Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/facebook.py

   1 import json
   2 import netrc
   3 import re
   4 import socket
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     compat_http_client,
   9     compat_str,
  10     compat_urllib_error,
  11     compat_urllib_parse,
  12     compat_urllib_request,
  13
  14     ExtractorError,
  15 )
  16
  17
  18 class FacebookIE(InfoExtractor):
  19     """Information Extractor for Facebook"""
  20
  21     _VALID_URL = r'^(?:https?://)?(?:\w+\.)?facebook\.com/(?:video/video|photo)\.php\?(?:.*?)v=(?P<ID>\d+)(?:.*)'
  22     _LOGIN_URL = 'https://www.facebook.com/login.php?next=http%3A%2F%2Ffacebook.com%2Fhome.php&login_attempt=1'
  23     _CHECKPOINT_URL = 'https://www.facebook.com/checkpoint/?next=http%3A%2F%2Ffacebook.com%2Fhome.php&_fb_noscript=1'
  24     _NETRC_MACHINE = 'facebook'
  25     IE_NAME = u'facebook'
  26     _TEST = {
  27         u'url': u'https://www.facebook.com/photo.php?v=120708114770723',
  28         u'file': u'120708114770723.mp4',
  29         u'md5': u'48975a41ccc4b7a581abd68651c1a5a8',
  30         u'info_dict': {
  31             u"duration": 279,
  32             u"title": u"PEOPLE ARE AWESOME 2013"
  33         }
  34     }
  35
  36     def report_login(self):
  37         """Report attempt to log in."""
  38         self.to_screen(u'Logging in')
  39
  40     def _login(self):
  41         (useremail, password) = self._get_login_info()
  42         if useremail is None:
  43             return
  44
  45         login_page_req = compat_urllib_request.Request(self._LOGIN_URL)
  46         login_page_req.add_header('Cookie', 'locale=en_US')
  47         self.report_login()
  48         login_page = self._download_webpage(login_page_req, None, note=False,
  49             errnote=u'Unable to download login page')
  50         lsd = self._search_regex(r'"lsd":"(\w*?)"', login_page, u'lsd')
  51         lgnrnd = self._search_regex(r'name="lgnrnd" value="([^"]*?)"', login_page, u'lgnrnd')
  52
  53         login_form = {
  54             'email': useremail,
  55             'pass': password,
  56             'lsd': lsd,
  57             'lgnrnd': lgnrnd,
  58             'next': 'http://facebook.com/home.php',
  59             'default_persistent': '0',
  60             'legacy_return': '1',
  61             'timezone': '-60',
  62             'trynum': '1',
  63             }
  64         request = compat_urllib_request.Request(self._LOGIN_URL, compat_urllib_parse.urlencode(login_form))
  65         request.add_header('Content-Type', 'application/x-www-form-urlencoded')
  66         try:
  67             login_results = compat_urllib_request.urlopen(request).read()
  68             if re.search(r'<form(.*)name="login"(.*)</form>', login_results) is not None:
  69                 self._downloader.report_warning(u'unable to log in: bad username/password, or exceded login rate limit (~3/min). Check credentials or wait.')
  70                 return
  71
  72             check_form = {
  73                 'fb_dtsg': self._search_regex(r'"fb_dtsg":"(.*?)"', login_results, u'fb_dtsg'),
  74                 'nh': self._search_regex(r'name="nh" value="(\w*?)"', login_results, u'nh'),
  75                 'name_action_selected': 'dont_save',
  76                 'submit[Continue]': self._search_regex(r'<input value="(.*?)" name="submit\[Continue\]"', login_results, u'continue'),
  77             }
  78             check_req = compat_urllib_request.Request(self._CHECKPOINT_URL, compat_urllib_parse.urlencode(check_form))
  79             check_req.add_header('Content-Type', 'application/x-www-form-urlencoded')
  80             check_response = compat_urllib_request.urlopen(check_req).read()
  81             if re.search(r'id="checkpointSubmitButton"', check_response) is not None:
  82                 self._downloader.report_warning(u'Unable to confirm login, you have to login in your brower and authorize the login.')
  83         except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
  84             self._downloader.report_warning(u'unable to log in: %s' % compat_str(err))
  85             return
  86
  87     def _real_initialize(self):
  88         self._login()
  89
  90     def _real_extract(self, url):
  91         mobj = re.match(self._VALID_URL, url)
  92         if mobj is None:
  93             raise ExtractorError(u'Invalid URL: %s' % url)
  94         video_id = mobj.group('ID')
  95
  96         url = 'https://www.facebook.com/video/video.php?v=%s' % video_id
  97         webpage = self._download_webpage(url, video_id)
  98
  99         BEFORE = '{swf.addParam(param[0], param[1]);});\n'
 100         AFTER = '.forEach(function(variable) {swf.addVariable(variable[0], variable[1]);});'
 101         m = re.search(re.escape(BEFORE) + '(.*?)' + re.escape(AFTER), webpage)
 102         if not m:
 103             m_msg = re.search(r'class="[^"]*uiInterstitialContent[^"]*"><div>(.*?)</div>', webpage)
 104             if m_msg is not None:
 105                 raise ExtractorError(
 106                     u'The video is not available, Facebook said: "%s"' % m_msg.group(1),
 107                     expected=True)
 108             else:
 109                 raise ExtractorError(u'Cannot parse data')
 110         data = dict(json.loads(m.group(1)))
 111         params_raw = compat_urllib_parse.unquote(data['params'])
 112         params = json.loads(params_raw)
 113         video_data = params['video_data'][0]
 114         video_url = video_data.get('hd_src')
 115         if not video_url:
 116             video_url = video_data['sd_src']
 117         if not video_url:
 118             raise ExtractorError(u'Cannot find video URL')
 119         video_duration = int(video_data['video_duration'])
 120         thumbnail = video_data['thumbnail_src']
 121
 122         video_title = self._html_search_regex(
 123             r'<h2 class="uiHeaderTitle">([^<]*)</h2>', webpage, u'title')
 124
 125         info = {
 126             'id': video_id,
 127             'title': video_title,
 128             'url': video_url,
 129             'ext': 'mp4',
 130             'duration': video_duration,
 131             'thumbnail': thumbnail,
 132         }
 133         return [info]