html_search_regex(
 174             r'<h2\s+[^>]*class="uiHeaderTitle"[^>]*>([^<]*)</h2>', webpage, 'title',
 175             default=None)
 176         if not video_title:
 177             video_title = self._html_search_regex(
 178                 r'(?s)<span class="fbPhotosPhotoCaption".*?id="fbPhotoPageCaption"><span class="hasCaption">(.*?)</span>',
 179                 webpage, 'alternative title', default=None)
 180             video_title = limit_length(video_title, 80)
 181         if not video_title:
 182             video_title = 'Facebook video #%s' % video_id
 183         uploader = clean_html(get_element_by_id('fbPhotoPageAuthorName', webpage))
 184
 185         return {
 186             'id': video_id,
 187             'title': video_title,
 188             'formats': formats,
 189             'uploader': uploader,
 190         }