_ 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 ..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\.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/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         'url': 'https://www.facebook.com/video.php?v=10204634152394104',
  42         'only_matching': True,
  43     }]
  44
  45     def _login(self):
  46         (useremail, password) = self._get_login_info()
  47         if useremail is None:
  48             return
  49
  50         login_page_req = compat_urllib_request.Request(self._LOGIN_URL)
  51         login_page_req.add_header('Cookie', 'locale=en_US')
  52         login_page = self._download_webpage(login_page_req, None,
  53             note='Downloading login page',
  54             errnote='Unable to download login page')
  55         lsd = self._search_regex(
  56             r'<input type="hidden" name="lsd" value="([^"]*)"',
  57             login_page, 'lsd')
  58         lgnrnd = self._search_regex(r'name="lgnrnd" value="([^"]*?)"', login_page, 'lgnrnd')
  59
  60         login_form = {
  61             'email': useremail,
  62             'pass': password,
  63             'lsd': lsd,
  64             'lgnrnd': lgnrnd,
  65             'next': 'http://facebook.com/home.php',
  66             'default_persistent': '0',
  67             'legacy_return': '1',
  68             'timezone': '-60',
  69             'trynum': '1',
  70             }
  71         request = compat_urllib_request.Request(self._LOGIN_URL, urlencode_postdata(login_form))
  72         request.add_header('Content-Type', 'application/x-www-form-urlencoded')
  73         try:
  74             login_results = self._download_webpage(request, None,
  75                 note='Logging in', errnote='unable to fetch login page')
  76             if re.search(r'<form(.*)name="login"(.*)</form>', login_results) is not None:
  77                 self._downloader.report_warning('unable to log in: bad username/password, or exceded login rate limit (~3/min). Check credentials or wait.')
  78                 return
  79
  80             check_form = {
  81                 'fb_dtsg': self._search_regex(r'name="fb_dtsg" value="(.+?)"', login_results, 'fb_dtsg'),
  82                 'h': self._search_regex(
  83                     r'name="h"\s+(?:\w+="[^"]+"\s+)*?value="([^"]+)"', login_results, 'h'),
  84                 'name_action_selected': 'dont_save',
  85             }
  86             check_req = compat_urllib_request.Request(self._CHECKPOINT_URL, urlencode_postdata(check_form))
  87             check_req.add_header('Content-Type', 'application/x-www-form-urlencoded')
  88             check_response = self._download_webpage(check_req, None,
  89                 note='Confirming login')
  90             if re.search(r'id="checkpointSubmitButton"', check_response) is not None:
  91                 self._downloader.report_warning('Unable to confirm login, you have to login in your brower and authorize the login.')
  92         except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
  93             self._downloader.report_warning('unable to log in: %s' % compat_str(err))
  94             return
  95
  96     def _real_initialize(self):
  97         self._login()
  98
  99     def _real_extract(self, url):
 100         mobj = re.match(self._VALID_URL, url)
 101         video_id = mobj.group('id')
 102
 103         url = 'https://www.facebook.com/video/video.php?v=%s' % video_id
 104         webpage = self._download_webpage(url, video_id)
 105
 106         BEFORE = '{swf.addParam(param[0], param[1]);});\n'
 107         AFTER = '.forEach(function(variable) {swf.addVariable(variable[0], variable[1]);});'
 108         m = re.search(re.escape(BEFORE) + '(.*?)' + re.escape(AFTER), webpage)
 109         if not m:
 110             m_msg = re.search(r'class="[^"]*uiInterstitialContent[^"]*"><div>(.*?)</div>', webpage)
 111             if m_msg is not None:
 112                 raise ExtractorError(
 113                     'The video is not available, Facebook said: "%s"' % m_msg.group(1),
 114                     expected=True)
 115             else:
 116                 raise ExtractorError('Cannot parse data')
 117         data = dict(json.loads(m.group(1)))
 118         params_raw = compat_urllib_parse.unquote(data['params'])
 119         params = json.loads(params_raw)
 120         video_data = params['video_data'][0]
 121         video_url = video_data.get('hd_src')
 122         if not video_url:
 123             video_url = video_data['sd_src']
 124         if not video_url:
 125             raise ExtractorError('Cannot find video URL')
 126
 127         video_title = self._html_search_regex(
 128             r'<h2 class="uiHeaderTitle">([^<]*)</h2>', webpage, 'title')
 129
 130         return {
 131             'id': video_id,
 132             'title': video_title,
 133             'url': video_url,
 134             'duration': int(video_data['video_duration']),
 135             'thumbnail': video_data['thumbnail_src'],
 136         }