Raphaël G. Git Repositories - youtubedl/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     ExtractorError,
  16     limit_length,
  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\.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     _TESTS = [{
  32         'url': 'https://www.facebook.com/video.php?v=637842556329505&fref=nf',
  33         'md5': '6a40d33c0eccbb1af76cf0485a052659',
  34         'info_dict': {
  35             'id': '637842556329505',
  36             'ext': 'mp4',
  37             'duration': 38,
  38             'title': 're:Did you know Kei Nishikori is the first Asian man to ever reach a Grand Slam',
  39         }
  40     }, {
  41         'note': 'Video without discernible title',
  42         'url': 'https://www.facebook.com/video.php?v=274175099429670',
  43         'info_dict': {
  44             'id': '274175099429670',
  45             'ext': 'mp4',
  46             'title': 'Facebook video #274175099429670',
  47         }
  48     }, {
  49         'url': 'https://www.facebook.com/video.php?v=10204634152394104',
  50         'only_matching': True,
  51     }]
  52
  53     def _login(self):
  54         (useremail, password) = self._get_login_info()
  55         if useremail is None:
  56             return
  57
  58         login_page_req = compat_urllib_request.Request(self._LOGIN_URL)
  59         login_page_req.add_header('Cookie', 'locale=en_US')
  60         login_page = self._download_webpage(login_page_req, None,
  61             note='Downloading login page',
  62             errnote='Unable to download login page')
  63         lsd = self._search_regex(
  64             r'<input type="hidden" name="lsd" value="([^"]*)"',
  65             login_page, 'lsd')
  66         lgnrnd = self._search_regex(r'name="lgnrnd" value="([^"]*?)"', login_page, 'lgnrnd')
  67
  68         login_form = {
  69             'email': useremail,
  70             'pass': password,
  71             'lsd': lsd,
  72             'lgnrnd': lgnrnd,
  73             'next': 'http://facebook.com/home.php',
  74             'default_persistent': '0',
  75             'legacy_return': '1',
  76             'timezone': '-60',
  77             'trynum': '1',
  78             }
  79         request = compat_urllib_request.Request(self._LOGIN_URL, urlencode_postdata(login_form))
  80         request.add_header('Content-Type', 'application/x-www-form-urlencoded')
  81         try:
  82             login_results = self._download_webpage(request, None,
  83                 note='Logging in', errnote='unable to fetch login page')
  84             if re.search(r'<form(.*)name="login"(.*)</form>', login_results) is not None:
  85                 self._downloader.report_warning('unable to log in: bad username/password, or exceded login rate limit (~3/min). Check credentials or wait.')
  86                 return
  87
  88             check_form = {
  89                 'fb_dtsg': self._search_regex(r'name="fb_dtsg" value="(.+?)"', login_results, 'fb_dtsg'),
  90                 'h': self._search_regex(
  91                     r'name="h"\s+(?:\w+="[^"]+"\s+)*?value="([^"]+)"', login_results, 'h'),
  92                 'name_action_selected': 'dont_save',
  93             }
  94             check_req = compat_urllib_request.Request(self._CHECKPOINT_URL, urlencode_postdata(check_form))
  95             check_req.add_header('Content-Type', 'application/x-www-form-urlencoded')
  96             check_response = self._download_webpage(check_req, None,
  97                 note='Confirming login')
  98             if re.search(r'id="checkpointSubmitButton"', check_response) is not None:
  99                 self._downloader.report_warning('Unable to confirm login, you have to login in your brower and authorize the login.')
 100         except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
 101             self._downloader.report_warning('unable to log in: %s' % compat_str(err))
 102             return
 103
 104     def _real_initialize(self):
 105         self._login()
 106
 107     def _real_extract(self, url):
 108         mobj = re.match(self._VALID_URL, url)
 109         video_id = mobj.group('id')
 110
 111         url = 'https://www.facebook.com/video/video.php?v=%s' % video_id
 112         webpage = self._download_webpage(url, video_id)
 113
 114         BEFORE = '{swf.addParam(param[0], param[1]);});\n'
 115         AFTER = '.forEach(function(variable) {swf.addVariable(variable[0], variable[1]);});'
 116         m = re.search(re.escape(BEFORE) + '(.*?)' + re.escape(AFTER), webpage)
 117         if not m:
 118             m_msg = re.search(r'class="[^"]*uiInterstitialContent[^"]*"><div>(.*?)</div>', webpage)
 119             if m_msg is not None:
 120                 raise ExtractorError(
 121                     'The video is not available, Facebook said: "%s"' % m_msg.group(1),
 122                     expected=True)
 123             else:
 124                 raise ExtractorError('Cannot parse data')
 125         data = dict(json.loads(m.group(1)))
 126         params_raw = compat_urllib_parse.unquote(data['params'])
 127         params = json.loads(params_raw)
 128         video_data = params['video_data'][0]
 129         video_url = video_data.get('hd_src')
 130         if not video_url:
 131             video_url = video_data['sd_src']
 132         if not video_url:
 133             raise ExtractorError('Cannot find video URL')
 134
 135         video_title = self._html_search_regex(
 136             r'<h2 class="uiHeaderTitle">([^<]*)</h2>', webpage, 'title',
 137             fatal=False)
 138         if not video_title:
 139             video_title = self._html_search_regex(
 140                 r'(?s)<span class="fbPhotosPhotoCaption".*?id="fbPhotoPageCaption"><span class="hasCaption">(.*?)</span>',
 141                 webpage, 'alternative title', default=None)
 142             video_title = limit_length(video_title, 80)
 143         if not video_title:
 144             video_title = 'Facebook video #%s' % video_id
 145
 146         return {
 147             'id': video_id,
 148             'title': video_title,
 149             'url': video_url,
 150             'duration': int(video_data['video_duration']),
 151             'thumbnail': video_data['thumbnail_src'],
 152         }