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

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     clean_html,
   9     determine_ext,
  10     int_or_none,
  11     js_to_json,
  12     qualities,
  13     unified_strdate,
  14     url_or_none,
  15 )
  16
  17
  18 class NovaEmbedIE(InfoExtractor):
  19     _VALID_URL = r'https?://media\.cms\.nova\.cz/embed/(?P<id>[^/?#&]+)'
  20     _TEST = {
  21         'url': 'https://media.cms.nova.cz/embed/8o0n0r?autoplay=1',
  22         'md5': 'ee009bafcc794541570edd44b71cbea3',
  23         'info_dict': {
  24             'id': '8o0n0r',
  25             'ext': 'mp4',
  26             'title': '2180. díl',
  27             'thumbnail': r're:^https?://.*\.jpg',
  28             'duration': 2578,
  29         },
  30     }
  31
  32     def _real_extract(self, url):
  33         video_id = self._match_id(url)
  34
  35         webpage = self._download_webpage(url, video_id)
  36
  37         duration = None
  38         formats = []
  39
  40         player = self._parse_json(
  41             self._search_regex(
  42                 r'Player\.init\s*\([^,]+,\s*({.+?})\s*,\s*{.+?}\s*\)\s*;',
  43                 webpage, 'player', default='{}'), video_id, fatal=False)
  44         if player:
  45             for format_id, format_list in player['tracks'].items():
  46                 if not isinstance(format_list, list):
  47                     format_list = [format_list]
  48                 for format_dict in format_list:
  49                     if not isinstance(format_dict, dict):
  50                         continue
  51                     format_url = url_or_none(format_dict.get('src'))
  52                     format_type = format_dict.get('type')
  53                     ext = determine_ext(format_url)
  54                     if (format_type == 'application/x-mpegURL'
  55                             or format_id == 'HLS' or ext == 'm3u8'):
  56                         formats.extend(self._extract_m3u8_formats(
  57                             format_url, video_id, 'mp4',
  58                             entry_protocol='m3u8_native', m3u8_id='hls',
  59                             fatal=False))
  60                     elif (format_type == 'application/dash+xml'
  61                           or format_id == 'DASH' or ext == 'mpd'):
  62                         formats.extend(self._extract_mpd_formats(
  63                             format_url, video_id, mpd_id='dash', fatal=False))
  64                     else:
  65                         formats.append({
  66                             'url': format_url,
  67                         })
  68             duration = int_or_none(player.get('duration'))
  69         else:
  70             # Old path, not actual as of 08.04.2020
  71             bitrates = self._parse_json(
  72                 self._search_regex(
  73                     r'(?s)(?:src|bitrates)\s*=\s*({.+?})\s*;', webpage, 'formats'),
  74                 video_id, transform_source=js_to_json)
  75
  76             QUALITIES = ('lq', 'mq', 'hq', 'hd')
  77             quality_key = qualities(QUALITIES)
  78
  79             for format_id, format_list in bitrates.items():
  80                 if not isinstance(format_list, list):
  81                     format_list = [format_list]
  82                 for format_url in format_list:
  83                     format_url = url_or_none(format_url)
  84                     if not format_url:
  85                         continue
  86                     if format_id == 'hls':
  87                         formats.extend(self._extract_m3u8_formats(
  88                             format_url, video_id, ext='mp4',
  89                             entry_protocol='m3u8_native', m3u8_id='hls',
  90                             fatal=False))
  91                         continue
  92                     f = {
  93                         'url': format_url,
  94                     }
  95                     f_id = format_id
  96                     for quality in QUALITIES:
  97                         if '%s.mp4' % quality in format_url:
  98                             f_id += '-%s' % quality
  99                             f.update({
 100                                 'quality': quality_key(quality),
 101                                 'format_note': quality.upper(),
 102                             })
 103                             break
 104                     f['format_id'] = f_id
 105                     formats.append(f)
 106
 107         self._sort_formats(formats)
 108
 109         title = self._og_search_title(
 110             webpage, default=None) or self._search_regex(
 111             (r'<value>(?P<title>[^<]+)',
 112              r'videoTitle\s*:\s*(["\'])(?P<value>(?:(?!\1).)+)\1'), webpage,
 113             'title', group='value')
 114         thumbnail = self._og_search_thumbnail(
 115             webpage, default=None) or self._search_regex(
 116             r'poster\s*:\s*(["\'])(?P<value>(?:(?!\1).)+)\1', webpage,
 117             'thumbnail', fatal=False, group='value')
 118         duration = int_or_none(self._search_regex(
 119             r'videoDuration\s*:\s*(\d+)', webpage, 'duration',
 120             default=duration))
 121
 122         return {
 123             'id': video_id,
 124             'title': title,
 125             'thumbnail': thumbnail,
 126             'duration': duration,
 127             'formats': formats,
 128         }
 129
 130
 131 class NovaIE(InfoExtractor):
 132     IE_DESC = 'TN.cz, Prásk.tv, Nova.cz, Novaplus.cz, FANDA.tv, Krásná.cz and Doma.cz'
 133     _VALID_URL = r'https?://(?:[^.]+\.)?(?P<site>tv(?:noviny)?|tn|novaplus|vymena|fanda|krasna|doma|prask)\.nova\.cz/(?:[^/]+/)+(?P<id>[^/]+?)(?:\.html|/|$)'
 134     _TESTS = [{
 135         'url': 'http://tn.nova.cz/clanek/tajemstvi-ukryte-v-podzemi-specialni-nemocnice-v-prazske-krci.html#player_13260',
 136         'md5': '249baab7d0104e186e78b0899c7d5f28',
 137         'info_dict': {
 138             'id': '1757139',
 139             'display_id': 'tajemstvi-ukryte-v-podzemi-specialni-nemocnice-v-prazske-krci',
 140             'ext': 'mp4',
 141             'title': 'Podzemní nemocnice v pražské Krči',
 142             'description': 'md5:f0a42dd239c26f61c28f19e62d20ef53',
 143             'thumbnail': r're:^https?://.*\.(?:jpg)',
 144         }
 145     }, {
 146         'url': 'http://fanda.nova.cz/clanek/fun-and-games/krvavy-epos-zaklinac-3-divoky-hon-vychazi-vyhrajte-ho-pro-sebe.html',
 147         'info_dict': {
 148             'id': '1753621',
 149             'ext': 'mp4',
 150             'title': 'Zaklínač 3: Divoký hon',
 151             'description': 're:.*Pokud se stejně jako my nemůžete.*',
 152             'thumbnail': r're:https?://.*\.jpg(\?.*)?',
 153             'upload_date': '20150521',
 154         },
 155         'params': {
 156             # rtmp download
 157             'skip_download': True,
 158         },
 159         'skip': 'gone',
 160     }, {
 161         # media.cms.nova.cz embed
 162         'url': 'https://novaplus.nova.cz/porad/ulice/epizoda/18760-2180-dil',
 163         'info_dict': {
 164             'id': '8o0n0r',
 165             'ext': 'mp4',
 166             'title': '2180. díl',
 167             'thumbnail': r're:^https?://.*\.jpg',
 168             'duration': 2578,
 169         },
 170         'params': {
 171             'skip_download': True,
 172         },
 173         'add_ie': [NovaEmbedIE.ie_key()],
 174         'skip': 'CHYBA 404: STRÁNKA NENALEZENA',
 175     }, {
 176         'url': 'http://sport.tn.nova.cz/clanek/sport/hokej/nhl/zivot-jde-dal-hodnotil-po-vyrazeni-z-playoff-jiri-sekac.html',
 177         'only_matching': True,
 178     }, {
 179         'url': 'http://fanda.nova.cz/clanek/fun-and-games/krvavy-epos-zaklinac-3-divoky-hon-vychazi-vyhrajte-ho-pro-sebe.html',
 180         'only_matching': True,
 181     }, {
 182         'url': 'http://doma.nova.cz/clanek/zdravi/prijdte-se-zapsat-do-registru-kostni-drene-jiz-ve-stredu-3-cervna.html',
 183         'only_matching': True,
 184     }, {
 185         'url': 'http://prask.nova.cz/clanek/novinky/co-si-na-sobe-nase-hvezdy-nechaly-pojistit.html',
 186         'only_matching': True,
 187     }, {
 188         'url': 'http://tv.nova.cz/clanek/novinky/zivot-je-zivot-bondovsky-trailer.html',
 189         'only_matching': True,
 190     }]
 191
 192     def _real_extract(self, url):
 193         mobj = re.match(self._VALID_URL, url)
 194         display_id = mobj.group('id')
 195         site = mobj.group('site')
 196
 197         webpage = self._download_webpage(url, display_id)
 198
 199         description = clean_html(self._og_search_description(webpage, default=None))
 200         if site == 'novaplus':
 201             upload_date = unified_strdate(self._search_regex(
 202                 r'(\d{1,2}-\d{1,2}-\d{4})$', display_id, 'upload date', default=None))
 203         elif site == 'fanda':
 204             upload_date = unified_strdate(self._search_regex(
 205                 r'<span class="date_time">(\d{1,2}\.\d{1,2}\.\d{4})', webpage, 'upload date', default=None))
 206         else:
 207             upload_date = None
 208
 209         # novaplus
 210         embed_id = self._search_regex(
 211             r'<iframe[^>]+\bsrc=["\'](?:https?:)?//media\.cms\.nova\.cz/embed/([^/?#&]+)',
 212             webpage, 'embed url', default=None)
 213         if embed_id:
 214             return {
 215                 '_type': 'url_transparent',
 216                 'url': 'https://media.cms.nova.cz/embed/%s' % embed_id,
 217                 'ie_key': NovaEmbedIE.ie_key(),
 218                 'id': embed_id,
 219                 'description': description,
 220                 'upload_date': upload_date
 221             }
 222
 223         video_id = self._search_regex(
 224             [r"(?:media|video_id)\s*:\s*'(\d+)'",
 225              r'media=(\d+)',
 226              r'id="article_video_(\d+)"',
 227              r'id="player_(\d+)"'],
 228             webpage, 'video id')
 229
 230         config_url = self._search_regex(
 231             r'src="(https?://(?:tn|api)\.nova\.cz/bin/player/videojs/config\.php\?[^"]+)"',
 232             webpage, 'config url', default=None)
 233         config_params = {}
 234
 235         if not config_url:
 236             player = self._parse_json(
 237                 self._search_regex(
 238                     r'(?s)Player\s*\(.+?\s*,\s*({.+?\bmedia\b["\']?\s*:\s*["\']?\d+.+?})\s*\)', webpage,
 239                     'player', default='{}'),
 240                 video_id, transform_source=js_to_json, fatal=False)
 241             if player:
 242                 config_url = url_or_none(player.get('configUrl'))
 243                 params = player.get('configParams')
 244                 if isinstance(params, dict):
 245                     config_params = params
 246
 247         if not config_url:
 248             DEFAULT_SITE_ID = '23000'
 249             SITES = {
 250                 'tvnoviny': DEFAULT_SITE_ID,
 251                 'novaplus': DEFAULT_SITE_ID,
 252                 'vymena': DEFAULT_SITE_ID,
 253                 'krasna': DEFAULT_SITE_ID,
 254                 'fanda': '30',
 255                 'tn': '30',
 256                 'doma': '30',
 257             }
 258
 259             site_id = self._search_regex(
 260                 r'site=(\d+)', webpage, 'site id', default=None) or SITES.get(
 261                 site, DEFAULT_SITE_ID)
 262
 263             config_url = 'https://api.nova.cz/bin/player/videojs/config.php'
 264             config_params = {
 265                 'site': site_id,
 266                 'media': video_id,
 267                 'quality': 3,
 268                 'version': 1,
 269             }
 270
 271         config = self._download_json(
 272             config_url, display_id,
 273             'Downloading config JSON', query=config_params,
 274             transform_source=lambda s: s[s.index('{'):s.rindex('}') + 1])
 275
 276         mediafile = config['mediafile']
 277         video_url = mediafile['src']
 278
 279         m = re.search(r'^(?P<url>rtmpe?://[^/]+/(?P<app>[^/]+?))/&*(?P<playpath>.+)$', video_url)
 280         if m:
 281             formats = [{
 282                 'url': m.group('url'),
 283                 'app': m.group('app'),
 284                 'play_path': m.group('playpath'),
 285                 'player_path': 'http://tvnoviny.nova.cz/static/shared/app/videojs/video-js.swf',
 286                 'ext': 'flv',
 287             }]
 288         else:
 289             formats = [{
 290                 'url': video_url,
 291             }]
 292         self._sort_formats(formats)
 293
 294         title = mediafile.get('meta', {}).get('title') or self._og_search_title(webpage)
 295         thumbnail = config.get('poster')
 296
 297         return {
 298             'id': video_id,
 299             'display_id': display_id,
 300             'title': title,
 301             'description': description,
 302             'upload_date': upload_date,
 303             'thumbnail': thumbnail,
 304             'formats': formats,
 305         }