2 from __future__ import unicode_literals
6 from .common import InfoExtractor
7 from ..compat import compat_str
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,
29 'Api-Auth': 'Bearer %s' % player['apiToken'],
32 def _extract_player(self, webpage, video_id, fatal=True):
33 return self._parse_json(
35 r'(?s)data-zdfplayer-jsb=(["\'])(?P<json>{.+?})\1', webpage,
36 'player JSON', default='{}' if not fatal else NO_DEFAULT,
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']
47 'url': 'https://www.zdf.de/dokumentation/terra-x/die-magie-der-farben-von-koenigspurpur-und-jeansblau-100.html',
49 'id': 'die-magie-der-farben-von-koenigspurpur-und-jeansblau-100',
51 'title': 'Die Magie der Farben (2/2)',
52 'description': 'md5:a89da10c928c6235401066b60a6d5c1a',
54 'timestamp': 1465021200,
55 'upload_date': '20160604',
58 'url': 'https://www.zdf.de/service-und-hilfe/die-neue-zdf-mediathek/zdfmediathek-trailer-100.html',
59 'only_matching': True,
61 'url': 'https://www.zdf.de/filme/taunuskrimi/die-lebenden-und-die-toten-1---ein-taunuskrimi-100.html',
62 'only_matching': True,
64 'url': 'https://www.zdf.de/dokumentation/planet-e/planet-e-uebersichtsseite-weitere-dokumentationen-von-planet-e-100.html',
65 'only_matching': True,
69 def _extract_subtitles(src):
71 for caption in try_get(src, lambda x: x['captions'], list) or []:
72 subtitle_url = url_or_none(caption.get('uri'))
74 lang = caption.get('language', 'deu')
75 subtitles.setdefault(lang, []).append({
80 def _extract_format(self, video_id, formats, format_urls, meta):
81 format_url = url_or_none(meta.get('url'))
84 if format_url in format_urls:
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))
97 f = parse_codecs(meta.get('mimeCodec'))
99 for p in (meta.get('type'), meta.get('quality')):
100 if p and isinstance(p, compat_str):
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')),
112 def _extract_entry(self, url, player, content, video_id):
113 title = content.get('title') or content['teaserHeadline']
115 t = content['mainVideoContent']['http://zdf.de/rels/target']
117 ptmd_path = t.get('http://zdf.de/rels/streams/ptmd')
121 'http://zdf.de/rels/streams/ptmd-template'].replace(
122 '{playerId}', 'portal')
124 ptmd = self._call_api(
125 urljoin(url, ptmd_path), player, url, video_id, 'metadata')
129 for p in ptmd['priorityList']:
130 formitaeten = p.get('formitaeten')
131 if not isinstance(formitaeten, list):
133 for f in formitaeten:
134 f_qualities = f.get('qualities')
135 if not isinstance(f_qualities, list):
137 for quality in f_qualities:
138 tracks = try_get(quality, lambda x: x['audio']['tracks'], list)
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'),
150 self._sort_formats(formats)
154 content, lambda x: x['teaserImageRef']['layouts'], dict)
156 for layout_key, layout_url in layouts.items():
157 layout_url = url_or_none(layout_url)
162 'format_id': layout_key,
164 mobj = re.search(r'(?P<width>\d+)x(?P<height>\d+)', layout_key)
167 'width': int(mobj.group('width')),
168 'height': int(mobj.group('height')),
170 thumbnails.append(thumbnail)
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),
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)
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']
193 title = document['titel']
197 for f in document['formitaeten']:
198 self._extract_format(video_id, formats, format_urls, f)
199 self._sort_formats(formats)
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)
209 'url': thumbnail_url,
211 'width': int_or_none(thumbnail.get('width')),
212 'height': int_or_none(thumbnail.get('height')),
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),
227 def _real_extract(self, url):
228 video_id = self._match_id(url)
230 webpage = self._download_webpage(url, video_id, fatal=False)
232 player = self._extract_player(webpage, url, fatal=False)
234 return self._extract_regular(url, player, video_id)
236 return self._extract_mobile(video_id)
239 class ZDFChannelIE(ZDFBaseIE):
240 _VALID_URL = r'https?://www\.zdf\.de/(?:[^/]+/)*(?P<id>[^/?#&]+)'
242 'url': 'https://www.zdf.de/sport/das-aktuelle-sportstudio',
244 'id': 'das-aktuelle-sportstudio',
245 'title': 'das aktuelle sportstudio | ZDF',
247 'playlist_count': 21,
249 'url': 'https://www.zdf.de/dokumentation/planet-e',
252 'title': 'planet e.',
256 'url': 'https://www.zdf.de/filme/taunuskrimi/',
257 'only_matching': True,
261 def suitable(cls, url):
262 return False if ZDFIE.suitable(url) else super(ZDFChannelIE, cls).suitable(url)
264 def _real_extract(self, url):
265 channel_id = self._match_id(url)
267 webpage = self._download_webpage(url, channel_id)
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))]
274 return self.playlist_result(
275 entries, channel_id, self._og_search_title(webpage, fatal=False))
278 player = self._extract_player(webpage, channel_id)
280 channel_id = self._search_regex(
281 r'docId\s*:\s*(["\'])(?P<id>(?!\1).+?)\1', webpage,
282 'channel id', group='id')
284 channel = self._call_api(
285 'https://api.zdf.de/content/documents/%s.json' % channel_id,
286 player, url, channel_id)
289 for module in channel['module']:
290 for teaser in try_get(module, lambda x: x['teaser'], list) or []:
292 teaser, lambda x: x['http://zdf.de/rels/target'], dict)
295 items.extend(try_get(
297 lambda x: x['resultsWithVideo']['http://zdf.de/rels/search/results'],
299 items.extend(try_get(
301 lambda x: x['filterRef']['resultsWithVideo']['http://zdf.de/rels/search/results'],
307 t = try_get(item, lambda x: x['http://zdf.de/rels/target'], dict)
310 sharing_url = t.get('http://zdf.de/rels/sharing-url')
311 if not sharing_url or not isinstance(sharing_url, compat_str):
313 if sharing_url in entry_urls:
315 entry_urls.add(sharing_url)
316 entries.append(self.url_result(
317 sharing_url, ie=ZDFIE.ie_key(), video_id=t.get('id')))
319 return self.playlist_result(entries, channel_id, channel.get('title'))