_ 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
  15     ExtractorError,
  16 )
  17
  18
  19 class FacebookIE(InfoExtractor):
  20     """Information Extractor for Facebook"""
  21
  22     _VALID_URL = r'''(?x)
  23         (?:https?://)?(?:\w+\.)?facebook\.com/
  24         (?:[^#?]*\#!/)?
  25         (?:video/video\.php|photo\.php|video/embed)\?(?:.*?)
  26         (?:v|video_id)=(?P<id>[0-9]+)
  27         (?:.*)'''
  28     _LOGIN_URL = 'https://www.facebook.com/login.php?next=http%3A%2F%2Ffacebook.com%2Fhome.php&login_attempt=1'
  29     _CHECKPOINT_URL = 'https://www.facebook.com/checkpoint/?next=http%3A%2F%2Ffacebook.com%2Fhome.php&_fb_noscript=1'
  30     _NETRC_MACHINE = 'facebook'
  31     IE_NAME = 'facebook'
  32     _TEST = {
  33         'url': 'https://www.facebook.com/photo.php?v=120708114770723',
  34         'md5': '48975a41ccc4b7a581abd68651c1a5a8',
  35         'info_dict': {
  36             'id': '120708114770723',
  37             'ext': 'mp4',
  38             'duration': 279,
  39             'title': 'PEOPLE ARE AWESOME 2013'
  40         }
  41     }
  42
  43     def report_login(self):
  44         """Report attempt to log in."""
  45         self.to_screen('Logging in')
  46
  47     def _login(self):
  48         (useremail, password) = self._get_login_info()
  49         if useremail is None:
  50             return
  51
  52         login_page_req = compat_urllib_request.Request(self._LOGIN_URL)
  53         login_page_req.add_header('Cookie', 'locale=en_US')
  54         self.report_login()
  55         login_page = self._download_webpage(login_page_req, None, note=False,
  56             errnote='Unable to download login page')
  57         lsd = self._search_regex(
  58             r'<input type="hidden" name="lsd" value="([^"]*)"',
  59             login_page, 'lsd')
  60         lgnrnd = self._search_regex(r'name="lgnrnd" value="([^"]*?)"', login_page, 'lgnrnd')
  61
  62         login_form = {
  63             'email': useremail,
  64             'pass': password,
  65             'lsd': lsd,
  66             'lgnrnd': lgnrnd,
  67             'next': 'http://facebook.com/home.php',
  68             'default_persistent': '0',
  69             'legacy_return': '1',
  70             'timezone': '-60',
  71             'trynum': '1',
  72             }
  73         request = compat_urllib_request.Request(self._LOGIN_URL, compat_urllib_parse.urlencode(login_form))
  74         request.add_header('Content-Type', 'application/x-www-form-urlencoded')
  75         try:
  76             login_results = compat_urllib_request.urlopen(request).read()
  77             if re.search(r'<form(.*)name="login"(.*)</form>', login_results) is not None:
  78                 self._downloader.report_warning('unable to log in: bad username/password, or exceded login rate limit (~3/min). Check credentials or wait.')
  79                 return
  80
  81             check_form = {
  82                 'fb_dtsg': self._search_regex(r'"fb_dtsg":"(.*?)"', login_results, 'fb_dtsg'),
  83                 'nh': self._search_regex(r'name="nh" value="(\w*?)"', login_results, 'nh'),
  84                 'name_action_selected': 'dont_save',
  85                 'submit[Continue]': self._search_regex(r'<input value="(.*?)" name="submit\[Continue\]"', login_results, 'continue'),
  86             }
  87             check_req = compat_urllib_request.Request(self._CHECKPOINT_URL, compat_urllib_parse.urlencode(check_form))
  88             check_req.add_header('Content-Type', 'application/x-www-form-urlencoded')
  89             check_response = compat_urllib_request.urlopen(check_req).read()
  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         if mobj is None:
 102             raise ExtractorError('Invalid URL: %s' % url)
 103         video_id = mobj.group('id')
 104
 105         url = 'https://www.facebook.com/video/video.php?v=%s' % video_id
 106         webpage = self._download_webpage(url, video_id)
 107
 108         BEFORE = '{swf.addParam(param[0], param[1]);});\n'
 109         AFTER = '.forEach(function(variable) {swf.addVariable(variable[0], variable[1]);});'
 110         m = re.search(re.escape(BEFORE) + '(.*?)' + re.escape(AFTER), webpage)
 111         if not m:
 112             m_msg = re.search(r'class="[^"]*uiInterstitialContent[^"]*"><div>(.*?)</div>', webpage)
 113             if m_msg is not None:
 114                 raise ExtractorError(
 115                     'The video is not available, Facebook said: "%s"' % m_msg.group(1),
 116                     expected=True)
 117             else:
 118                 raise ExtractorError('Cannot parse data')
 119         data = dict(json.loads(m.group(1)))
 120         params_raw = compat_urllib_parse.unquote(data['params'])
 121         params = json.loads(params_raw)
 122         video_data = params['video_data'][0]
 123         video_url = video_data.get('hd_src')
 124         if not video_url:
 125             video_url = video_data['sd_src']
 126         if not video_url:
 127             raise ExtractorError('Cannot find video URL')
 128         video_duration = int(video_data['video_duration'])
 129         thumbnail = video_data['thumbnail_src']
 130
 131         video_title = self._html_search_regex(
 132             r'<h2 class="uiHeaderTitle">([^<]*)</h2>', webpage, 'title')
 133
 134         info = {
 135             'id': video_id,
 136             'title': video_title,
 137             'url': video_url,
 138             'ext': 'mp4',
 139             'duration': video_duration,
 140             'thumbnail': thumbnail,
 141         }
 142         return [info]