yt_dlp/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     float_or_none,
  11     int_or_none,
  12     merge_dicts,
  13     NO_DEFAULT,
  14     orderedSet,
  15     parse_codecs,
  16     qualities,
  17     try_get,
  18     unified_timestamp,
  19     update_url_query,
  20     url_or_none,
  21     urljoin,
  22 )
  23
  24
  25 class ZDFBaseIE(InfoExtractor):
  26     _GEO_COUNTRIES = ['DE']
  27     _QUALITIES = ('auto', 'low', 'med', 'high', 'veryhigh', 'hd')
  28
  29     def _call_api(self, url, video_id, item, api_token=None, referrer=None):
  30         headers = {}
  31         if api_token:
  32             headers['Api-Auth'] = 'Bearer %s' % api_token
  33         if referrer:
  34             headers['Referer'] = referrer
  35         return self._download_json(
  36             url, video_id, 'Downloading JSON %s' % item, headers=headers)
  37
  38     @staticmethod
  39     def _extract_subtitles(src):
  40         subtitles = {}
  41         for caption in try_get(src, lambda x: x['captions'], list) or []:
  42             subtitle_url = url_or_none(caption.get('uri'))
  43             if subtitle_url:
  44                 lang = caption.get('language', 'deu')
  45                 subtitles.setdefault(lang, []).append({
  46                     'url': subtitle_url,
  47                 })
  48         return subtitles
  49
  50     def _extract_format(self, video_id, formats, format_urls, meta):
  51         format_url = url_or_none(meta.get('url'))
  52         if not format_url:
  53             return
  54         if format_url in format_urls:
  55             return
  56         format_urls.add(format_url)
  57         mime_type = meta.get('mimeType')
  58         ext = determine_ext(format_url)
  59         if mime_type == 'application/x-mpegURL' or ext == 'm3u8':
  60             formats.extend(self._extract_m3u8_formats(
  61                 format_url, video_id, 'mp4', m3u8_id='hls',
  62                 entry_protocol='m3u8_native', fatal=False))
  63         elif mime_type == 'application/f4m+xml' or ext == 'f4m':
  64             formats.extend(self._extract_f4m_formats(
  65                 update_url_query(format_url, {'hdcore': '3.7.0'}), video_id, f4m_id='hds', fatal=False))
  66         else:
  67             f = parse_codecs(meta.get('mimeCodec'))
  68             format_id = ['http']
  69             for p in (meta.get('type'), meta.get('quality')):
  70                 if p and isinstance(p, compat_str):
  71                     format_id.append(p)
  72             f.update({
  73                 'url': format_url,
  74                 'format_id': '-'.join(format_id),
  75                 'format_note': meta.get('quality'),
  76                 'language': meta.get('language'),
  77                 'quality': qualities(self._QUALITIES)(meta.get('quality')),
  78                 'preference': -10,
  79             })
  80             formats.append(f)
  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                                 'language': track.get('language'),
 110                             })
 111         self._sort_formats(formats)
 112
 113         duration = float_or_none(try_get(
 114             ptmd, lambda x: x['attributes']['duration']['value']), scale=1000)
 115
 116         return {
 117             'extractor_key': ZDFIE.ie_key(),
 118             'id': content_id,
 119             'duration': duration,
 120             'formats': formats,
 121             'subtitles': self._extract_subtitles(ptmd),
 122         }
 123
 124     def _extract_player(self, webpage, video_id, fatal=True):
 125         return self._parse_json(
 126             self._search_regex(
 127                 r'(?s)data-zdfplayer-jsb=(["\'])(?P<json>{.+?})\1', webpage,
 128                 'player JSON', default='{}' if not fatal else NO_DEFAULT,
 129                 group='json'),
 130             video_id)
 131
 132
 133 class ZDFIE(ZDFBaseIE):
 134     _VALID_URL = r'https?://www\.zdf\.de/(?:[^/]+/)*(?P<id>[^/?#&]+)\.html'
 135     _TESTS = [{
 136         # Same as https://www.phoenix.de/sendungen/ereignisse/corona-nachgehakt/wohin-fuehrt-der-protest-in-der-pandemie-a-2050630.html
 137         'url': 'https://www.zdf.de/politik/phoenix-sendungen/wohin-fuehrt-der-protest-in-der-pandemie-100.html',
 138         'md5': '34ec321e7eb34231fd88616c65c92db0',
 139         'info_dict': {
 140             'id': '210222_phx_nachgehakt_corona_protest',
 141             'ext': 'mp4',
 142             'title': 'Wohin führt der Protest in der Pandemie?',
 143             'description': 'md5:7d643fe7f565e53a24aac036b2122fbd',
 144             'duration': 1691,
 145             'timestamp': 1613948400,
 146             'upload_date': '20210221',
 147         },
 148     }, {
 149         # Same as https://www.3sat.de/film/ab-18/10-wochen-sommer-108.html
 150         'url': 'https://www.zdf.de/dokumentation/ab-18/10-wochen-sommer-102.html',
 151         'md5': '0aff3e7bc72c8813f5e0fae333316a1d',
 152         'info_dict': {
 153             'id': '141007_ab18_10wochensommer_film',
 154             'ext': 'mp4',
 155             'title': 'Ab 18! - 10 Wochen Sommer',
 156             'description': 'md5:8253f41dc99ce2c3ff892dac2d65fe26',
 157             'duration': 2660,
 158             'timestamp': 1608604200,
 159             'upload_date': '20201222',
 160         },
 161     }, {
 162         'url': 'https://www.zdf.de/dokumentation/terra-x/die-magie-der-farben-von-koenigspurpur-und-jeansblau-100.html',
 163         'info_dict': {
 164             'id': '151025_magie_farben2_tex',
 165             'ext': 'mp4',
 166             'title': 'Die Magie der Farben (2/2)',
 167             'description': 'md5:a89da10c928c6235401066b60a6d5c1a',
 168             'duration': 2615,
 169             'timestamp': 1465021200,
 170             'upload_date': '20160604',
 171         },
 172     }, {
 173         # Same as https://www.phoenix.de/sendungen/dokumentationen/gesten-der-maechtigen-i-a-89468.html?ref=suche
 174         'url': 'https://www.zdf.de/politik/phoenix-sendungen/die-gesten-der-maechtigen-100.html',
 175         'only_matching': True,
 176     }, {
 177         # Same as https://www.3sat.de/film/spielfilm/der-hauptmann-100.html
 178         'url': 'https://www.zdf.de/filme/filme-sonstige/der-hauptmann-112.html',
 179         'only_matching': True,
 180     }, {
 181         # Same as https://www.3sat.de/wissen/nano/nano-21-mai-2019-102.html, equal media ids
 182         'url': 'https://www.zdf.de/wissen/nano/nano-21-mai-2019-102.html',
 183         'only_matching': True,
 184     }, {
 185         'url': 'https://www.zdf.de/service-und-hilfe/die-neue-zdf-mediathek/zdfmediathek-trailer-100.html',
 186         'only_matching': True,
 187     }, {
 188         'url': 'https://www.zdf.de/filme/taunuskrimi/die-lebenden-und-die-toten-1---ein-taunuskrimi-100.html',
 189         'only_matching': True,
 190     }, {
 191         'url': 'https://www.zdf.de/dokumentation/planet-e/planet-e-uebersichtsseite-weitere-dokumentationen-von-planet-e-100.html',
 192         'only_matching': True,
 193     }]
 194
 195     def _extract_entry(self, url, player, content, video_id):
 196         title = content.get('title') or content['teaserHeadline']
 197
 198         t = content['mainVideoContent']['http://zdf.de/rels/target']
 199
 200         ptmd_path = t.get('http://zdf.de/rels/streams/ptmd')
 201
 202         if not ptmd_path:
 203             ptmd_path = t[
 204                 'http://zdf.de/rels/streams/ptmd-template'].replace(
 205                 '{playerId}', 'ngplayer_2_4')
 206
 207         info = self._extract_ptmd(
 208             urljoin(url, ptmd_path), video_id, player['apiToken'], url)
 209
 210         thumbnails = []
 211         layouts = try_get(
 212             content, lambda x: x['teaserImageRef']['layouts'], dict)
 213         if layouts:
 214             for layout_key, layout_url in layouts.items():
 215                 layout_url = url_or_none(layout_url)
 216                 if not layout_url:
 217                     continue
 218                 thumbnail = {
 219                     'url': layout_url,
 220                     'format_id': layout_key,
 221                 }
 222                 mobj = re.search(r'(?P<width>\d+)x(?P<height>\d+)', layout_key)
 223                 if mobj:
 224                     thumbnail.update({
 225                         'width': int(mobj.group('width')),
 226                         'height': int(mobj.group('height')),
 227                     })
 228                 thumbnails.append(thumbnail)
 229
 230         return merge_dicts(info, {
 231             'title': title,
 232             'description': content.get('leadParagraph') or content.get('teasertext'),
 233             'duration': int_or_none(t.get('duration')),
 234             'timestamp': unified_timestamp(content.get('editorialDate')),
 235             'thumbnails': thumbnails,
 236         })
 237
 238     def _extract_regular(self, url, player, video_id):
 239         content = self._call_api(
 240             player['content'], video_id, 'content', player['apiToken'], url)
 241         return self._extract_entry(player['content'], player, content, video_id)
 242
 243     def _extract_mobile(self, video_id):
 244         video = self._download_json(
 245             'https://zdf-cdn.live.cellular.de/mediathekV2/document/%s' % video_id,
 246             video_id)
 247
 248         document = video['document']
 249
 250         title = document['titel']
 251         content_id = document['basename']
 252
 253         formats = []
 254         format_urls = set()
 255         for f in document['formitaeten']:
 256             self._extract_format(content_id, formats, format_urls, f)
 257         self._sort_formats(formats)
 258
 259         thumbnails = []
 260         teaser_bild = document.get('teaserBild')
 261         if isinstance(teaser_bild, dict):
 262             for thumbnail_key, thumbnail in teaser_bild.items():
 263                 thumbnail_url = try_get(
 264                     thumbnail, lambda x: x['url'], compat_str)
 265                 if thumbnail_url:
 266                     thumbnails.append({
 267                         'url': thumbnail_url,
 268                         'id': thumbnail_key,
 269                         'width': int_or_none(thumbnail.get('width')),
 270                         'height': int_or_none(thumbnail.get('height')),
 271                     })
 272
 273         return {
 274             'id': content_id,
 275             'title': title,
 276             'description': document.get('beschreibung'),
 277             'duration': int_or_none(document.get('length')),
 278             'timestamp': unified_timestamp(document.get('date')) or unified_timestamp(
 279                 try_get(video, lambda x: x['meta']['editorialDate'], compat_str)),
 280             'thumbnails': thumbnails,
 281             'subtitles': self._extract_subtitles(document),
 282             'formats': formats,
 283         }
 284
 285     def _real_extract(self, url):
 286         video_id = self._match_id(url)
 287
 288         webpage = self._download_webpage(url, video_id, fatal=False)
 289         if webpage:
 290             player = self._extract_player(webpage, url, fatal=False)
 291             if player:
 292                 return self._extract_regular(url, player, video_id)
 293
 294         return self._extract_mobile(video_id)
 295
 296
 297 class ZDFChannelIE(ZDFBaseIE):
 298     _VALID_URL = r'https?://www\.zdf\.de/(?:[^/]+/)*(?P<id>[^/?#&]+)'
 299     _TESTS = [{
 300         'url': 'https://www.zdf.de/sport/das-aktuelle-sportstudio',
 301         'info_dict': {
 302             'id': 'das-aktuelle-sportstudio',
 303             'title': 'das aktuelle sportstudio | ZDF',
 304         },
 305         'playlist_mincount': 23,
 306     }, {
 307         'url': 'https://www.zdf.de/dokumentation/planet-e',
 308         'info_dict': {
 309             'id': 'planet-e',
 310             'title': 'planet e.',
 311         },
 312         'playlist_mincount': 50,
 313     }, {
 314         'url': 'https://www.zdf.de/filme/taunuskrimi/',
 315         'only_matching': True,
 316     }]
 317
 318     @classmethod
 319     def suitable(cls, url):
 320         return False if ZDFIE.suitable(url) else super(ZDFChannelIE, cls).suitable(url)
 321
 322     def _real_extract(self, url):
 323         channel_id = self._match_id(url)
 324
 325         webpage = self._download_webpage(url, channel_id)
 326
 327         entries = [
 328             self.url_result(item_url, ie=ZDFIE.ie_key())
 329             for item_url in orderedSet(re.findall(
 330                 r'data-plusbar-url=["\'](http.+?\.html)', webpage))]
 331
 332         return self.playlist_result(
 333             entries, channel_id, self._og_search_title(webpage, fatal=False))
 334
 335         r"""
 336         player = self._extract_player(webpage, channel_id)
 337
 338         channel_id = self._search_regex(
 339             r'docId\s*:\s*(["\'])(?P<id>(?!\1).+?)\1', webpage,
 340             'channel id', group='id')
 341
 342         channel = self._call_api(
 343             'https://api.zdf.de/content/documents/%s.json' % channel_id,
 344             player, url, channel_id)
 345
 346         items = []
 347         for module in channel['module']:
 348             for teaser in try_get(module, lambda x: x['teaser'], list) or []:
 349                 t = try_get(
 350                     teaser, lambda x: x['http://zdf.de/rels/target'], dict)
 351                 if not t:
 352                     continue
 353                 items.extend(try_get(
 354                     t,
 355                     lambda x: x['resultsWithVideo']['http://zdf.de/rels/search/results'],
 356                     list) or [])
 357             items.extend(try_get(
 358                 module,
 359                 lambda x: x['filterRef']['resultsWithVideo']['http://zdf.de/rels/search/results'],
 360                 list) or [])
 361
 362         entries = []
 363         entry_urls = set()
 364         for item in items:
 365             t = try_get(item, lambda x: x['http://zdf.de/rels/target'], dict)
 366             if not t:
 367                 continue
 368             sharing_url = t.get('http://zdf.de/rels/sharing-url')
 369             if not sharing_url or not isinstance(sharing_url, compat_str):
 370                 continue
 371             if sharing_url in entry_urls:
 372                 continue
 373             entry_urls.add(sharing_url)
 374             entries.append(self.url_result(
 375                 sharing_url, ie=ZDFIE.ie_key(), video_id=t.get('id')))
 376
 377         return self.playlist_result(entries, channel_id, channel.get('title'))
 378         """