yt_dlp/extractor/zdf.py

   1 import re
   2
   3 from .common import InfoExtractor
   4 from ..compat import compat_str
   5 from ..utils import (
   6     determine_ext,
   7     float_or_none,
   8     int_or_none,
   9     join_nonempty,
  10     merge_dicts,
  11     NO_DEFAULT,
  12     orderedSet,
  13     parse_codecs,
  14     qualities,
  15     traverse_obj,
  16     try_get,
  17     unified_timestamp,
  18     update_url_query,
  19     url_or_none,
  20     urljoin,
  21 )
  22
  23
  24 class ZDFBaseIE(InfoExtractor):
  25     _GEO_COUNTRIES = ['DE']
  26     _QUALITIES = ('auto', 'low', 'med', 'high', 'veryhigh', 'hd')
  27
  28     def _call_api(self, url, video_id, item, api_token=None, referrer=None):
  29         headers = {}
  30         if api_token:
  31             headers['Api-Auth'] = 'Bearer %s' % api_token
  32         if referrer:
  33             headers['Referer'] = referrer
  34         return self._download_json(
  35             url, video_id, 'Downloading JSON %s' % item, headers=headers)
  36
  37     @staticmethod
  38     def _extract_subtitles(src):
  39         subtitles = {}
  40         for caption in try_get(src, lambda x: x['captions'], list) or []:
  41             subtitle_url = url_or_none(caption.get('uri'))
  42             if subtitle_url:
  43                 lang = caption.get('language', 'deu')
  44                 subtitles.setdefault(lang, []).append({
  45                     'url': subtitle_url,
  46                 })
  47         return subtitles
  48
  49     def _extract_format(self, video_id, formats, format_urls, meta):
  50         format_url = url_or_none(meta.get('url'))
  51         if not format_url or format_url in format_urls:
  52             return
  53         format_urls.add(format_url)
  54
  55         mime_type, ext = meta.get('mimeType'), determine_ext(format_url)
  56         if mime_type == 'application/x-mpegURL' or ext == 'm3u8':
  57             new_formats = self._extract_m3u8_formats(
  58                 format_url, video_id, 'mp4', m3u8_id='hls',
  59                 entry_protocol='m3u8_native', fatal=False)
  60         elif mime_type == 'application/f4m+xml' or ext == 'f4m':
  61             new_formats = self._extract_f4m_formats(
  62                 update_url_query(format_url, {'hdcore': '3.7.0'}), video_id, f4m_id='hds', fatal=False)
  63         else:
  64             f = parse_codecs(meta.get('mimeCodec'))
  65             if not f and meta.get('type'):
  66                 data = meta['type'].split('_')
  67                 if try_get(data, lambda x: x[2]) == ext:
  68                     f = {'vcodec': data[0], 'acodec': data[1]}
  69             f.update({
  70                 'url': format_url,
  71                 'format_id': join_nonempty('http', meta.get('type'), meta.get('quality')),
  72                 'tbr': int_or_none(self._search_regex(r'_(\d+)k_', format_url, 'tbr', default=None))
  73             })
  74             new_formats = [f]
  75         formats.extend(merge_dicts(f, {
  76             'format_note': join_nonempty('quality', 'class', from_dict=meta, delim=', '),
  77             'language': meta.get('language'),
  78             'language_preference': 10 if meta.get('class') == 'main' else -10 if meta.get('class') == 'ad' else -1,
  79             'quality': qualities(self._QUALITIES)(meta.get('quality')),
  80         }) for f in new_formats)
  81
  82     def _extract_ptmd(self, ptmd_url, video_id, api_token, referrer):
  83         ptmd = self._call_api(
  84             ptmd_url, video_id, 'metadata', api_token, referrer)
  85
  86         content_id = ptmd.get('basename') or ptmd_url.split('/')[-1]
  87
  88         formats = []
  89         track_uris = set()
  90         for p in ptmd['priorityList']:
  91             formitaeten = p.get('formitaeten')
  92             if not isinstance(formitaeten, list):
  93                 continue
  94             for f in formitaeten:
  95                 f_qualities = f.get('qualities')
  96                 if not isinstance(f_qualities, list):
  97                     continue
  98                 for quality in f_qualities:
  99                     tracks = try_get(quality, lambda x: x['audio']['tracks'], list)
 100                     if not tracks:
 101                         continue
 102                     for track in tracks:
 103                         self._extract_format(
 104                             content_id, formats, track_uris, {
 105                                 'url': track.get('uri'),
 106                                 'type': f.get('type'),
 107                                 'mimeType': f.get('mimeType'),
 108                                 'quality': quality.get('quality'),
 109                                 'class': track.get('class'),
 110                                 'language': track.get('language'),
 111                             })
 112         self._sort_formats(formats, ('tbr', 'res', 'quality', 'language_preference'))
 113
 114         duration = float_or_none(try_get(
 115             ptmd, lambda x: x['attributes']['duration']['value']), scale=1000)
 116
 117         return {
 118             'extractor_key': ZDFIE.ie_key(),
 119             'id': content_id,
 120             'duration': duration,
 121             'formats': formats,
 122             'subtitles': self._extract_subtitles(ptmd),
 123         }
 124
 125     def _extract_player(self, webpage, video_id, fatal=True):
 126         return self._parse_json(
 127             self._search_regex(
 128                 r'(?s)data-zdfplayer-jsb=(["\'])(?P<json>{.+?})\1', webpage,
 129                 'player JSON', default='{}' if not fatal else NO_DEFAULT,
 130                 group='json'),
 131             video_id)
 132
 133
 134 class ZDFIE(ZDFBaseIE):
 135     _VALID_URL = r'https?://www\.zdf\.de/(?:[^/]+/)*(?P<id>[^/?#&]+)\.html'
 136     _TESTS = [{
 137         # Same as https://www.phoenix.de/sendungen/ereignisse/corona-nachgehakt/wohin-fuehrt-der-protest-in-der-pandemie-a-2050630.html
 138         'url': 'https://www.zdf.de/politik/phoenix-sendungen/wohin-fuehrt-der-protest-in-der-pandemie-100.html',
 139         'md5': '34ec321e7eb34231fd88616c65c92db0',
 140         'info_dict': {
 141             'id': '210222_phx_nachgehakt_corona_protest',
 142             'ext': 'mp4',
 143             'title': 'Wohin führt der Protest in der Pandemie?',
 144             'description': 'md5:7d643fe7f565e53a24aac036b2122fbd',
 145             'duration': 1691,
 146             'timestamp': 1613948400,
 147             'upload_date': '20210221',
 148         },
 149         'skip': 'No longer available: "Diese Seite wurde leider nicht gefunden"',
 150     }, {
 151         # Same as https://www.3sat.de/film/ab-18/10-wochen-sommer-108.html
 152         'url': 'https://www.zdf.de/dokumentation/ab-18/10-wochen-sommer-102.html',
 153         'md5': '0aff3e7bc72c8813f5e0fae333316a1d',
 154         'info_dict': {
 155             'id': '141007_ab18_10wochensommer_film',
 156             'ext': 'mp4',
 157             'title': 'Ab 18! - 10 Wochen Sommer',
 158             'description': 'md5:8253f41dc99ce2c3ff892dac2d65fe26',
 159             'duration': 2660,
 160             'timestamp': 1608604200,
 161             'upload_date': '20201222',
 162         },
 163         'skip': 'No longer available: "Diese Seite wurde leider nicht gefunden"',
 164     }, {
 165         'url': 'https://www.zdf.de/nachrichten/heute-journal/heute-journal-vom-30-12-2021-100.html',
 166         'info_dict': {
 167             'id': '211230_sendung_hjo',
 168             'ext': 'mp4',
 169             'description': 'md5:47dff85977bde9fb8cba9e9c9b929839',
 170             'duration': 1890.0,
 171             'upload_date': '20211230',
 172             'chapters': list,
 173             'thumbnail': 'md5:e65f459f741be5455c952cd820eb188e',
 174             'title': 'heute journal vom 30.12.2021',
 175             'timestamp': 1640897100,
 176         }
 177     }, {
 178         'url': 'https://www.zdf.de/dokumentation/terra-x/die-magie-der-farben-von-koenigspurpur-und-jeansblau-100.html',
 179         'info_dict': {
 180             'id': '151025_magie_farben2_tex',
 181             'ext': 'mp4',
 182             'title': 'Die Magie der Farben (2/2)',
 183             'description': 'md5:a89da10c928c6235401066b60a6d5c1a',
 184             'duration': 2615,
 185             'timestamp': 1465021200,
 186             'upload_date': '20160604',
 187             'thumbnail': 'https://www.zdf.de/assets/mauve-im-labor-100~768x432?cb=1464909117806',
 188         },
 189     }, {
 190         'url': 'https://www.zdf.de/funk/druck-11790/funk-alles-ist-verzaubert-102.html',
 191         'md5': '57af4423db0455a3975d2dc4578536bc',
 192         'info_dict': {
 193             'ext': 'mp4',
 194             'id': 'video_funk_1770473',
 195             'duration': 1278,
 196             'description': 'Die Neue an der Schule verdreht Ismail den Kopf.',
 197             'title': 'Alles ist verzaubert',
 198             'timestamp': 1635520560,
 199             'upload_date': '20211029',
 200             'thumbnail': 'https://www.zdf.de/assets/teaser-funk-alles-ist-verzaubert-100~1920x1080?cb=1636466431799',
 201         },
 202     }, {
 203         # Same as https://www.phoenix.de/sendungen/dokumentationen/gesten-der-maechtigen-i-a-89468.html?ref=suche
 204         'url': 'https://www.zdf.de/politik/phoenix-sendungen/die-gesten-der-maechtigen-100.html',
 205         'only_matching': True,
 206     }, {
 207         # Same as https://www.3sat.de/film/spielfilm/der-hauptmann-100.html
 208         'url': 'https://www.zdf.de/filme/filme-sonstige/der-hauptmann-112.html',
 209         'only_matching': True,
 210     }, {
 211         # Same as https://www.3sat.de/wissen/nano/nano-21-mai-2019-102.html, equal media ids
 212         'url': 'https://www.zdf.de/wissen/nano/nano-21-mai-2019-102.html',
 213         'only_matching': True,
 214     }, {
 215         'url': 'https://www.zdf.de/service-und-hilfe/die-neue-zdf-mediathek/zdfmediathek-trailer-100.html',
 216         'only_matching': True,
 217     }, {
 218         'url': 'https://www.zdf.de/filme/taunuskrimi/die-lebenden-und-die-toten-1---ein-taunuskrimi-100.html',
 219         'only_matching': True,
 220     }, {
 221         'url': 'https://www.zdf.de/dokumentation/planet-e/planet-e-uebersichtsseite-weitere-dokumentationen-von-planet-e-100.html',
 222         'only_matching': True,
 223     }, {
 224         'url': 'https://www.zdf.de/arte/todliche-flucht/page-video-artede-toedliche-flucht-16-100.html',
 225         'info_dict': {
 226             'id': 'video_artede_083871-001-A',
 227             'ext': 'mp4',
 228             'title': 'Tödliche Flucht (1/6)',
 229             'description': 'md5:e34f96a9a5f8abd839ccfcebad3d5315',
 230             'duration': 3193.0,
 231             'timestamp': 1641355200,
 232             'upload_date': '20220105',
 233         },
 234         'skip': 'No longer available "Diese Seite wurde leider nicht gefunden"'
 235     }, {
 236         'url': 'https://www.zdf.de/serien/soko-stuttgart/das-geld-anderer-leute-100.html',
 237         'info_dict': {
 238             'id': '191205_1800_sendung_sok8',
 239             'ext': 'mp4',
 240             'title': 'Das Geld anderer Leute',
 241             'description': 'md5:cb6f660850dc5eb7d1ab776ea094959d',
 242             'duration': 2581.0,
 243             'timestamp': 1654790700,
 244             'upload_date': '20220609',
 245             'thumbnail': 'https://epg-image.zdf.de/fotobase-webdelivery/images/e2d7e55a-09f0-424e-ac73-6cac4dd65f35?layout=2400x1350',
 246         },
 247     }]
 248
 249     def _extract_entry(self, url, player, content, video_id):
 250         title = content.get('title') or content['teaserHeadline']
 251
 252         t = content['mainVideoContent']['http://zdf.de/rels/target']
 253
 254         ptmd_path = t.get('http://zdf.de/rels/streams/ptmd')
 255
 256         if not ptmd_path:
 257             ptmd_path = traverse_obj(
 258                 t, ('streams', 'default', 'http://zdf.de/rels/streams/ptmd-template'),
 259                 'http://zdf.de/rels/streams/ptmd-template').replace(
 260                 '{playerId}', 'ngplayer_2_4')
 261
 262         info = self._extract_ptmd(
 263             urljoin(url, ptmd_path), video_id, player['apiToken'], url)
 264
 265         thumbnails = []
 266         layouts = try_get(
 267             content, lambda x: x['teaserImageRef']['layouts'], dict)
 268         if layouts:
 269             for layout_key, layout_url in layouts.items():
 270                 layout_url = url_or_none(layout_url)
 271                 if not layout_url:
 272                     continue
 273                 thumbnail = {
 274                     'url': layout_url,
 275                     'format_id': layout_key,
 276                 }
 277                 mobj = re.search(r'(?P<width>\d+)x(?P<height>\d+)', layout_key)
 278                 if mobj:
 279                     thumbnail.update({
 280                         'width': int(mobj.group('width')),
 281                         'height': int(mobj.group('height')),
 282                     })
 283                 thumbnails.append(thumbnail)
 284
 285         chapter_marks = t.get('streamAnchorTag') or []
 286         chapter_marks.append({'anchorOffset': int_or_none(t.get('duration'))})
 287         chapters = [{
 288             'start_time': chap.get('anchorOffset'),
 289             'end_time': next_chap.get('anchorOffset'),
 290             'title': chap.get('anchorLabel')
 291         } for chap, next_chap in zip(chapter_marks, chapter_marks[1:])]
 292
 293         return merge_dicts(info, {
 294             'title': title,
 295             'description': content.get('leadParagraph') or content.get('teasertext'),
 296             'duration': int_or_none(t.get('duration')),
 297             'timestamp': unified_timestamp(content.get('editorialDate')),
 298             'thumbnails': thumbnails,
 299             'chapters': chapters or None
 300         })
 301
 302     def _extract_regular(self, url, player, video_id):
 303         content = self._call_api(
 304             player['content'], video_id, 'content', player['apiToken'], url)
 305         return self._extract_entry(player['content'], player, content, video_id)
 306
 307     def _extract_mobile(self, video_id):
 308         video = self._download_json(
 309             'https://zdf-cdn.live.cellular.de/mediathekV2/document/%s' % video_id,
 310             video_id)
 311
 312         document = video['document']
 313
 314         title = document['titel']
 315         content_id = document['basename']
 316
 317         formats = []
 318         format_urls = set()
 319         for f in document['formitaeten']:
 320             self._extract_format(content_id, formats, format_urls, f)
 321         self._sort_formats(formats)
 322
 323         thumbnails = []
 324         teaser_bild = document.get('teaserBild')
 325         if isinstance(teaser_bild, dict):
 326             for thumbnail_key, thumbnail in teaser_bild.items():
 327                 thumbnail_url = try_get(
 328                     thumbnail, lambda x: x['url'], compat_str)
 329                 if thumbnail_url:
 330                     thumbnails.append({
 331                         'url': thumbnail_url,
 332                         'id': thumbnail_key,
 333                         'width': int_or_none(thumbnail.get('width')),
 334                         'height': int_or_none(thumbnail.get('height')),
 335                     })
 336
 337         return {
 338             'id': content_id,
 339             'title': title,
 340             'description': document.get('beschreibung'),
 341             'duration': int_or_none(document.get('length')),
 342             'timestamp': unified_timestamp(document.get('date')) or unified_timestamp(
 343                 try_get(video, lambda x: x['meta']['editorialDate'], compat_str)),
 344             'thumbnails': thumbnails,
 345             'subtitles': self._extract_subtitles(document),
 346             'formats': formats,
 347         }
 348
 349     def _real_extract(self, url):
 350         video_id = self._match_id(url)
 351
 352         webpage = self._download_webpage(url, video_id, fatal=False)
 353         if webpage:
 354             player = self._extract_player(webpage, url, fatal=False)
 355             if player:
 356                 return self._extract_regular(url, player, video_id)
 357
 358         return self._extract_mobile(video_id)
 359
 360
 361 class ZDFChannelIE(ZDFBaseIE):
 362     _VALID_URL = r'https?://www\.zdf\.de/(?:[^/]+/)*(?P<id>[^/?#&]+)'
 363     _TESTS = [{
 364         'url': 'https://www.zdf.de/sport/das-aktuelle-sportstudio',
 365         'info_dict': {
 366             'id': 'das-aktuelle-sportstudio',
 367             'title': 'das aktuelle sportstudio | ZDF',
 368         },
 369         'playlist_mincount': 23,
 370     }, {
 371         'url': 'https://www.zdf.de/dokumentation/planet-e',
 372         'info_dict': {
 373             'id': 'planet-e',
 374             'title': 'planet e.',
 375         },
 376         'playlist_mincount': 50,
 377     }, {
 378         'url': 'https://www.zdf.de/filme/taunuskrimi/',
 379         'only_matching': True,
 380     }]
 381
 382     @classmethod
 383     def suitable(cls, url):
 384         return False if ZDFIE.suitable(url) else super(ZDFChannelIE, cls).suitable(url)
 385
 386     def _real_extract(self, url):
 387         channel_id = self._match_id(url)
 388
 389         webpage = self._download_webpage(url, channel_id)
 390
 391         entries = [
 392             self.url_result(item_url, ie=ZDFIE.ie_key())
 393             for item_url in orderedSet(re.findall(
 394                 r'data-plusbar-url=["\'](http.+?\.html)', webpage))]
 395
 396         return self.playlist_result(
 397             entries, channel_id, self._og_search_title(webpage, fatal=False))
 398
 399         r"""
 400         player = self._extract_player(webpage, channel_id)
 401
 402         channel_id = self._search_regex(
 403             r'docId\s*:\s*(["\'])(?P<id>(?!\1).+?)\1', webpage,
 404             'channel id', group='id')
 405
 406         channel = self._call_api(
 407             'https://api.zdf.de/content/documents/%s.json' % channel_id,
 408             player, url, channel_id)
 409
 410         items = []
 411         for module in channel['module']:
 412             for teaser in try_get(module, lambda x: x['teaser'], list) or []:
 413                 t = try_get(
 414                     teaser, lambda x: x['http://zdf.de/rels/target'], dict)
 415                 if not t:
 416                     continue
 417                 items.extend(try_get(
 418                     t,
 419                     lambda x: x['resultsWithVideo']['http://zdf.de/rels/search/results'],
 420                     list) or [])
 421             items.extend(try_get(
 422                 module,
 423                 lambda x: x['filterRef']['resultsWithVideo']['http://zdf.de/rels/search/results'],
 424                 list) or [])
 425
 426         entries = []
 427         entry_urls = set()
 428         for item in items:
 429             t = try_get(item, lambda x: x['http://zdf.de/rels/target'], dict)
 430             if not t:
 431                 continue
 432             sharing_url = t.get('http://zdf.de/rels/sharing-url')
 433             if not sharing_url or not isinstance(sharing_url, compat_str):
 434                 continue
 435             if sharing_url in entry_urls:
 436                 continue
 437             entry_urls.add(sharing_url)
 438             entries.append(self.url_result(
 439                 sharing_url, ie=ZDFIE.ie_key(), video_id=t.get('id')))
 440
 441         return self.playlist_result(entries, channel_id, channel.get('title'))
 442         """