yt_dlp/extractor/rozhlas.py

   1 import itertools
   2
   3 from .common import InfoExtractor
   4 from ..networking.exceptions import HTTPError
   5 from ..utils import (
   6     ExtractorError,
   7     extract_attributes,
   8     int_or_none,
   9     remove_start,
  10     str_or_none,
  11     traverse_obj,
  12     unified_timestamp,
  13     url_or_none,
  14 )
  15
  16
  17 class RozhlasIE(InfoExtractor):
  18     _VALID_URL = r'https?://(?:www\.)?prehravac\.rozhlas\.cz/audio/(?P<id>[0-9]+)'
  19     _TESTS = [{
  20         'url': 'http://prehravac.rozhlas.cz/audio/3421320',
  21         'md5': '504c902dbc9e9a1fd50326eccf02a7e2',
  22         'info_dict': {
  23             'id': '3421320',
  24             'ext': 'mp3',
  25             'title': 'Echo Pavla Klusáka (30.06.2015 21:00)',
  26             'description': 'Osmdesátiny Terryho Rileyho jsou skvělou příležitostí proletět se elektronickými i akustickými díly zakladatatele minimalismu, který je aktivní už přes padesát let',
  27         },
  28     }, {
  29         'url': 'http://prehravac.rozhlas.cz/audio/3421320/embed',
  30         'only_matching': True,
  31     }]
  32
  33     def _real_extract(self, url):
  34         audio_id = self._match_id(url)
  35
  36         webpage = self._download_webpage(
  37             f'http://prehravac.rozhlas.cz/audio/{audio_id}', audio_id)
  38
  39         title = self._html_search_regex(
  40             r'<h3>(.+?)</h3>\s*<p[^>]*>.*?</p>\s*<div[^>]+id=["\']player-track',
  41             webpage, 'title', default=None) or remove_start(
  42             self._og_search_title(webpage), 'Radio Wave - ')
  43         description = self._html_search_regex(
  44             r'<p[^>]+title=(["\'])(?P<url>(?:(?!\1).)+)\1[^>]*>.*?</p>\s*<div[^>]+id=["\']player-track',
  45             webpage, 'description', fatal=False, group='url')
  46         duration = int_or_none(self._search_regex(
  47             r'data-duration=["\'](\d+)', webpage, 'duration', default=None))
  48
  49         return {
  50             'id': audio_id,
  51             'url': f'http://media.rozhlas.cz/_audio/{audio_id}.mp3',
  52             'title': title,
  53             'description': description,
  54             'duration': duration,
  55             'vcodec': 'none',
  56         }
  57
  58
  59 class RozhlasBaseIE(InfoExtractor):
  60     def _extract_formats(self, entry, audio_id):
  61         formats = []
  62         for audio in traverse_obj(entry, ('audioLinks', lambda _, v: url_or_none(v['url']))):
  63             ext = audio.get('variant')
  64             for retry in self.RetryManager():
  65                 if retry.attempt > 1:
  66                     self._sleep(1, audio_id)
  67                 try:
  68                     if ext == 'dash':
  69                         formats.extend(self._extract_mpd_formats(
  70                             audio['url'], audio_id, mpd_id=ext))
  71                     elif ext == 'hls':
  72                         formats.extend(self._extract_m3u8_formats(
  73                             audio['url'], audio_id, 'm4a', m3u8_id=ext))
  74                     else:
  75                         formats.append({
  76                             'url': audio['url'],
  77                             'ext': ext,
  78                             'format_id': ext,
  79                             'abr': int_or_none(audio.get('bitrate')),
  80                             'acodec': ext,
  81                             'vcodec': 'none',
  82                         })
  83                 except ExtractorError as e:
  84                     if isinstance(e.cause, HTTPError) and e.cause.status == 429:
  85                         retry.error = e.cause
  86                     else:
  87                         self.report_warning(e.msg)
  88
  89         return formats
  90
  91
  92 class RozhlasVltavaIE(RozhlasBaseIE):
  93     _VALID_URL = r'https?://(?:\w+\.rozhlas|english\.radio)\.cz/[\w-]+-(?P<id>\d+)'
  94     _TESTS = [{
  95         'url': 'https://wave.rozhlas.cz/papej-masicko-porcujeme-a-bilancujeme-filmy-a-serialy-ktere-letos-zabily-8891337',
  96         'md5': 'ba2fdbc1242fc16771c7695d271ec355',
  97         'info_dict': {
  98             'id': '8891337',
  99             'title': 'md5:21f99739d04ab49d8c189ec711eef4ec',
 100         },
 101         'playlist_count': 1,
 102         'playlist': [{
 103             'md5': 'ba2fdbc1242fc16771c7695d271ec355',
 104             'info_dict': {
 105                 'id': '10520988',
 106                 'ext': 'mp3',
 107                 'title': 'Papej masíčko! Porcujeme a bilancujeme filmy a seriály, které to letos zabily',
 108                 'description': 'md5:1c6d29fb9564e1f17fc1bb83ae7da0bc',
 109                 'duration': 1574,
 110                 'artist': 'Aleš Stuchlý',
 111                 'channel_id': 'radio-wave',
 112             },
 113         }],
 114     }, {
 115         'url': 'https://wave.rozhlas.cz/poslechnete-si-neklid-podcastovy-thriller-o-vine-strachu-a-vztahu-ktery-zasel-8554744',
 116         'info_dict': {
 117             'id': '8554744',
 118             'title': 'Poslechněte si Neklid. Podcastový thriller o vině, strachu a vztahu, který zašel příliš daleko',
 119         },
 120         'playlist_count': 5,
 121         'playlist': [{
 122             'md5': '93d4109cf8f40523699ae9c1d4600bdd',
 123             'info_dict': {
 124                 'id': '9890713',
 125                 'ext': 'mp3',
 126                 'title': 'Neklid #1',
 127                 'description': '1. díl: Neklid: 1. díl',
 128                 'duration': 1025,
 129                 'artist': 'Josef Kokta',
 130                 'channel_id': 'radio-wave',
 131                 'chapter': 'Neklid #1',
 132                 'chapter_number': 1,
 133             },
 134         }, {
 135             'md5': 'e9763235be4a6dcf94bc8a5bac1ca126',
 136             'info_dict': {
 137                 'id': '9890716',
 138                 'ext': 'mp3',
 139                 'title': 'Neklid #2',
 140                 'description': '2. díl: Neklid: 2. díl',
 141                 'duration': 768,
 142                 'artist': 'Josef Kokta',
 143                 'channel_id': 'radio-wave',
 144                 'chapter': 'Neklid #2',
 145                 'chapter_number': 2,
 146             },
 147         }, {
 148             'md5': '00b642ea94b78cc949ac84da09f87895',
 149             'info_dict': {
 150                 'id': '9890722',
 151                 'ext': 'mp3',
 152                 'title': 'Neklid #3',
 153                 'description': '3. díl: Neklid: 3. díl',
 154                 'duration': 607,
 155                 'artist': 'Josef Kokta',
 156                 'channel_id': 'radio-wave',
 157                 'chapter': 'Neklid #3',
 158                 'chapter_number': 3,
 159             },
 160         }, {
 161             'md5': 'faef97b1b49da7df874740f118c19dea',
 162             'info_dict': {
 163                 'id': '9890728',
 164                 'ext': 'mp3',
 165                 'title': 'Neklid #4',
 166                 'description': '4. díl: Neklid: 4. díl',
 167                 'duration': 621,
 168                 'artist': 'Josef Kokta',
 169                 'channel_id': 'radio-wave',
 170                 'chapter': 'Neklid #4',
 171                 'chapter_number': 4,
 172             },
 173         }, {
 174             'md5': '6e729fa39b647325b868d419c76f3efa',
 175             'info_dict': {
 176                 'id': '9890734',
 177                 'ext': 'mp3',
 178                 'title': 'Neklid #5',
 179                 'description': '5. díl: Neklid: 5. díl',
 180                 'duration': 908,
 181                 'artist': 'Josef Kokta',
 182                 'channel_id': 'radio-wave',
 183                 'chapter': 'Neklid #5',
 184                 'chapter_number': 5,
 185             },
 186         }],
 187     }, {
 188         'url': 'https://dvojka.rozhlas.cz/karel-siktanc-cerny-jezdec-bily-kun-napinava-pohadka-o-tajemnem-prizraku-8946969',
 189         'info_dict': {
 190             'id': '8946969',
 191             'title': 'Karel Šiktanc: Černý jezdec, bílý kůň. Napínavá pohádka o tajemném přízraku',
 192         },
 193         'playlist_count': 1,
 194         'playlist': [{
 195             'info_dict': {
 196                 'id': '10631121',
 197                 'ext': 'm4a',
 198                 'title': 'Karel Šiktanc: Černý jezdec, bílý kůň. Napínavá pohádka o tajemném přízraku',
 199                 'description': 'Karel Šiktanc: Černý jezdec, bílý kůň',
 200                 'duration': 2656,
 201                 'artist': 'Tvůrčí skupina Drama a literatura',
 202                 'channel_id': 'dvojka',
 203             },
 204         }],
 205         'params': {'skip_download': 'dash'},
 206     }]
 207
 208     def _extract_video(self, entry):
 209         audio_id = entry['meta']['ga']['contentId']
 210         chapter_number = traverse_obj(entry, ('meta', 'ga', 'contentSerialPart', {int_or_none}))
 211
 212         return {
 213             'id': audio_id,
 214             'chapter': traverse_obj(entry, ('meta', 'ga', 'contentNameShort')) if chapter_number else None,
 215             'chapter_number': chapter_number,
 216             'formats': self._extract_formats(entry, audio_id),
 217             **traverse_obj(entry, {
 218                 'title': ('meta', 'ga', 'contentName'),
 219                 'description': 'title',
 220                 'duration': ('duration', {int_or_none}),
 221                 'artist': ('meta', 'ga', 'contentAuthor'),
 222                 'channel_id': ('meta', 'ga', 'contentCreator'),
 223             }),
 224         }
 225
 226     def _real_extract(self, url):
 227         video_id = self._match_id(url)
 228         webpage = self._download_webpage(url, video_id)
 229
 230         # FIXME: Use get_element_text_and_html_by_tag when it accepts less strict html
 231         data = self._parse_json(extract_attributes(self._search_regex(
 232             r'(<div class="mujRozhlasPlayer" data-player=\'[^\']+\'>)',
 233             webpage, 'player'))['data-player'], video_id)['data']
 234
 235         return {
 236             '_type': 'playlist',
 237             'id': str_or_none(data.get('embedId')) or video_id,
 238             'title': traverse_obj(data, ('series', 'title')),
 239             'entries': map(self._extract_video, data['playlist']),
 240         }
 241
 242
 243 class MujRozhlasIE(RozhlasBaseIE):
 244     _VALID_URL = r'https?://(?:www\.)?mujrozhlas\.cz/(?:[^/]+/)*(?P<id>[^/?#&]+)'
 245     _TESTS = [{
 246         # single episode extraction
 247         'url': 'https://www.mujrozhlas.cz/vykopavky/ach-jo-zase-teleci-rizek-je-mnohem-min-cesky-nez-jsme-si-mysleli',
 248         'md5': '6f8fd68663e64936623e67c152a669e0',
 249         'info_dict': {
 250             'id': '10787730',
 251             'ext': 'mp3',
 252             'title': 'Ach jo, zase to telecí! Řízek je mnohem míň český, než jsme si mysleli',
 253             'description': 'md5:db7141e9caaedc9041ec7cefb9a62908',
 254             'timestamp': 1684915200,
 255             'modified_timestamp': 1687550432,
 256             'series': 'Vykopávky',
 257             'thumbnail': 'https://portal.rozhlas.cz/sites/default/files/images/84377046610af6ddc54d910b1dd7a22b.jpg',
 258             'channel_id': 'radio-wave',
 259             'upload_date': '20230524',
 260             'modified_date': '20230623',
 261         },
 262     }, {
 263         # serial extraction
 264         'url': 'https://www.mujrozhlas.cz/radiokniha/jaroslava-janackova-pribeh-tajemneho-psani-o-pramenech-genezi-babicky',
 265         'playlist_mincount': 7,
 266         'info_dict': {
 267             'id': 'bb2b5f4e-ffb4-35a6-a34a-046aa62d6f6b',
 268             'title': 'Jaroslava Janáčková: Příběh tajemného psaní. O pramenech a genezi Babičky',
 269             'description': 'md5:7434d8fac39ac9fee6df098e11dfb1be',
 270         },
 271     }, {
 272         # show extraction
 273         'url': 'https://www.mujrozhlas.cz/nespavci',
 274         'playlist_mincount': 14,
 275         'info_dict': {
 276             'id': '09db9b37-d0f4-368c-986a-d3439f741f08',
 277             'title': 'Nespavci',
 278             'description': 'md5:c430adcbf9e2b9eac88b745881e814dc',
 279         },
 280     }, {
 281         # serialPart
 282         'url': 'https://www.mujrozhlas.cz/povidka/gustavo-adolfo-becquer-hora-duchu',
 283         'info_dict': {
 284             'id': '8889035',
 285             'ext': 'm4a',
 286             'title': 'Gustavo Adolfo Bécquer: Hora duchů',
 287             'description': 'md5:343a15257b376c276e210b78e900ffea',
 288             'chapter': 'Hora duchů a Polibek – dva tajemné příběhy Gustava Adolfa Bécquera',
 289             'thumbnail': 'https://portal.rozhlas.cz/sites/default/files/images/2adfe1387fb140634be725c1ccf26214.jpg',
 290             'timestamp': 1708173000,
 291             'episode': 'Episode 1',
 292             'episode_number': 1,
 293             'series': 'Povídka',
 294             'modified_date': '20240217',
 295             'upload_date': '20240217',
 296             'modified_timestamp': 1708173198,
 297             'channel_id': 'vltava',
 298         },
 299         'params': {'skip_download': 'dash'},
 300     }]
 301
 302     def _call_api(self, path, item_id, msg='API JSON'):
 303         return self._download_json(
 304             f'https://api.mujrozhlas.cz/{path}/{item_id}', item_id,
 305             note=f'Downloading {msg}', errnote=f'Failed to download {msg}')['data']
 306
 307     def _extract_audio_entry(self, entry):
 308         audio_id = entry['meta']['ga']['contentId']
 309
 310         return {
 311             'id': audio_id,
 312             'formats': self._extract_formats(entry['attributes'], audio_id),
 313             **traverse_obj(entry, {
 314                 'title': ('attributes', 'title'),
 315                 'description': ('attributes', 'description'),
 316                 'episode_number': ('attributes', 'part'),
 317                 'series': ('attributes', 'mirroredShow', 'title'),
 318                 'chapter': ('attributes', 'mirroredSerial', 'title'),
 319                 'artist': ('meta', 'ga', 'contentAuthor'),
 320                 'channel_id': ('meta', 'ga', 'contentCreator'),
 321                 'timestamp': ('attributes', 'since', {unified_timestamp}),
 322                 'modified_timestamp': ('attributes', 'updated', {unified_timestamp}),
 323                 'thumbnail': ('attributes', 'asset', 'url', {url_or_none}),
 324             }),
 325         }
 326
 327     def _entries(self, api_url, playlist_id):
 328         for page in itertools.count(1):
 329             episodes = self._download_json(
 330                 api_url, playlist_id, note=f'Downloading episodes page {page}',
 331                 errnote=f'Failed to download episodes page {page}', fatal=False)
 332             for episode in traverse_obj(episodes, ('data', lambda _, v: v['meta']['ga']['contentId'])):
 333                 yield self._extract_audio_entry(episode)
 334             api_url = traverse_obj(episodes, ('links', 'next', {url_or_none}))
 335             if not api_url:
 336                 break
 337
 338     def _real_extract(self, url):
 339         display_id = self._match_id(url)
 340         webpage = self._download_webpage(url, display_id)
 341         info = self._search_json(r'\bvar\s+dl\s*=', webpage, 'info json', display_id)
 342
 343         entity = info['siteEntityBundle']
 344
 345         if entity in ('episode', 'serialPart'):
 346             return self._extract_audio_entry(self._call_api(
 347                 'episodes', info['contentId'], 'episode info API JSON'))
 348
 349         elif entity in ('show', 'serial'):
 350             playlist_id = info['contentShow'].split(':')[0] if entity == 'show' else info['contentId']
 351             data = self._call_api(f'{entity}s', playlist_id, f'{entity} playlist JSON')
 352             api_url = data['relationships']['episodes']['links']['related']
 353             return self.playlist_result(
 354                 self._entries(api_url, playlist_id), playlist_id,
 355                 **traverse_obj(data, ('attributes', {
 356                     'title': 'title',
 357                     'description': 'description',
 358                 })))
 359
 360         else:
 361             # `entity == 'person'` not implemented yet by API, ref:
 362             # https://api.mujrozhlas.cz/persons/8367e456-2a57-379a-91bb-e699619bea49/participation
 363             raise ExtractorError(f'Unsupported entity type "{entity}"')