yt-dlp/yt_dlp/extractor/zdf.py

401 lines
16 KiB
Python
Raw Normal View History

# coding: utf-8
from __future__ import unicode_literals
import re
from .common import InfoExtractor
from ..compat import compat_str
from ..utils import (
2015-12-28 17:49:59 +01:00
determine_ext,
2021-03-02 09:05:59 +01:00
float_or_none,
int_or_none,
2021-11-06 02:05:24 +01:00
join_nonempty,
2021-03-02 09:05:59 +01:00
merge_dicts,
NO_DEFAULT,
orderedSet,
parse_codecs,
qualities,
traverse_obj,
try_get,
unified_timestamp,
update_url_query,
2018-07-21 14:08:28 +02:00
url_or_none,
urljoin,
)
class ZDFBaseIE(InfoExtractor):
2019-09-18 20:11:52 +02:00
_GEO_COUNTRIES = ['DE']
2021-03-02 09:05:59 +01:00
_QUALITIES = ('auto', 'low', 'med', 'high', 'veryhigh', 'hd')
2021-03-02 09:05:59 +01:00
def _call_api(self, url, video_id, item, api_token=None, referrer=None):
headers = {}
if api_token:
headers['Api-Auth'] = 'Bearer %s' % api_token
if referrer:
headers['Referer'] = referrer
return self._download_json(
url, video_id, 'Downloading JSON %s' % item, headers=headers)
2015-12-28 17:49:59 +01:00
@staticmethod
def _extract_subtitles(src):
subtitles = {}
for caption in try_get(src, lambda x: x['captions'], list) or []:
2018-07-21 14:08:28 +02:00
subtitle_url = url_or_none(caption.get('uri'))
if subtitle_url:
lang = caption.get('language', 'deu')
subtitles.setdefault(lang, []).append({
'url': subtitle_url,
})
return subtitles
def _extract_format(self, video_id, formats, format_urls, meta):
2018-07-21 14:08:28 +02:00
format_url = url_or_none(meta.get('url'))
if not format_url or format_url in format_urls:
return
format_urls.add(format_url)
mime_type, ext = meta.get('mimeType'), determine_ext(format_url)
if mime_type == 'application/x-mpegURL' or ext == 'm3u8':
new_formats = 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':
new_formats = 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'))
if not f and meta.get('type'):
data = meta['type'].split('_')
if try_get(data, lambda x: x[2]) == ext:
f = {'vcodec': data[0], 'acodec': data[1]}
f.update({
'url': format_url,
2021-11-06 02:05:24 +01:00
'format_id': join_nonempty('http', meta.get('type'), meta.get('quality')),
})
new_formats = [f]
formats.extend(merge_dicts(f, {
2021-11-06 02:05:24 +01:00
'format_note': join_nonempty('quality', 'class', from_dict=meta, delim=', '),
'language': meta.get('language'),
'language_preference': 10 if meta.get('class') == 'main' else -10 if meta.get('class') == 'ad' else -1,
'quality': qualities(self._QUALITIES)(meta.get('quality')),
}) for f in new_formats)
2021-03-02 09:05:59 +01:00
def _extract_ptmd(self, ptmd_url, video_id, api_token, referrer):
2017-02-13 19:00:06 +01:00
ptmd = self._call_api(
2021-03-02 09:05:59 +01:00
ptmd_url, video_id, 'metadata', api_token, referrer)
content_id = ptmd.get('basename') or ptmd_url.split('/')[-1]
2015-12-28 17:49:59 +01:00
formats = []
track_uris = set()
for p in ptmd['priorityList']:
formitaeten = p.get('formitaeten')
if not isinstance(formitaeten, list):
2015-12-28 17:49:59 +01:00
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(
2021-03-02 09:05:59 +01:00
content_id, formats, track_uris, {
'url': track.get('uri'),
'type': f.get('type'),
'mimeType': f.get('mimeType'),
'quality': quality.get('quality'),
'class': track.get('class'),
'language': track.get('language'),
})
self._sort_formats(formats, ('hasaud', 'res', 'quality', 'language_preference'))
2021-03-02 09:05:59 +01:00
duration = float_or_none(try_get(
ptmd, lambda x: x['attributes']['duration']['value']), scale=1000)
return {
'extractor_key': ZDFIE.ie_key(),
'id': content_id,
'duration': duration,
'formats': formats,
'subtitles': self._extract_subtitles(ptmd),
}
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)
class ZDFIE(ZDFBaseIE):
_VALID_URL = r'https?://www\.zdf\.de/(?:[^/]+/)*(?P<id>[^/?#&]+)\.html'
_TESTS = [{
'url': 'https://www.zdf.de/nachrichten/heute-journal/heute-journal-vom-30-12-2021-100.html',
2021-03-02 09:05:59 +01:00
'info_dict': {
'id': '211230_sendung_hjo',
2021-03-02 09:05:59 +01:00
'ext': 'mp4',
'description': 'md5:47dff85977bde9fb8cba9e9c9b929839',
'duration': 1890.0,
'upload_date': '20211230',
'chapters': list,
'thumbnail': 'md5:e65f459f741be5455c952cd820eb188e',
'title': 'heute journal vom 30.12.2021',
'timestamp': 1640897100,
}
2021-03-02 09:05:59 +01:00
}, {
'url': 'https://www.zdf.de/dokumentation/terra-x/die-magie-der-farben-von-koenigspurpur-und-jeansblau-100.html',
'info_dict': {
'id': '151025_magie_farben2_tex',
'ext': 'mp4',
'title': 'Die Magie der Farben (2/2)',
'description': 'md5:a89da10c928c6235401066b60a6d5c1a',
'duration': 2615,
'timestamp': 1465021200,
'upload_date': '20160604',
'thumbnail': 'https://www.zdf.de/assets/mauve-im-labor-100~768x432?cb=1464909117806',
2021-03-02 09:05:59 +01:00
},
}, {
'url': 'https://www.zdf.de/funk/druck-11790/funk-alles-ist-verzaubert-102.html',
'md5': '3d6f1049e9682178a11c54b91f3dd065',
'info_dict': {
'ext': 'mp4',
'id': 'video_funk_1770473',
'duration': 1278,
'description': 'Die Neue an der Schule verdreht Ismail den Kopf.',
'title': 'Alles ist verzaubert',
'timestamp': 1635520560,
'upload_date': '20211029',
'thumbnail': 'https://www.zdf.de/assets/teaser-funk-alles-ist-verzaubert-100~1920x1080?cb=1636466431799',
},
2021-03-02 09:05:59 +01:00
}, {
# Same as https://www.phoenix.de/sendungen/dokumentationen/gesten-der-maechtigen-i-a-89468.html?ref=suche
'url': 'https://www.zdf.de/politik/phoenix-sendungen/die-gesten-der-maechtigen-100.html',
'only_matching': True,
}, {
# Same as https://www.3sat.de/film/spielfilm/der-hauptmann-100.html
'url': 'https://www.zdf.de/filme/filme-sonstige/der-hauptmann-112.html',
'only_matching': True,
}, {
# Same as https://www.3sat.de/wissen/nano/nano-21-mai-2019-102.html, equal media ids
'url': 'https://www.zdf.de/wissen/nano/nano-21-mai-2019-102.html',
'only_matching': True,
}, {
'url': 'https://www.zdf.de/service-und-hilfe/die-neue-zdf-mediathek/zdfmediathek-trailer-100.html',
'only_matching': True,
}, {
'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,
}, {
# Same as https://www.phoenix.de/sendungen/ereignisse/corona-nachgehakt/wohin-fuehrt-der-protest-in-der-pandemie-a-2050630.html
'url': 'https://www.zdf.de/politik/phoenix-sendungen/wohin-fuehrt-der-protest-in-der-pandemie-100.html',
'only_matching': True
}, {
# Same as https://www.3sat.de/film/ab-18/10-wochen-sommer-108.html
'url': 'https://www.zdf.de/dokumentation/ab-18/10-wochen-sommer-102.html',
'only_matching': True
2021-03-02 09:05:59 +01:00
}]
def _extract_entry(self, url, player, 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 = traverse_obj(
t, ('streams', 'default', 'http://zdf.de/rels/streams/ptmd-template'),
'http://zdf.de/rels/streams/ptmd-template').replace(
2021-03-02 09:05:59 +01:00
'{playerId}', 'ngplayer_2_4')
info = self._extract_ptmd(
urljoin(url, ptmd_path), video_id, player['apiToken'], url)
thumbnails = []
layouts = try_get(
content, lambda x: x['teaserImageRef']['layouts'], dict)
if layouts:
for layout_key, layout_url in layouts.items():
2018-07-21 14:08:28 +02:00
layout_url = url_or_none(layout_url)
if not layout_url:
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)
2015-12-28 17:49:59 +01:00
chapter_marks = t.get('streamAnchorTag') or []
chapter_marks.append({'anchorOffset': int_or_none(t.get('duration'))})
chapters = [{
'start_time': chap.get('anchorOffset'),
'end_time': next_chap.get('anchorOffset'),
'title': chap.get('anchorLabel')
} for chap, next_chap in zip(chapter_marks, chapter_marks[1:])]
2021-03-02 09:05:59 +01:00
return merge_dicts(info, {
'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,
'chapters': chapters or None
2021-03-02 09:05:59 +01:00
})
def _extract_regular(self, url, player, video_id):
2017-02-13 19:00:06 +01:00
content = self._call_api(
2021-03-02 09:05:59 +01:00
player['content'], video_id, 'content', player['apiToken'], url)
2017-02-13 19:00:06 +01:00
return self._extract_entry(player['content'], player, content, video_id)
2016-10-29 14:06:07 +02:00
def _extract_mobile(self, video_id):
2021-03-02 09:05:59 +01:00
video = self._download_json(
'https://zdf-cdn.live.cellular.de/mediathekV2/document/%s' % video_id,
2021-03-02 09:05:59 +01:00
video_id)
document = video['document']
2016-10-29 14:06:07 +02:00
title = document['titel']
2021-03-02 09:05:59 +01:00
content_id = document['basename']
2016-10-29 14:06:07 +02:00
formats = []
format_urls = set()
for f in document['formitaeten']:
2021-03-02 09:05:59 +01:00
self._extract_format(content_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')),
})
2016-10-29 14:06:07 +02:00
return {
2021-03-02 09:05:59 +01:00
'id': content_id,
'title': title,
'description': document.get('beschreibung'),
'duration': int_or_none(document.get('length')),
2021-03-02 09:05:59 +01:00
'timestamp': unified_timestamp(document.get('date')) or unified_timestamp(
try_get(video, lambda x: x['meta']['editorialDate'], compat_str)),
'thumbnails': thumbnails,
'subtitles': self._extract_subtitles(document),
'formats': formats,
}
2016-10-29 14:06:07 +02:00
def _real_extract(self, url):
video_id = self._match_id(url)
2016-10-29 14:06:07 +02:00
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)
2016-10-29 14:06:07 +02:00
return self._extract_mobile(video_id)
2016-10-29 14:06:07 +02:00
class ZDFChannelIE(ZDFBaseIE):
_VALID_URL = r'https?://www\.zdf\.de/(?:[^/]+/)*(?P<id>[^/?#&]+)'
2015-12-24 15:44:49 +01:00
_TESTS = [{
'url': 'https://www.zdf.de/sport/das-aktuelle-sportstudio',
2014-12-10 17:29:03 +01:00
'info_dict': {
'id': 'das-aktuelle-sportstudio',
'title': 'das aktuelle sportstudio | ZDF',
2014-12-10 17:29:03 +01:00
},
2020-02-29 17:06:36 +01:00
'playlist_mincount': 23,
2015-12-24 15:44:49 +01:00
}, {
'url': 'https://www.zdf.de/dokumentation/planet-e',
'info_dict': {
'id': 'planet-e',
'title': 'planet e.',
},
2020-02-29 17:06:36 +01:00
'playlist_mincount': 50,
2015-12-24 15:44:49 +01:00
}, {
'url': 'https://www.zdf.de/filme/taunuskrimi/',
2015-12-24 15:44:49 +01:00
'only_matching': True,
}]
@classmethod
def suitable(cls, url):
return False if ZDFIE.suitable(url) else super(ZDFChannelIE, cls).suitable(url)
2014-12-11 17:07:59 +01:00
def _real_extract(self, url):
channel_id = self._match_id(url)
2014-12-10 17:29:03 +01:00
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'))
"""