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