# coding: utf-8
+from __future__ import unicode_literals
import re
from .common import InfoExtractor
+from ..compat import compat_str
from ..utils import (
+ determine_ext,
int_or_none,
- unified_strdate,
+ NO_DEFAULT,
+ orderedSet,
+ parse_codecs,
+ qualities,
+ try_get,
+ unified_timestamp,
+ update_url_query,
+ urljoin,
)
-class ZDFIE(InfoExtractor):
- _VALID_URL = r'^https?://www\.zdf\.de/ZDFmediathek(?P<hash>#)?/(.*beitrag/(?:video/)?)(?P<video_id>[0-9]+)(?:/[^/?]+)?(?:\?.*)?'
+class ZDFBaseIE(InfoExtractor):
+ def _call_api(self, url, player, referrer, video_id):
+ return self._download_json(
+ url, video_id, 'Downloading JSON content',
+ headers={
+ 'Referer': referrer,
+ 'Api-Auth': 'Bearer %s' % player['apiToken'],
+ })
- _TEST = {
- u"url": u"http://www.zdf.de/ZDFmediathek/beitrag/video/2037704/ZDFspezial---Ende-des-Machtpokers--?bc=sts;stt",
- u"file": u"2037704.webm",
- u"info_dict": {
- u"upload_date": u"20131127",
- u"description": u"Union und SPD haben sich auf einen Koalitionsvertrag geeinigt. Aber was bedeutet das für die Bürger? Sehen Sie hierzu das ZDFspezial \"Ende des Machtpokers - Große Koalition für Deutschland\".",
- u"uploader": u"spezial",
- u"title": u"ZDFspezial - Ende des Machtpokers"
- },
- u"skip": u"Videos on ZDF.de are depublicised in short order",
- }
+ def _extract_player(self, webpage, video_id, fatal=True):
+ return self._parse_json(
+ self._search_regex(
+ r'(?s)data-zdfplayer-jsb=(["\'])(?P<json>{.+?})\1', webpage,
+ 'player JSON', default='{}' if not fatal else NO_DEFAULT,
+ group='json'),
+ video_id)
- def _real_extract(self, url):
- mobj = re.match(self._VALID_URL, url)
- video_id = mobj.group('video_id')
-
- xml_url = u'http://www.zdf.de/ZDFmediathek/xmlservice/web/beitragsDetails?ak=web&id=%s' % video_id
- doc = self._download_xml(
- xml_url, video_id,
- note=u'Downloading video info',
- errnote=u'Failed to download video info')
-
- title = doc.find('.//information/title').text
- description = doc.find('.//information/detail').text
- uploader_node = doc.find('.//details/originChannelTitle')
- uploader = None if uploader_node is None else uploader_node.text
- duration_str = doc.find('.//details/length').text
- duration_m = re.match(r'''(?x)^
- (?P<hours>[0-9]{2})
- :(?P<minutes>[0-9]{2})
- :(?P<seconds>[0-9]{2})
- (?:\.(?P<ms>[0-9]+)?)
- ''', duration_str)
- duration = (
- (
- (int(duration_m.group('hours')) * 60 * 60) +
- (int(duration_m.group('minutes')) * 60) +
- int(duration_m.group('seconds'))
- )
- if duration_m
- else None
- )
- upload_date = unified_strdate(doc.find('.//details/airtime').text)
-
- def xml_to_format(fnode):
- video_url = fnode.find('url').text
- is_available = u'http://www.metafilegenerator' not in video_url
-
- format_id = fnode.attrib['basetype']
- format_m = re.match(r'''(?x)
- (?P<vcodec>[^_]+)_(?P<acodec>[^_]+)_(?P<container>[^_]+)_
- (?P<proto>[^_]+)_(?P<index>[^_]+)_(?P<indexproto>[^_]+)
- ''', format_id)
-
- ext = format_m.group('container')
- proto = format_m.group('proto').lower()
-
- quality = fnode.find('./quality').text
- abr = int(fnode.find('./audioBitrate').text) // 1000
- vbr = int(fnode.find('./videoBitrate').text) // 1000
-
- format_note = u''
- if not format_note:
- format_note = None
-
- return {
- 'format_id': format_id + u'-' + quality,
- 'url': video_url,
- 'ext': ext,
- 'acodec': format_m.group('acodec'),
- 'vcodec': format_m.group('vcodec'),
- 'abr': abr,
- 'vbr': vbr,
- 'width': int_or_none(fnode.find('./width').text),
- 'height': int_or_none(fnode.find('./height').text),
- 'filesize': int_or_none(fnode.find('./filesize').text),
- 'format_note': format_note,
- 'protocol': proto,
- '_available': is_available,
- }
-
- format_nodes = doc.findall('.//formitaeten/formitaet')
- formats = list(filter(
- lambda f: f['_available'],
- map(xml_to_format, format_nodes)))
+class ZDFIE(ZDFBaseIE):
+ _VALID_URL = r'https?://www\.zdf\.de/(?:[^/]+/)*(?P<id>[^/?]+)\.html'
+ _QUALITIES = ('auto', 'low', 'med', 'high', 'veryhigh')
+
+ _TESTS = [{
+ 'url': 'https://www.zdf.de/service-und-hilfe/die-neue-zdf-mediathek/zdfmediathek-trailer-100.html',
+ 'info_dict': {
+ 'id': 'zdfmediathek-trailer-100',
+ 'ext': 'mp4',
+ 'title': 'Die neue ZDFmediathek',
+ 'description': 'md5:3003d36487fb9a5ea2d1ff60beb55e8d',
+ 'duration': 30,
+ 'timestamp': 1477627200,
+ 'upload_date': '20161028',
+ }
+ }, {
+ 'url': 'https://www.zdf.de/filme/taunuskrimi/die-lebenden-und-die-toten-1---ein-taunuskrimi-100.html',
+ 'only_matching': True,
+ }, {
+ 'url': 'https://www.zdf.de/dokumentation/planet-e/planet-e-uebersichtsseite-weitere-dokumentationen-von-planet-e-100.html',
+ 'only_matching': True,
+ }]
+
+ @staticmethod
+ def _extract_subtitles(src):
+ subtitles = {}
+ for caption in try_get(src, lambda x: x['captions'], list) or []:
+ subtitle_url = caption.get('uri')
+ if subtitle_url and isinstance(subtitle_url, compat_str):
+ lang = caption.get('language', 'deu')
+ subtitles.setdefault(lang, []).append({
+ 'url': subtitle_url,
+ })
+ return subtitles
+
+ def _extract_format(self, video_id, formats, format_urls, meta):
+ format_url = meta.get('url')
+ if not format_url or not isinstance(format_url, compat_str):
+ return
+ if format_url in format_urls:
+ return
+ format_urls.add(format_url)
+ mime_type = meta.get('mimeType')
+ ext = determine_ext(format_url)
+ if mime_type == 'application/x-mpegURL' or ext == 'm3u8':
+ formats.extend(self._extract_m3u8_formats(
+ format_url, video_id, 'mp4', m3u8_id='hls',
+ entry_protocol='m3u8_native', fatal=False))
+ elif mime_type == 'application/f4m+xml' or ext == 'f4m':
+ formats.extend(self._extract_f4m_formats(
+ update_url_query(format_url, {'hdcore': '3.7.0'}), video_id, f4m_id='hds', fatal=False))
+ else:
+ f = parse_codecs(meta.get('mimeCodec'))
+ format_id = ['http']
+ for p in (meta.get('type'), meta.get('quality')):
+ if p and isinstance(p, compat_str):
+ format_id.append(p)
+ f.update({
+ 'url': format_url,
+ 'format_id': '-'.join(format_id),
+ 'format_note': meta.get('quality'),
+ 'language': meta.get('language'),
+ 'quality': qualities(self._QUALITIES)(meta.get('quality')),
+ 'preference': -10,
+ })
+ formats.append(f)
+
+ def _extract_entry(self, url, content, video_id):
+ title = content.get('title') or content['teaserHeadline']
+
+ t = content['mainVideoContent']['http://zdf.de/rels/target']
+
+ ptmd_path = t.get('http://zdf.de/rels/streams/ptmd')
+
+ if not ptmd_path:
+ ptmd_path = t[
+ 'http://zdf.de/rels/streams/ptmd-template'].replace(
+ '{playerId}', 'portal')
+
+ ptmd = self._download_json(urljoin(url, ptmd_path), video_id)
+
+ formats = []
+ track_uris = set()
+ for p in ptmd['priorityList']:
+ formitaeten = p.get('formitaeten')
+ if not isinstance(formitaeten, list):
+ continue
+ for f in formitaeten:
+ f_qualities = f.get('qualities')
+ if not isinstance(f_qualities, list):
+ continue
+ for quality in f_qualities:
+ tracks = try_get(quality, lambda x: x['audio']['tracks'], list)
+ if not tracks:
+ continue
+ for track in tracks:
+ self._extract_format(
+ video_id, formats, track_uris, {
+ 'url': track.get('uri'),
+ 'type': f.get('type'),
+ 'mimeType': f.get('mimeType'),
+ 'quality': quality.get('quality'),
+ 'language': track.get('language'),
+ })
self._sort_formats(formats)
+ thumbnails = []
+ layouts = try_get(
+ content, lambda x: x['teaserImageRef']['layouts'], dict)
+ if layouts:
+ for layout_key, layout_url in layouts.items():
+ if not isinstance(layout_url, compat_str):
+ continue
+ thumbnail = {
+ 'url': layout_url,
+ 'format_id': layout_key,
+ }
+ mobj = re.search(r'(?P<width>\d+)x(?P<height>\d+)', layout_key)
+ if mobj:
+ thumbnail.update({
+ 'width': int(mobj.group('width')),
+ 'height': int(mobj.group('height')),
+ })
+ thumbnails.append(thumbnail)
+
return {
'id': video_id,
'title': title,
+ 'description': content.get('leadParagraph') or content.get('teasertext'),
+ 'duration': int_or_none(t.get('duration')),
+ 'timestamp': unified_timestamp(content.get('editorialDate')),
+ 'thumbnails': thumbnails,
+ 'subtitles': self._extract_subtitles(ptmd),
'formats': formats,
- 'description': description,
- 'uploader': uploader,
- 'duration': duration,
- 'upload_date': upload_date,
}
+
+ def _extract_regular(self, url, player, video_id):
+ content = self._call_api(player['content'], player, url, video_id)
+ return self._extract_entry(player['content'], content, video_id)
+
+ def _extract_mobile(self, video_id):
+ document = self._download_json(
+ 'https://zdf-cdn.live.cellular.de/mediathekV2/document/%s' % video_id,
+ video_id)['document']
+
+ title = document['titel']
+
+ formats = []
+ format_urls = set()
+ for f in document['formitaeten']:
+ self._extract_format(video_id, formats, format_urls, f)
+ self._sort_formats(formats)
+
+ thumbnails = []
+ teaser_bild = document.get('teaserBild')
+ if isinstance(teaser_bild, dict):
+ for thumbnail_key, thumbnail in teaser_bild.items():
+ thumbnail_url = try_get(
+ thumbnail, lambda x: x['url'], compat_str)
+ if thumbnail_url:
+ thumbnails.append({
+ 'url': thumbnail_url,
+ 'id': thumbnail_key,
+ 'width': int_or_none(thumbnail.get('width')),
+ 'height': int_or_none(thumbnail.get('height')),
+ })
+
+ return {
+ 'id': video_id,
+ 'title': title,
+ 'description': document.get('beschreibung'),
+ 'duration': int_or_none(document.get('length')),
+ 'timestamp': unified_timestamp(try_get(
+ document, lambda x: x['meta']['editorialDate'], compat_str)),
+ 'thumbnails': thumbnails,
+ 'subtitles': self._extract_subtitles(document),
+ 'formats': formats,
+ }
+
+ def _real_extract(self, url):
+ video_id = self._match_id(url)
+
+ webpage = self._download_webpage(url, video_id, fatal=False)
+ if webpage:
+ player = self._extract_player(webpage, url, fatal=False)
+ if player:
+ return self._extract_regular(url, player, video_id)
+
+ return self._extract_mobile(video_id)
+
+
+class ZDFChannelIE(ZDFBaseIE):
+ _VALID_URL = r'https?://www\.zdf\.de/(?:[^/]+/)*(?P<id>[^/?#&]+)'
+ _TESTS = [{
+ 'url': 'https://www.zdf.de/sport/das-aktuelle-sportstudio',
+ 'info_dict': {
+ 'id': 'das-aktuelle-sportstudio',
+ 'title': 'das aktuelle sportstudio | ZDF',
+ },
+ 'playlist_count': 21,
+ }, {
+ 'url': 'https://www.zdf.de/dokumentation/planet-e',
+ 'info_dict': {
+ 'id': 'planet-e',
+ 'title': 'planet e.',
+ },
+ 'playlist_count': 4,
+ }, {
+ 'url': 'https://www.zdf.de/filme/taunuskrimi/',
+ 'only_matching': True,
+ }]
+
+ @classmethod
+ def suitable(cls, url):
+ return False if ZDFIE.suitable(url) else super(ZDFChannelIE, cls).suitable(url)
+
+ def _real_extract(self, url):
+ channel_id = self._match_id(url)
+
+ webpage = self._download_webpage(url, channel_id)
+
+ entries = [
+ self.url_result(item_url, ie=ZDFIE.ie_key())
+ for item_url in orderedSet(re.findall(
+ r'data-plusbar-url=["\'](http.+?\.html)', webpage))]
+
+ return self.playlist_result(
+ entries, channel_id, self._og_search_title(webpage, fatal=False))
+
+ r"""
+ player = self._extract_player(webpage, channel_id)
+
+ channel_id = self._search_regex(
+ r'docId\s*:\s*(["\'])(?P<id>(?!\1).+?)\1', webpage,
+ 'channel id', group='id')
+
+ channel = self._call_api(
+ 'https://api.zdf.de/content/documents/%s.json' % channel_id,
+ player, url, channel_id)
+
+ items = []
+ for module in channel['module']:
+ for teaser in try_get(module, lambda x: x['teaser'], list) or []:
+ t = try_get(
+ teaser, lambda x: x['http://zdf.de/rels/target'], dict)
+ if not t:
+ continue
+ items.extend(try_get(
+ t,
+ lambda x: x['resultsWithVideo']['http://zdf.de/rels/search/results'],
+ list) or [])
+ items.extend(try_get(
+ module,
+ lambda x: x['filterRef']['resultsWithVideo']['http://zdf.de/rels/search/results'],
+ list) or [])
+
+ entries = []
+ entry_urls = set()
+ for item in items:
+ t = try_get(item, lambda x: x['http://zdf.de/rels/target'], dict)
+ if not t:
+ continue
+ sharing_url = t.get('http://zdf.de/rels/sharing-url')
+ if not sharing_url or not isinstance(sharing_url, compat_str):
+ continue
+ if sharing_url in entry_urls:
+ continue
+ entry_urls.add(sharing_url)
+ entries.append(self.url_result(
+ sharing_url, ie=ZDFIE.ie_key(), video_id=t.get('id')))
+
+ return self.playlist_result(entries, channel_id, channel.get('title'))
+ """
Rapsys Git / youtubedl / blobdiff