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

   1 # encoding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5 import time
   6 import hashlib
   7
   8 from .common import InfoExtractor
   9 from ..compat import (
  10     compat_str,
  11     compat_urlparse,
  12 )
  13 from ..utils import (
  14     clean_html,
  15     ExtractorError,
  16     int_or_none,
  17     float_or_none,
  18     parse_iso8601,
  19     sanitized_Request,
  20     urlencode_postdata,
  21 )
  22
  23
  24 class NocoIE(InfoExtractor):
  25     _VALID_URL = r'https?://(?:(?:www\.)?noco\.tv/emission/|player\.noco\.tv/\?idvideo=)(?P<id>\d+)'
  26     _LOGIN_URL = 'http://noco.tv/do.php'
  27     _API_URL_TEMPLATE = 'https://api.noco.tv/1.1/%s?ts=%s&tk=%s'
  28     _SUB_LANG_TEMPLATE = '&sub_lang=%s'
  29     _NETRC_MACHINE = 'noco'
  30
  31     _TESTS = [
  32         {
  33             'url': 'http://noco.tv/emission/11538/nolife/ami-ami-idol-hello-france/',
  34             'md5': '0a993f0058ddbcd902630b2047ef710e',
  35             'info_dict': {
  36                 'id': '11538',
  37                 'ext': 'mp4',
  38                 'title': 'Ami Ami Idol - Hello! France',
  39                 'description': 'md5:4eaab46ab68fa4197a317a88a53d3b86',
  40                 'upload_date': '20140412',
  41                 'uploader': 'Nolife',
  42                 'uploader_id': 'NOL',
  43                 'duration': 2851.2,
  44             },
  45             'skip': 'Requires noco account',
  46         },
  47         {
  48             'url': 'http://noco.tv/emission/12610/lbl42/the-guild/s01e01-wake-up-call',
  49             'md5': 'c190f1f48e313c55838f1f412225934d',
  50             'info_dict': {
  51                 'id': '12610',
  52                 'ext': 'mp4',
  53                 'title': 'The Guild #1 - Wake-Up Call',
  54                 'timestamp': 1403863200,
  55                 'upload_date': '20140627',
  56                 'uploader': 'LBL42',
  57                 'uploader_id': 'LBL',
  58                 'duration': 233.023,
  59             },
  60             'skip': 'Requires noco account',
  61         }
  62     ]
  63
  64     def _real_initialize(self):
  65         self._login()
  66
  67     def _login(self):
  68         (username, password) = self._get_login_info()
  69         if username is None:
  70             return
  71
  72         login_form = {
  73             'a': 'login',
  74             'cookie': '1',
  75             'username': username,
  76             'password': password,
  77         }
  78         request = sanitized_Request(self._LOGIN_URL, urlencode_postdata(login_form))
  79         request.add_header('Content-Type', 'application/x-www-form-urlencoded; charset=UTF-8')
  80
  81         login = self._download_json(request, None, 'Logging in as %s' % username)
  82
  83         if 'erreur' in login:
  84             raise ExtractorError('Unable to login: %s' % clean_html(login['erreur']), expected=True)
  85
  86     @staticmethod
  87     def _ts():
  88         return int(time.time() * 1000)
  89
  90     def _call_api(self, path, video_id, note, sub_lang=None):
  91         ts = compat_str(self._ts() + self._ts_offset)
  92         tk = hashlib.md5((hashlib.md5(ts.encode('ascii')).hexdigest() + '#8S?uCraTedap6a').encode('ascii')).hexdigest()
  93         url = self._API_URL_TEMPLATE % (path, ts, tk)
  94         if sub_lang:
  95             url += self._SUB_LANG_TEMPLATE % sub_lang
  96
  97         request = sanitized_Request(url)
  98         request.add_header('Referer', self._referer)
  99
 100         resp = self._download_json(request, video_id, note)
 101
 102         if isinstance(resp, dict) and resp.get('error'):
 103             self._raise_error(resp['error'], resp['description'])
 104
 105         return resp
 106
 107     def _raise_error(self, error, description):
 108         raise ExtractorError(
 109             '%s returned error: %s - %s' % (self.IE_NAME, error, description),
 110             expected=True)
 111
 112     def _real_extract(self, url):
 113         video_id = self._match_id(url)
 114
 115         # Timestamp adjustment offset between server time and local time
 116         # must be calculated in order to use timestamps closest to server's
 117         # in all API requests (see https://github.com/rg3/youtube-dl/issues/7864)
 118         webpage = self._download_webpage(url, video_id)
 119
 120         player_url = self._search_regex(
 121             r'(["\'])(?P<player>https?://noco\.tv/(?:[^/]+/)+NocoPlayer.+?\.swf.*?)\1',
 122             webpage, 'noco player', group='player',
 123             default='http://noco.tv/cdata/js/player/NocoPlayer-v1.2.40.swf')
 124
 125         qs = compat_urlparse.parse_qs(compat_urlparse.urlparse(player_url).query)
 126         ts = int_or_none(qs.get('ts', [None])[0])
 127         self._ts_offset = ts - self._ts() if ts else 0
 128         self._referer = player_url
 129
 130         medias = self._call_api(
 131             'shows/%s/medias' % video_id,
 132             video_id, 'Downloading video JSON')
 133
 134         show = self._call_api(
 135             'shows/by_id/%s' % video_id,
 136             video_id, 'Downloading show JSON')[0]
 137
 138         options = self._call_api(
 139             'users/init', video_id,
 140             'Downloading user options JSON')['options']
 141         audio_lang_pref = options.get('audio_language') or options.get('language', 'fr')
 142
 143         if audio_lang_pref == 'original':
 144             audio_lang_pref = show['original_lang']
 145         if len(medias) == 1:
 146             audio_lang_pref = list(medias.keys())[0]
 147         elif audio_lang_pref not in medias:
 148             audio_lang_pref = 'fr'
 149
 150         qualities = self._call_api(
 151             'qualities',
 152             video_id, 'Downloading qualities JSON')
 153
 154         formats = []
 155
 156         for audio_lang, audio_lang_dict in medias.items():
 157             preference = 1 if audio_lang == audio_lang_pref else 0
 158             for sub_lang, lang_dict in audio_lang_dict['video_list'].items():
 159                 for format_id, fmt in lang_dict['quality_list'].items():
 160                     format_id_extended = 'audio-%s_sub-%s_%s' % (audio_lang, sub_lang, format_id)
 161
 162                     video = self._call_api(
 163                         'shows/%s/video/%s/%s' % (video_id, format_id.lower(), audio_lang),
 164                         video_id, 'Downloading %s video JSON' % format_id_extended,
 165                         sub_lang if sub_lang != 'none' else None)
 166
 167                     file_url = video['file']
 168                     if not file_url:
 169                         continue
 170
 171                     if file_url in ['forbidden', 'not found']:
 172                         popmessage = video['popmessage']
 173                         self._raise_error(popmessage['title'], popmessage['message'])
 174
 175                     formats.append({
 176                         'url': file_url,
 177                         'format_id': format_id_extended,
 178                         'width': int_or_none(fmt.get('res_width')),
 179                         'height': int_or_none(fmt.get('res_lines')),
 180                         'abr': int_or_none(fmt.get('audiobitrate'), 1000),
 181                         'vbr': int_or_none(fmt.get('videobitrate'), 1000),
 182                         'filesize': int_or_none(fmt.get('filesize')),
 183                         'format_note': qualities[format_id].get('quality_name'),
 184                         'quality': qualities[format_id].get('priority'),
 185                         'preference': preference,
 186                     })
 187
 188         self._sort_formats(formats)
 189
 190         timestamp = parse_iso8601(show.get('online_date_start_utc'), ' ')
 191
 192         if timestamp is not None and timestamp < 0:
 193             timestamp = None
 194
 195         uploader = show.get('partner_name')
 196         uploader_id = show.get('partner_key')
 197         duration = float_or_none(show.get('duration_ms'), 1000)
 198
 199         thumbnails = []
 200         for thumbnail_key, thumbnail_url in show.items():
 201             m = re.search(r'^screenshot_(?P<width>\d+)x(?P<height>\d+)$', thumbnail_key)
 202             if not m:
 203                 continue
 204             thumbnails.append({
 205                 'url': thumbnail_url,
 206                 'width': int(m.group('width')),
 207                 'height': int(m.group('height')),
 208             })
 209
 210         episode = show.get('show_TT') or show.get('show_OT')
 211         family = show.get('family_TT') or show.get('family_OT')
 212         episode_number = show.get('episode_number')
 213
 214         title = ''
 215         if family:
 216             title += family
 217         if episode_number:
 218             title += ' #' + compat_str(episode_number)
 219         if episode:
 220             title += ' - ' + compat_str(episode)
 221
 222         description = show.get('show_resume') or show.get('family_resume')
 223
 224         return {
 225             'id': video_id,
 226             'title': title,
 227             'description': description,
 228             'thumbnails': thumbnails,
 229             'timestamp': timestamp,
 230             'uploader': uploader,
 231             'uploader_id': uploader_id,
 232             'duration': duration,
 233             'formats': formats,
 234         }