Raphaël G. Git Repositories - youtubedl/blob - 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     int_or_none,
  11     NO_DEFAULT,
  12     orderedSet,
  13     parse_codecs,
  14     qualities,
  15     try_get,
  16     unified_timestamp,
  17     update_url_query,
  18     url_or_none,
  19     urljoin,
  20 )
  21
  22
  23 class ZDFBaseIE(InfoExtractor):
  24     def _call_api(self, url, player, referrer, video_id, item):
  25         return self._download_json(
  26             url, video_id, 'Downloading JSON %s' % item,
  27             headers={
  28                 'Referer': referrer,
  29                 'Api-Auth': 'Bearer %s' % player['apiToken'],
  30             })
  31
  32     def _extract_player(self, webpage, video_id, fatal=True):
  33         return self._parse_json(
  34             self._search_regex(
  35                 r'(?s)data-zdfplayer-jsb=(["\'])(?P<json>{.+?})\1', webpage,
  36                 'player JSON', default='{}' if not fatal else NO_DEFAULT,
  37                 group='json'),
  38             video_id)
  39
  40
  41 class ZDFIE(ZDFBaseIE):
  42     _VALID_URL = r'https?://www\.zdf\.de/(?:[^/]+/)*(?P<id>[^/?]+)\.html'
  43     _QUALITIES = ('auto', 'low', 'med', 'high', 'veryhigh')
  44     _GEO_COUNTRIES = ['DE']
  45
  46     _TESTS = [{
  47         'url': 'https://www.zdf.de/dokumentation/terra-x/die-magie-der-farben-von-koenigspurpur-und-jeansblau-100.html',
  48         'info_dict': {
  49             'id': 'die-magie-der-farben-von-koenigspurpur-und-jeansblau-100',
  50             'ext': 'mp4',
  51             'title': 'Die Magie der Farben (2/2)',
  52             'description': 'md5:a89da10c928c6235401066b60a6d5c1a',
  53             'duration': 2615,
  54             'timestamp': 1465021200,
  55             'upload_date': '20160604',
  56         },
  57     }, {
  58         'url': 'https://www.zdf.de/service-und-hilfe/die-neue-zdf-mediathek/zdfmediathek-trailer-100.html',
  59         'only_matching': True,
  60     }, {
  61         'url': 'https://www.zdf.de/filme/taunuskrimi/die-lebenden-und-die-toten-1---ein-taunuskrimi-100.html',
  62         'only_matching': True,
  63     }, {
  64         'url': 'https://www.zdf.de/dokumentation/planet-e/planet-e-uebersichtsseite-weitere-dokumentationen-von-planet-e-100.html',
  65         'only_matching': True,
  66     }]
  67
  68     @staticmethod
  69     def _extract_subtitles(src):
  70         subtitles = {}
  71         for caption in try_get(src, lambda x: x['captions'], list) or []:
  72             subtitle_url = url_or_none(caption.get('uri'))
  73             if subtitle_url:
  74                 lang = caption.get('language', 'deu')
  75                 subtitles.setdefault(lang, []).append({
  76                     'url': subtitle_url,
  77                 })
  78         return subtitles
  79
  80     def _extract_format(self, video_id, formats, format_urls, meta):
  81         format_url = url_or_none(meta.get('url'))
  82         if not format_url:
  83             return
  84         if format_url in format_urls:
  85             return
  86         format_urls.add(format_url)
  87         mime_type = meta.get('mimeType')
  88         ext = determine_ext(format_url)
  89         if mime_type == 'application/x-mpegURL' or ext == 'm3u8':
  90             formats.extend(self._extract_m3u8_formats(
  91                 format_url, video_id, 'mp4', m3u8_id='hls',
  92                 entry_protocol='m3u8_native', fatal=False))
  93         elif mime_type == 'application/f4m+xml' or ext == 'f4m':
  94             formats.extend(self._extract_f4m_formats(
  95                 update_url_query(format_url, {'hdcore': '3.7.0'}), video_id, f4m_id='hds', fatal=False))
  96         else:
  97             f = parse_codecs(meta.get('mimeCodec'))
  98             format_id = ['http']
  99             for p in (meta.get('type'), meta.get('quality')):
 100                 if p and isinstance(p, compat_str):
 101                     format_id.append(p)
 102             f.update({
 103                 'url': format_url,
 104                 'format_id': '-'.join(format_id),
 105                 'format_note': meta.get('quality'),
 106                 'language': meta.get('language'),
 107                 'quality': qualities(self._QUALITIES)(meta.get('quality')),
 108                 'preference': -10,
 109             })
 110             formats.append(f)
 111
 112     def _extract_entry(self, url, player, content, video_id):
 113         title = content.get('title') or content['teaserHeadline']
 114
 115         t = content['mainVideoContent']['http://zdf.de/rels/target']
 116
 117         ptmd_path = t.get('http://zdf.de/rels/streams/ptmd')
 118
 119         if not ptmd_path:
 120             ptmd_path = t[
 121                 'http://zdf.de/rels/streams/ptmd-template'].replace(
 122                 '{playerId}', 'portal')
 123
 124         ptmd = self._call_api(
 125             urljoin(url, ptmd_path), player, url, video_id, 'metadata')
 126
 127         formats = []
 128         track_uris = set()
 129         for p in ptmd['priorityList']:
 130             formitaeten = p.get('formitaeten')
 131             if not isinstance(formitaeten, list):
 132                 continue
 133             for f in formitaeten:
 134                 f_qualities = f.get('qualities')
 135                 if not isinstance(f_qualities, list):
 136                     continue
 137                 for quality in f_qualities:
 138                     tracks = try_get(quality, lambda x: x['audio']['tracks'], list)
 139                     if not tracks:
 140                         continue
 141                     for track in tracks:
 142                         self._extract_format(
 143                             video_id, formats, track_uris, {
 144                                 'url': track.get('uri'),
 145                                 'type': f.get('type'),
 146                                 'mimeType': f.get('mimeType'),
 147                                 'quality': quality.get('quality'),
 148                                 'language': track.get('language'),
 149                             })
 150         self._sort_formats(formats)
 151
 152         thumbnails = []
 153         layouts = try_get(
 154             content, lambda x: x['teaserImageRef']['layouts'], dict)
 155         if layouts:
 156             for layout_key, layout_url in layouts.items():
 157                 layout_url = url_or_none(layout_url)
 158                 if not layout_url:
 159                     continue
 160                 thumbnail = {
 161                     'url': layout_url,
 162                     'format_id': layout_key,
 163                 }
 164                 mobj = re.search(r'(?P<width>\d+)x(?P<height>\d+)', layout_key)
 165                 if mobj:
 166                     thumbnail.update({
 167                         'width': int(mobj.group('width')),
 168                         'height': int(mobj.group('height')),
 169                     })
 170                 thumbnails.append(thumbnail)
 171
 172         return {
 173             'id': video_id,
 174             'title': title,
 175             'description': content.get('leadParagraph') or content.get('teasertext'),
 176             'duration': int_or_none(t.get('duration')),
 177             'timestamp': unified_timestamp(content.get('editorialDate')),
 178             'thumbnails': thumbnails,
 179             'subtitles': self._extract_subtitles(ptmd),
 180             'formats': formats,
 181         }
 182
 183     def _extract_regular(self, url, player, video_id):
 184         content = self._call_api(
 185             player['content'], player, url, video_id, 'content')
 186         return self._extract_entry(player['content'], player, content, video_id)
 187
 188     def _extract_mobile(self, video_id):
 189         document = self._download_json(
 190             'https://zdf-cdn.live.cellular.de/mediathekV2/document/%s' % video_id,
 191             video_id)['document']
 192
 193         title = document['titel']
 194
 195         formats = []
 196         format_urls = set()
 197         for f in document['formitaeten']:
 198             self._extract_format(video_id, formats, format_urls, f)
 199         self._sort_formats(formats)
 200
 201         thumbnails = []
 202         teaser_bild = document.get('teaserBild')
 203         if isinstance(teaser_bild, dict):
 204             for thumbnail_key, thumbnail in teaser_bild.items():
 205                 thumbnail_url = try_get(
 206                     thumbnail, lambda x: x['url'], compat_str)
 207                 if thumbnail_url:
 208                     thumbnails.append({
 209                         'url': thumbnail_url,
 210                         'id': thumbnail_key,
 211                         'width': int_or_none(thumbnail.get('width')),
 212                         'height': int_or_none(thumbnail.get('height')),
 213                     })
 214
 215         return {
 216             'id': video_id,
 217             'title': title,
 218             'description': document.get('beschreibung'),
 219             'duration': int_or_none(document.get('length')),
 220             'timestamp': unified_timestamp(try_get(
 221                 document, lambda x: x['meta']['editorialDate'], compat_str)),
 222             'thumbnails': thumbnails,
 223             'subtitles': self._extract_subtitles(document),
 224             'formats': formats,
 225         }
 226
 227     def _real_extract(self, url):
 228         video_id = self._match_id(url)
 229
 230         webpage = self._download_webpage(url, video_id, fatal=False)
 231         if webpage:
 232             player = self._extract_player(webpage, url, fatal=False)
 233             if player:
 234                 return self._extract_regular(url, player, video_id)
 235
 236         return self._extract_mobile(video_id)
 237
 238
 239 class ZDFChannelIE(ZDFBaseIE):
 240     _VALID_URL = r'https?://www\.zdf\.de/(?:[^/]+/)*(?P<id>[^/?#&]+)'
 241     _TESTS = [{
 242         'url': 'https://www.zdf.de/sport/das-aktuelle-sportstudio',
 243         'info_dict': {
 244             'id': 'das-aktuelle-sportstudio',
 245             'title': 'das aktuelle sportstudio | ZDF',
 246         },
 247         'playlist_mincount': 23,
 248     }, {
 249         'url': 'https://www.zdf.de/dokumentation/planet-e',
 250         'info_dict': {
 251             'id': 'planet-e',
 252             'title': 'planet e.',
 253         },
 254         'playlist_mincount': 50,
 255     }, {
 256         'url': 'https://www.zdf.de/filme/taunuskrimi/',
 257         'only_matching': True,
 258     }]
 259
 260     @classmethod
 261     def suitable(cls, url):
 262         return False if ZDFIE.suitable(url) else super(ZDFChannelIE, cls).suitable(url)
 263
 264     def _real_extract(self, url):
 265         channel_id = self._match_id(url)
 266
 267         webpage = self._download_webpage(url, channel_id)
 268
 269         entries = [
 270             self.url_result(item_url, ie=ZDFIE.ie_key())
 271             for item_url in orderedSet(re.findall(
 272                 r'data-plusbar-url=["\'](http.+?\.html)', webpage))]
 273
 274         return self.playlist_result(
 275             entries, channel_id, self._og_search_title(webpage, fatal=False))
 276
 277         r"""
 278         player = self._extract_player(webpage, channel_id)
 279
 280         channel_id = self._search_regex(
 281             r'docId\s*:\s*(["\'])(?P<id>(?!\1).+?)\1', webpage,
 282             'channel id', group='id')
 283
 284         channel = self._call_api(
 285             'https://api.zdf.de/content/documents/%s.json' % channel_id,
 286             player, url, channel_id)
 287
 288         items = []
 289         for module in channel['module']:
 290             for teaser in try_get(module, lambda x: x['teaser'], list) or []:
 291                 t = try_get(
 292                     teaser, lambda x: x['http://zdf.de/rels/target'], dict)
 293                 if not t:
 294                     continue
 295                 items.extend(try_get(
 296                     t,
 297                     lambda x: x['resultsWithVideo']['http://zdf.de/rels/search/results'],
 298                     list) or [])
 299             items.extend(try_get(
 300                 module,
 301                 lambda x: x['filterRef']['resultsWithVideo']['http://zdf.de/rels/search/results'],
 302                 list) or [])
 303
 304         entries = []
 305         entry_urls = set()
 306         for item in items:
 307             t = try_get(item, lambda x: x['http://zdf.de/rels/target'], dict)
 308             if not t:
 309                 continue
 310             sharing_url = t.get('http://zdf.de/rels/sharing-url')
 311             if not sharing_url or not isinstance(sharing_url, compat_str):
 312                 continue
 313             if sharing_url in entry_urls:
 314                 continue
 315             entry_urls.add(sharing_url)
 316             entries.append(self.url_result(
 317                 sharing_url, ie=ZDFIE.ie_key(), video_id=t.get('id')))
 318
 319         return self.playlist_result(entries, channel_id, channel.get('title'))
 320         """