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