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