logo

youtube-dl

[mirror] Download/Watch videos from video hostersgit clone https://hacktivis.me/git/mirror/youtube-dl.git

zdf.py (17212B)

    

  1. # coding: utf-8
  2. from __future__ import unicode_literals
  4. import re
  6. from .common import InfoExtractor
  7. from ..compat import compat_str
  8. from ..utils import (
  9.     determine_ext,
  10.     ExtractorError,
  11.     extract_attributes,
  12.     float_or_none,
  13.     int_or_none,
  14.     merge_dicts,
  15.     NO_DEFAULT,
  16.     parse_codecs,
  17.     qualities,
  18.     str_or_none,
  19.     try_get,
  20.     unified_timestamp,
  21.     update_url_query,
  22.     url_or_none,
  23.     urljoin,
  24. )
  27. class ZDFBaseIE(InfoExtractor):
  28.     _GEO_COUNTRIES = ['DE']
  29.     _QUALITIES = ('auto', 'low', 'med', 'high', 'veryhigh', 'hd')
  31.     def _call_api(self, url, video_id, item, api_token=None, referrer=None):
  32.         headers = {}
  33.         if api_token:
  34.             headers['Api-Auth'] = 'Bearer %s' % api_token
  35.         if referrer:
  36.             headers['Referer'] = referrer
  37.         return self._download_json(
  38.             url, video_id, 'Downloading JSON %s' % item, headers=headers)
  40.     @staticmethod
  41.     def _extract_subtitles(src):
  42.         subtitles = {}
  43.         for caption in try_get(src, lambda x: x['captions'], list) or []:
  44.             subtitle_url = url_or_none(caption.get('uri'))
  45.             if subtitle_url:
  46.                 lang = caption.get('language', 'deu')
  47.                 subtitles.setdefault(lang, []).append({
  48.                     'url': subtitle_url,
  49.                 })
  50.         return subtitles
  52.     def _extract_format(self, video_id, formats, format_urls, meta):
  53.         format_url = url_or_none(meta.get('url'))
  54.         if not format_url:
  55.             return
  56.         if format_url in format_urls:
  57.             return
  58.         format_urls.add(format_url)
  59.         mime_type = meta.get('mimeType')
  60.         ext = determine_ext(format_url)
  62.         join_nonempty = lambda s, l: s.join(filter(None, l))
  63.         meta_map = lambda t: map(lambda x: str_or_none(meta.get(x)), t)
  65.         if mime_type == 'application/x-mpegURL' or ext == 'm3u8':
  66.             new_formats = self._extract_m3u8_formats(
  67.                 format_url, video_id, 'mp4', m3u8_id='hls',
  68.                 entry_protocol='m3u8_native', fatal=False)
  69.         elif mime_type == 'application/f4m+xml' or ext == 'f4m':
  70.             new_formats = self._extract_f4m_formats(
  71.                 update_url_query(format_url, {'hdcore': '3.7.0'}), video_id, f4m_id='hds', fatal=False)
  72.         else:
  73.             f = parse_codecs(meta.get('mimeCodec'))
  74.             if not f:
  75.                 data = meta.get('type', '').split('_')
  76.                 if try_get(data, lambda x: x[2]) == ext:
  77.                     f = dict(zip(('vcodec', 'acodec'), data[1]))
  79.             format_id = ['http']
  80.             format_id.extend(join_nonempty('-', meta_map(('type', 'quality'))))
  81.             f.update({
  82.                 'url': format_url,
  83.                 'format_id': '-'.join(format_id),
  84.                 'tbr': int_or_none(self._search_regex(r'_(\d+)k_', format_url, 'tbr', default=None))
  85.             })
  86.             new_formats = [f]
  88.         formats.extend(merge_dicts(f, {
  89.             'format_note': join_nonempty(',', meta_map(('quality', 'class'))),
  90.             'language': meta.get('language'),
  91.             'language_preference': 10 if meta.get('class') == 'main' else -10 if meta.get('class') == 'ad' else -1,
  92.             'quality': qualities(self._QUALITIES)(meta.get('quality')),
  93.         }) for f in new_formats)
  95.     def _extract_ptmd(self, ptmd_url, video_id, api_token, referrer):
  96.         ptmd = self._call_api(
  97.             ptmd_url, video_id, 'metadata', api_token, referrer)
  99.         content_id = ptmd.get('basename') or ptmd_url.split('/')[-1]
  101.         formats = []
  102.         track_uris = set()
  103.         for p in ptmd['priorityList']:
  104.             formitaeten = p.get('formitaeten')
  105.             if not isinstance(formitaeten, list):
  106.                 continue
  107.             for f in formitaeten:
  108.                 f_qualities = f.get('qualities')
  109.                 if not isinstance(f_qualities, list):
  110.                     continue
  111.                 for quality in f_qualities:
  112.                     tracks = try_get(quality, lambda x: x['audio']['tracks'], list)
  113.                     if not tracks:
  114.                         continue
  115.                     for track in tracks:
  116.                         self._extract_format(
  117.                             content_id, formats, track_uris, {
  118.                                 'url': track.get('uri'),
  119.                                 'type': f.get('type'),
  120.                                 'mimeType': f.get('mimeType'),
  121.                                 'quality': quality.get('quality'),
  122.                                 'class': track.get('class'),
  123.                                 'language': track.get('language'),
  124.                             })
  125.         self._sort_formats(formats)
  127.         duration = float_or_none(try_get(
  128.             ptmd, lambda x: x['attributes']['duration']['value']), scale=1000)
  130.         return {
  131.             'extractor_key': ZDFIE.ie_key(),
  132.             'id': content_id,
  133.             'duration': duration,
  134.             'formats': formats,
  135.             'subtitles': self._extract_subtitles(ptmd),
  136.         }
  138.     def _extract_player(self, webpage, video_id, fatal=True):
  139.         return self._parse_json(
  140.             self._search_regex(
  141.                 r'(?s)data-zdfplayer-jsb=(["\'])(?P<json>{.+?})\1', webpage,
  142.                 'player JSON', default='{}' if not fatal else NO_DEFAULT,
  143.                 group='json'),
  144.             video_id)
  147. class ZDFIE(ZDFBaseIE):
  148.     _VALID_URL = r'https?://www\.zdf\.de/(?:[^/]+/)*(?P<id>[^/?#&]+)\.html'
  149.     _TESTS = [{
  150.         # Same as https://www.phoenix.de/sendungen/ereignisse/corona-nachgehakt/wohin-fuehrt-der-protest-in-der-pandemie-a-2050630.html
  151.         'url': 'https://www.zdf.de/politik/phoenix-sendungen/wohin-fuehrt-der-protest-in-der-pandemie-100.html',
  152.         'md5': '34ec321e7eb34231fd88616c65c92db0',
  153.         'info_dict': {
  154.             'id': '210222_phx_nachgehakt_corona_protest',
  155.             'ext': 'mp4',
  156.             'title': 'Wohin führt der Protest in der Pandemie?',
  157.             'description': 'md5:7d643fe7f565e53a24aac036b2122fbd',
  158.             'duration': 1691,
  159.             'timestamp': 1613948400,
  160.             'upload_date': '20210221',
  161.         },
  162.         'skip': 'No longer available: "Diese Seite wurde leider nicht gefunden"',
  163.     }, {
  164.         # Same as https://www.3sat.de/film/ab-18/10-wochen-sommer-108.html
  165.         'url': 'https://www.zdf.de/dokumentation/ab-18/10-wochen-sommer-102.html',
  166.         'md5': '0aff3e7bc72c8813f5e0fae333316a1d',
  167.         'info_dict': {
  168.             'id': '141007_ab18_10wochensommer_film',
  169.             'ext': 'mp4',
  170.             'title': 'Ab 18! - 10 Wochen Sommer',
  171.             'description': 'md5:8253f41dc99ce2c3ff892dac2d65fe26',
  172.             'duration': 2660,
  173.             'timestamp': 1608604200,
  174.             'upload_date': '20201222',
  175.         },
  176.         'skip': 'No longer available: "Diese Seite wurde leider nicht gefunden"',
  177.     }, {
  178.         'url': 'https://www.zdf.de/dokumentation/terra-x/die-magie-der-farben-von-koenigspurpur-und-jeansblau-100.html',
  179.         'info_dict': {
  180.             'id': '151025_magie_farben2_tex',
  181.             'ext': 'mp4',
  182.             'title': 'Die Magie der Farben (2/2)',
  183.             'description': 'md5:a89da10c928c6235401066b60a6d5c1a',
  184.             'duration': 2615,
  185.             'timestamp': 1465021200,
  186.             'upload_date': '20160604',
  187.             'thumbnail': 'https://www.zdf.de/assets/mauve-im-labor-100~768x432?cb=1464909117806',
  188.         },
  189.     }, {
  190.         'url': 'https://www.zdf.de/funk/druck-11790/funk-alles-ist-verzaubert-102.html',
  191.         'md5': '1b93bdec7d02fc0b703c5e7687461628',
  192.         'info_dict': {
  193.             'ext': 'mp4',
  194.             'id': 'video_funk_1770473',
  195.             'duration': 1278,
  196.             'description': 'Die Neue an der Schule verdreht Ismail den Kopf.',
  197.             'title': 'Alles ist verzaubert',
  198.             'timestamp': 1635520560,
  199.             'upload_date': '20211029',
  200.             'thumbnail': 'https://www.zdf.de/assets/teaser-funk-alles-ist-verzaubert-100~1920x1080?cb=1636466431799',
  201.         },
  202.     }, {
  203.         # Same as https://www.phoenix.de/sendungen/dokumentationen/gesten-der-maechtigen-i-a-89468.html?ref=suche
  204.         'url': 'https://www.zdf.de/politik/phoenix-sendungen/die-gesten-der-maechtigen-100.html',
  205.         'only_matching': True,
  206.     }, {
  207.         # Same as https://www.3sat.de/film/spielfilm/der-hauptmann-100.html
  208.         'url': 'https://www.zdf.de/filme/filme-sonstige/der-hauptmann-112.html',
  209.         'only_matching': True,
  210.     }, {
  211.         # Same as https://www.3sat.de/wissen/nano/nano-21-mai-2019-102.html, equal media ids
  212.         'url': 'https://www.zdf.de/wissen/nano/nano-21-mai-2019-102.html',
  213.         'only_matching': True,
  214.     }, {
  215.         'url': 'https://www.zdf.de/service-und-hilfe/die-neue-zdf-mediathek/zdfmediathek-trailer-100.html',
  216.         'only_matching': True,
  217.     }, {
  218.         'url': 'https://www.zdf.de/filme/taunuskrimi/die-lebenden-und-die-toten-1---ein-taunuskrimi-100.html',
  219.         'only_matching': True,
  220.     }, {
  221.         'url': 'https://www.zdf.de/dokumentation/planet-e/planet-e-uebersichtsseite-weitere-dokumentationen-von-planet-e-100.html',
  222.         'only_matching': True,
  223.     }, {
  224.         'url': 'https://www.zdf.de/arte/todliche-flucht/page-video-artede-toedliche-flucht-16-100.html',
  225.         'info_dict': {
  226.             'id': 'video_artede_083871-001-A',
  227.             'ext': 'mp4',
  228.             'title': 'Tödliche Flucht (1/6)',
  229.             'description': 'md5:e34f96a9a5f8abd839ccfcebad3d5315',
  230.             'duration': 3193.0,
  231.             'timestamp': 1641355200,
  232.             'upload_date': '20220105',
  233.         },
  234.         'skip': 'No longer available "Diese Seite wurde leider nicht gefunden"'
  235.     }, {
  236.         'url': 'https://www.zdf.de/serien/soko-stuttgart/das-geld-anderer-leute-100.html',
  237.         'info_dict': {
  238.             'id': '191205_1800_sendung_sok8',
  239.             'ext': 'mp4',
  240.             'title': 'Das Geld anderer Leute',
  241.             'description': 'md5:cb6f660850dc5eb7d1ab776ea094959d',
  242.             'duration': 2581.0,
  243.             'timestamp': 1654790700,
  244.             'upload_date': '20220609',
  245.             'thumbnail': 'https://epg-image.zdf.de/fotobase-webdelivery/images/e2d7e55a-09f0-424e-ac73-6cac4dd65f35?layout=2400x1350',
  246.         },
  247.     }]
  249.     def _extract_entry(self, url, player, content, video_id):
  250.         title = content.get('title') or content['teaserHeadline']
  252.         t = content['mainVideoContent']['http://zdf.de/rels/target']
  254.         def get_ptmd_path(d):
  255.             return (
  256.                 d.get('http://zdf.de/rels/streams/ptmd')
  257.                 or d.get('http://zdf.de/rels/streams/ptmd-template',
  258.                          '').replace('{playerId}', 'ngplayer_2_4'))
  260.         ptmd_path = get_ptmd_path(try_get(t, lambda x: x['streams']['default'], dict) or {})
  261.         if not ptmd_path:
  262.             ptmd_path = get_ptmd_path(t)
  264.         if not ptmd_path:
  265.             raise ExtractorError('Could not extract ptmd_path')
  267.         info = self._extract_ptmd(
  268.             urljoin(url, ptmd_path), video_id, player['apiToken'], url)
  270.         thumbnails = []
  271.         layouts = try_get(
  272.             content, lambda x: x['teaserImageRef']['layouts'], dict)
  273.         if layouts:
  274.             for layout_key, layout_url in layouts.items():
  275.                 layout_url = url_or_none(layout_url)
  276.                 if not layout_url:
  277.                     continue
  278.                 thumbnail = {
  279.                     'url': layout_url,
  280.                     'format_id': layout_key,
  281.                 }
  282.                 mobj = re.search(r'(?P<width>\d+)x(?P<height>\d+)', layout_key)
  283.                 if mobj:
  284.                     thumbnail.update({
  285.                         'width': int(mobj.group('width')),
  286.                         'height': int(mobj.group('height')),
  287.                     })
  288.                 thumbnails.append(thumbnail)
  290.         return merge_dicts(info, {
  291.             'title': title,
  292.             'description': content.get('leadParagraph') or content.get('teasertext'),
  293.             'duration': int_or_none(t.get('duration')),
  294.             'timestamp': unified_timestamp(content.get('editorialDate')),
  295.             'thumbnails': thumbnails,
  296.         })
  298.     def _extract_regular(self, url, player, video_id):
  299.         content = self._call_api(
  300.             player['content'], video_id, 'content', player['apiToken'], url)
  301.         return self._extract_entry(player['content'], player, content, video_id)
  303.     def _extract_mobile(self, video_id):
  304.         video = self._download_json(
  305.             'https://zdf-cdn.live.cellular.de/mediathekV2/document/%s' % video_id,
  306.             video_id)
  308.         formats = []
  309.         formitaeten = try_get(video, lambda x: x['document']['formitaeten'], list)
  310.         document = formitaeten and video['document']
  311.         if formitaeten:
  312.             title = document['titel']
  313.             content_id = document['basename']
  315.             format_urls = set()
  316.             for f in formitaeten or []:
  317.                 self._extract_format(content_id, formats, format_urls, f)
  318.         self._sort_formats(formats)
  320.         thumbnails = []
  321.         teaser_bild = document.get('teaserBild')
  322.         if isinstance(teaser_bild, dict):
  323.             for thumbnail_key, thumbnail in teaser_bild.items():
  324.                 thumbnail_url = try_get(
  325.                     thumbnail, lambda x: x['url'], compat_str)
  326.                 if thumbnail_url:
  327.                     thumbnails.append({
  328.                         'url': thumbnail_url,
  329.                         'id': thumbnail_key,
  330.                         'width': int_or_none(thumbnail.get('width')),
  331.                         'height': int_or_none(thumbnail.get('height')),
  332.                     })
  334.         return {
  335.             'id': content_id,
  336.             'title': title,
  337.             'description': document.get('beschreibung'),
  338.             'duration': int_or_none(document.get('length')),
  339.             'timestamp': unified_timestamp(document.get('date')) or unified_timestamp(
  340.                 try_get(video, lambda x: x['meta']['editorialDate'], compat_str)),
  341.             'thumbnails': thumbnails,
  342.             'subtitles': self._extract_subtitles(document),
  343.             'formats': formats,
  344.         }
  346.     def _real_extract(self, url):
  347.         video_id = self._match_id(url)
  349.         webpage = self._download_webpage(url, video_id, fatal=False)
  350.         if webpage:
  351.             player = self._extract_player(webpage, url, fatal=False)
  352.             if player:
  353.                 return self._extract_regular(url, player, video_id)
  355.         return self._extract_mobile(video_id)
  358. class ZDFChannelIE(ZDFBaseIE):
  359.     _VALID_URL = r'https?://www\.zdf\.de/(?:[^/]+/)*(?P<id>[^/?#&]+)'
  360.     _TESTS = [{
  361.         'url': 'https://www.zdf.de/sport/das-aktuelle-sportstudio',
  362.         'info_dict': {
  363.             'id': 'das-aktuelle-sportstudio',
  364.             'title': 'das aktuelle sportstudio',
  365.         },
  366.         'playlist_mincount': 18,
  367.     }, {
  368.         'url': 'https://www.zdf.de/dokumentation/planet-e',
  369.         'info_dict': {
  370.             'id': 'planet-e',
  371.             'title': 'planet e.',
  372.         },
  373.         'playlist_mincount': 50,
  374.     }, {
  375.         'url': 'https://www.zdf.de/gesellschaft/aktenzeichen-xy-ungeloest',
  376.         'info_dict': {
  377.             'id': 'aktenzeichen-xy-ungeloest',
  378.             'title': 'Aktenzeichen XY... ungelöst',
  379.             'entries': "lambda x: not any('xy580-fall1-kindermoerder-gesucht-100' in e['url'] for e in x)",
  380.         },
  381.         'playlist_mincount': 2,
  382.     }, {
  383.         'url': 'https://www.zdf.de/filme/taunuskrimi/',
  384.         'only_matching': True,
  385.     }]
  387.     @classmethod
  388.     def suitable(cls, url):
  389.         return False if ZDFIE.suitable(url) else super(ZDFChannelIE, cls).suitable(url)
  391.     def _og_search_title(self, webpage, fatal=False):
  392.         title = super(ZDFChannelIE, self)._og_search_title(webpage, fatal=fatal)
  393.         return re.split(r'\s+[-|]\s+ZDF(?:mediathek)?$', title or '')[0] or None
  395.     def _real_extract(self, url):
  396.         channel_id = self._match_id(url)
  398.         webpage = self._download_webpage(url, channel_id)
  400.         matches = re.finditer(
  401.             r'''<div\b[^>]*?\sdata-plusbar-id\s*=\s*(["'])(?P<p_id>[\w-]+)\1[^>]*?\sdata-plusbar-url=\1(?P<url>%s)\1''' % ZDFIE._VALID_URL,
  402.             webpage)
  404.         if self._downloader.params.get('noplaylist', False):
  405.             entry = next(
  406.                 (self.url_result(m.group('url'), ie=ZDFIE.ie_key()) for m in matches),
  407.                 None)
  408.             self.to_screen('Downloading just the main video because of --no-playlist')
  409.             if entry:
  410.                 return entry
  411.         else:
  412.             self.to_screen('Downloading playlist %s - add --no-playlist to download just the main video' % (channel_id, ))
  414.         def check_video(m):
  415.             v_ref = self._search_regex(
  416.                 r'''(<a\b[^>]*?\shref\s*=[^>]+?\sdata-target-id\s*=\s*(["'])%s\2[^>]*>)''' % (m.group('p_id'), ),
  417.                 webpage, 'check id', default='')
  418.             v_ref = extract_attributes(v_ref)
  419.             return v_ref.get('data-target-video-type') != 'novideo'
  421.         return self.playlist_from_matches(
  422.             (m.group('url') for m in matches if check_video(m)),
  423.             channel_id, self._og_search_title(webpage, fatal=False))