youtube_dl/extractor/zdf.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..compat import compat_str
   8 from ..utils import (
   9     determine_ext,
  10     ExtractorError,
  11     float_or_none,
  12     int_or_none,
  13     merge_dicts,
  14     NO_DEFAULT,
  15     orderedSet,
  16     parse_codecs,
  17     qualities,
  18     try_get,
  19     unified_timestamp,
  20     update_url_query,
  21     url_or_none,
  22     urljoin,
  23 )
  24
  25
  26 class ZDFBaseIE(InfoExtractor):
  27     _GEO_COUNTRIES = ['DE']
  28     _QUALITIES = ('auto', 'low', 'med', 'high', 'veryhigh', 'hd')
  29
  30     def _call_api(self, url, video_id, item, api_token=None, referrer=None):
  31         headers = {}
  32         if api_token:
  33             headers['Api-Auth'] = 'Bearer %s' % api_token
  34         if referrer:
  35             headers['Referer'] = referrer
  36         return self._download_json(
  37             url, video_id, 'Downloading JSON %s' % item, headers=headers)
  38
  39     @staticmethod
  40     def _extract_subtitles(src):
  41         subtitles = {}
  42         for caption in try_get(src, lambda x: x['captions'], list) or []:
  43             subtitle_url = url_or_none(caption.get('uri'))
  44             if subtitle_url:
  45                 lang = caption.get('language', 'deu')
  46                 subtitles.setdefault(lang, []).append({
  47                     'url': subtitle_url,
  48                 })
  49         return subtitles
  50
  51     def _extract_format(self, video_id, formats, format_urls, meta):
  52         format_url = url_or_none(meta.get('url'))
  53         if not format_url:
  54             return
  55         if format_url in format_urls:
  56             return
  57         format_urls.add(format_url)
  58         mime_type = meta.get('mimeType')
  59         ext = determine_ext(format_url)
  60         if mime_type == 'application/x-mpegURL' or ext == 'm3u8':
  61             formats.extend(self._extract_m3u8_formats(
  62                 format_url, video_id, 'mp4', m3u8_id='hls',
  63                 entry_protocol='m3u8_native', fatal=False))
  64         elif mime_type == 'application/f4m+xml' or ext == 'f4m':
  65             formats.extend(self._extract_f4m_formats(
  66                 update_url_query(format_url, {'hdcore': '3.7.0'}), video_id, f4m_id='hds', fatal=False))
  67         else:
  68             f = parse_codecs(meta.get('mimeCodec'))
  69             format_id = ['http']
  70             for p in (meta.get('type'), meta.get('quality')):
  71                 if p and isinstance(p, compat_str):
  72                     format_id.append(p)
  73             f.update({
  74                 'url': format_url,
  75                 'format_id': '-'.join(format_id),
  76                 'format_note': meta.get('quality'),
  77                 'language': meta.get('language'),
  78                 'quality': qualities(self._QUALITIES)(meta.get('quality')),
  79                 'preference': -10,
  80             })
  81             formats.append(f)
  82
  83     def _extract_ptmd(self, ptmd_url, video_id, api_token, referrer):
  84         ptmd = self._call_api(
  85             ptmd_url, video_id, 'metadata', api_token, referrer)
  86
  87         content_id = ptmd.get('basename') or ptmd_url.split('/')[-1]
  88
  89         formats = []
  90         track_uris = set()
  91         for p in ptmd['priorityList']:
  92             formitaeten = p.get('formitaeten')
  93             if not isinstance(formitaeten, list):
  94                 continue
  95             for f in formitaeten:
  96                 f_qualities = f.get('qualities')
  97                 if not isinstance(f_qualities, list):
  98                     continue
  99                 for quality in f_qualities:
 100                     tracks = try_get(quality, lambda x: x['audio']['tracks'], list)
 101                     if not tracks:
 102                         continue
 103                     for track in tracks:
 104                         self._extract_format(
 105                             content_id, formats, track_uris, {
 106                                 'url': track.get('uri'),
 107                                 'type': f.get('type'),
 108                                 'mimeType': f.get('mimeType'),
 109                                 'quality': quality.get('quality'),
 110                                 'language': track.get('language'),
 111                             })
 112         self._sort_formats(formats)
 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/dokumentation/terra-x/die-magie-der-farben-von-koenigspurpur-und-jeansblau-100.html',
 166         'info_dict': {
 167             'id': '151025_magie_farben2_tex',
 168             'ext': 'mp4',
 169             'title': 'Die Magie der Farben (2/2)',
 170             'description': 'md5:a89da10c928c6235401066b60a6d5c1a',
 171             'duration': 2615,
 172             'timestamp': 1465021200,
 173             'upload_date': '20160604',
 174         },
 175     }, {
 176         # Same as https://www.phoenix.de/sendungen/dokumentationen/gesten-der-maechtigen-i-a-89468.html?ref=suche
 177         'url': 'https://www.zdf.de/politik/phoenix-sendungen/die-gesten-der-maechtigen-100.html',
 178         'only_matching': True,
 179     }, {
 180         # Same as https://www.3sat.de/film/spielfilm/der-hauptmann-100.html
 181         'url': 'https://www.zdf.de/filme/filme-sonstige/der-hauptmann-112.html',
 182         'only_matching': True,
 183     }, {
 184         # Same as https://www.3sat.de/wissen/nano/nano-21-mai-2019-102.html, equal media ids
 185         'url': 'https://www.zdf.de/wissen/nano/nano-21-mai-2019-102.html',
 186         'only_matching': True,
 187     }, {
 188         'url': 'https://www.zdf.de/service-und-hilfe/die-neue-zdf-mediathek/zdfmediathek-trailer-100.html',
 189         'only_matching': True,
 190     }, {
 191         'url': 'https://www.zdf.de/filme/taunuskrimi/die-lebenden-und-die-toten-1---ein-taunuskrimi-100.html',
 192         'only_matching': True,
 193     }, {
 194         'url': 'https://www.zdf.de/dokumentation/planet-e/planet-e-uebersichtsseite-weitere-dokumentationen-von-planet-e-100.html',
 195         'only_matching': True,
 196     }, {
 197         'url': 'https://www.zdf.de/arte/todliche-flucht/page-video-artede-toedliche-flucht-16-100.html',
 198         'info_dict': {
 199             'id': 'video_artede_083871-001-A',
 200             'ext': 'mp4',
 201             'title': 'Tödliche Flucht (1/6)',
 202             'description': 'md5:e34f96a9a5f8abd839ccfcebad3d5315',
 203             'duration': 3193.0,
 204             'timestamp': 1641355200,
 205             'upload_date': '20220105',
 206         },
 207     }]
 208
 209     def _extract_entry(self, url, player, content, video_id):
 210         title = content.get('title') or content['teaserHeadline']
 211
 212         t = content['mainVideoContent']['http://zdf.de/rels/target']
 213
 214         def get_ptmd_path(d):
 215             return (
 216                 d.get('http://zdf.de/rels/streams/ptmd')
 217                 or d.get('http://zdf.de/rels/streams/ptmd-template',
 218                          '').replace('{playerId}', 'ngplayer_2_4'))
 219
 220         ptmd_path = get_ptmd_path(try_get(t, lambda x: x['streams']['default'], dict) or {})
 221         if not ptmd_path:
 222             ptmd_path = get_ptmd_path(t)
 223
 224         if not ptmd_path:
 225             raise ExtractorError('Could not extract ptmd_path')
 226
 227         info = self._extract_ptmd(
 228             urljoin(url, ptmd_path), video_id, player['apiToken'], url)
 229
 230         thumbnails = []
 231         layouts = try_get(
 232             content, lambda x: x['teaserImageRef']['layouts'], dict)
 233         if layouts:
 234             for layout_key, layout_url in layouts.items():
 235                 layout_url = url_or_none(layout_url)
 236                 if not layout_url:
 237                     continue
 238                 thumbnail = {
 239                     'url': layout_url,
 240                     'format_id': layout_key,
 241                 }
 242                 mobj = re.search(r'(?P<width>\d+)x(?P<height>\d+)', layout_key)
 243                 if mobj:
 244                     thumbnail.update({
 245                         'width': int(mobj.group('width')),
 246                         'height': int(mobj.group('height')),
 247                     })
 248                 thumbnails.append(thumbnail)
 249
 250         return merge_dicts(info, {
 251             'title': title,
 252             'description': content.get('leadParagraph') or content.get('teasertext'),
 253             'duration': int_or_none(t.get('duration')),
 254             'timestamp': unified_timestamp(content.get('editorialDate')),
 255             'thumbnails': thumbnails,
 256         })
 257
 258     def _extract_regular(self, url, player, video_id):
 259         content = self._call_api(
 260             player['content'], video_id, 'content', player['apiToken'], url)
 261         return self._extract_entry(player['content'], player, content, video_id)
 262
 263     def _extract_mobile(self, video_id):
 264         video = self._download_json(
 265             'https://zdf-cdn.live.cellular.de/mediathekV2/document/%s' % video_id,
 266             video_id)
 267
 268         document = video['document']
 269
 270         title = document['titel']
 271         content_id = document['basename']
 272
 273         formats = []
 274         format_urls = set()
 275         for f in document['formitaeten']:
 276             self._extract_format(content_id, formats, format_urls, f)
 277         self._sort_formats(formats)
 278
 279         thumbnails = []
 280         teaser_bild = document.get('teaserBild')
 281         if isinstance(teaser_bild, dict):
 282             for thumbnail_key, thumbnail in teaser_bild.items():
 283                 thumbnail_url = try_get(
 284                     thumbnail, lambda x: x['url'], compat_str)
 285                 if thumbnail_url:
 286                     thumbnails.append({
 287                         'url': thumbnail_url,
 288                         'id': thumbnail_key,
 289                         'width': int_or_none(thumbnail.get('width')),
 290                         'height': int_or_none(thumbnail.get('height')),
 291                     })
 292
 293         return {
 294             'id': content_id,
 295             'title': title,
 296             'description': document.get('beschreibung'),
 297             'duration': int_or_none(document.get('length')),
 298             'timestamp': unified_timestamp(document.get('date')) or unified_timestamp(
 299                 try_get(video, lambda x: x['meta']['editorialDate'], compat_str)),
 300             'thumbnails': thumbnails,
 301             'subtitles': self._extract_subtitles(document),
 302             'formats': formats,
 303         }
 304
 305     def _real_extract(self, url):
 306         video_id = self._match_id(url)
 307
 308         webpage = self._download_webpage(url, video_id, fatal=False)
 309         if webpage:
 310             player = self._extract_player(webpage, url, fatal=False)
 311             if player:
 312                 return self._extract_regular(url, player, video_id)
 313
 314         return self._extract_mobile(video_id)
 315
 316
 317 class ZDFChannelIE(ZDFBaseIE):
 318     _VALID_URL = r'https?://www\.zdf\.de/(?:[^/]+/)*(?P<id>[^/?#&]+)'
 319     _TESTS = [{
 320         'url': 'https://www.zdf.de/sport/das-aktuelle-sportstudio',
 321         'info_dict': {
 322             'id': 'das-aktuelle-sportstudio',
 323             'title': 'das aktuelle sportstudio | ZDF',
 324         },
 325         'playlist_mincount': 23,
 326     }, {
 327         'url': 'https://www.zdf.de/dokumentation/planet-e',
 328         'info_dict': {
 329             'id': 'planet-e',
 330             'title': 'planet e.',
 331         },
 332         'playlist_mincount': 50,
 333     }, {
 334         'url': 'https://www.zdf.de/filme/taunuskrimi/',
 335         'only_matching': True,
 336     }]
 337
 338     @classmethod
 339     def suitable(cls, url):
 340         return False if ZDFIE.suitable(url) else super(ZDFChannelIE, cls).suitable(url)
 341
 342     def _real_extract(self, url):
 343         channel_id = self._match_id(url)
 344
 345         webpage = self._download_webpage(url, channel_id)
 346
 347         entries = [
 348             self.url_result(item_url, ie=ZDFIE.ie_key())
 349             for item_url in orderedSet(re.findall(
 350                 r'data-plusbar-url=["\'](http.+?\.html)', webpage))]
 351
 352         return self.playlist_result(
 353             entries, channel_id, self._og_search_title(webpage, fatal=False))
 354
 355         r"""
 356         player = self._extract_player(webpage, channel_id)
 357
 358         channel_id = self._search_regex(
 359             r'docId\s*:\s*(["\'])(?P<id>(?!\1).+?)\1', webpage,
 360             'channel id', group='id')
 361
 362         channel = self._call_api(
 363             'https://api.zdf.de/content/documents/%s.json' % channel_id,
 364             player, url, channel_id)
 365
 366         items = []
 367         for module in channel['module']:
 368             for teaser in try_get(module, lambda x: x['teaser'], list) or []:
 369                 t = try_get(
 370                     teaser, lambda x: x['http://zdf.de/rels/target'], dict)
 371                 if not t:
 372                     continue
 373                 items.extend(try_get(
 374                     t,
 375                     lambda x: x['resultsWithVideo']['http://zdf.de/rels/search/results'],
 376                     list) or [])
 377             items.extend(try_get(
 378                 module,
 379                 lambda x: x['filterRef']['resultsWithVideo']['http://zdf.de/rels/search/results'],
 380                 list) or [])
 381
 382         entries = []
 383         entry_urls = set()
 384         for item in items:
 385             t = try_get(item, lambda x: x['http://zdf.de/rels/target'], dict)
 386             if not t:
 387                 continue
 388             sharing_url = t.get('http://zdf.de/rels/sharing-url')
 389             if not sharing_url or not isinstance(sharing_url, compat_str):
 390                 continue
 391             if sharing_url in entry_urls:
 392                 continue
 393             entry_urls.add(sharing_url)
 394             entries.append(self.url_result(
 395                 sharing_url, ie=ZDFIE.ie_key(), video_id=t.get('id')))
 396
 397         return self.playlist_result(entries, channel_id, channel.get('title'))
 398         """