Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/tvnow.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     ExtractorError,
  10     int_or_none,
  11     parse_iso8601,
  12     parse_duration,
  13     try_get,
  14     update_url_query,
  15 )
  16
  17
  18 class TVNowBaseIE(InfoExtractor):
  19     _VIDEO_FIELDS = (
  20         'id', 'title', 'free', 'geoblocked', 'articleLong', 'articleShort',
  21         'broadcastStartDate', 'isDrm', 'duration', 'season', 'episode',
  22         'manifest.dashclear', 'manifest.hlsclear', 'manifest.smoothclear',
  23         'format.title', 'format.defaultImage169Format', 'format.defaultImage169Logo')
  24
  25     def _call_api(self, path, video_id, query):
  26         return self._download_json(
  27             'https://api.tvnow.de/v3/' + path,
  28             video_id, query=query)
  29
  30     def _extract_video(self, info, display_id):
  31         video_id = compat_str(info['id'])
  32         title = info['title']
  33
  34         paths = []
  35         for manifest_url in (info.get('manifest') or {}).values():
  36             if not manifest_url:
  37                 continue
  38             manifest_url = update_url_query(manifest_url, {'filter': ''})
  39             path = self._search_regex(r'https?://[^/]+/(.+?)\.ism/', manifest_url, 'path')
  40             if path in paths:
  41                 continue
  42             paths.append(path)
  43
  44             def url_repl(proto, suffix):
  45                 return re.sub(
  46                     r'(?:hls|dash|hss)([.-])', proto + r'\1', re.sub(
  47                         r'\.ism/(?:[^.]*\.(?:m3u8|mpd)|[Mm]anifest)',
  48                         '.ism/' + suffix, manifest_url))
  49
  50             formats = self._extract_mpd_formats(
  51                 url_repl('dash', '.mpd'), video_id,
  52                 mpd_id='dash', fatal=False)
  53             formats.extend(self._extract_ism_formats(
  54                 url_repl('hss', 'Manifest'),
  55                 video_id, ism_id='mss', fatal=False))
  56             formats.extend(self._extract_m3u8_formats(
  57                 url_repl('hls', '.m3u8'), video_id, 'mp4',
  58                 'm3u8_native', m3u8_id='hls', fatal=False))
  59             if formats:
  60                 break
  61         else:
  62             if info.get('isDrm'):
  63                 raise ExtractorError(
  64                     'Video %s is DRM protected' % video_id, expected=True)
  65             if info.get('geoblocked'):
  66                 raise self.raise_geo_restricted()
  67             if not info.get('free', True):
  68                 raise ExtractorError(
  69                     'Video %s is not available for free' % video_id, expected=True)
  70         self._sort_formats(formats)
  71
  72         description = info.get('articleLong') or info.get('articleShort')
  73         timestamp = parse_iso8601(info.get('broadcastStartDate'), ' ')
  74         duration = parse_duration(info.get('duration'))
  75
  76         f = info.get('format', {})
  77
  78         thumbnails = [{
  79             'url': 'https://aistvnow-a.akamaihd.net/tvnow/movie/%s' % video_id,
  80         }]
  81         thumbnail = f.get('defaultImage169Format') or f.get('defaultImage169Logo')
  82         if thumbnail:
  83             thumbnails.append({
  84                 'url': thumbnail,
  85             })
  86
  87         return {
  88             'id': video_id,
  89             'display_id': display_id,
  90             'title': title,
  91             'description': description,
  92             'thumbnails': thumbnails,
  93             'timestamp': timestamp,
  94             'duration': duration,
  95             'series': f.get('title'),
  96             'season_number': int_or_none(info.get('season')),
  97             'episode_number': int_or_none(info.get('episode')),
  98             'episode': title,
  99             'formats': formats,
 100         }
 101
 102
 103 class TVNowIE(TVNowBaseIE):
 104     _VALID_URL = r'''(?x)
 105                     https?://
 106                         (?:www\.)?tvnow\.(?:de|at|ch)/(?P<station>[^/]+)/
 107                         (?P<show_id>[^/]+)/
 108                         (?!(?:list|jahr)(?:/|$))(?P<id>[^/?\#&]+)
 109                     '''
 110
 111     _TESTS = [{
 112         'url': 'https://www.tvnow.de/rtl2/grip-das-motormagazin/der-neue-porsche-911-gt-3/player',
 113         'info_dict': {
 114             'id': '331082',
 115             'display_id': 'grip-das-motormagazin/der-neue-porsche-911-gt-3',
 116             'ext': 'mp4',
 117             'title': 'Der neue Porsche 911 GT 3',
 118             'description': 'md5:6143220c661f9b0aae73b245e5d898bb',
 119             'thumbnail': r're:^https?://.*\.jpg$',
 120             'timestamp': 1495994400,
 121             'upload_date': '20170528',
 122             'duration': 5283,
 123             'series': 'GRIP - Das Motormagazin',
 124             'season_number': 14,
 125             'episode_number': 405,
 126             'episode': 'Der neue Porsche 911 GT 3',
 127         },
 128     }, {
 129         # rtl2
 130         'url': 'https://www.tvnow.de/rtl2/armes-deutschland/episode-0008/player',
 131         'only_matching': True,
 132     }, {
 133         # rtlnitro
 134         'url': 'https://www.tvnow.de/nitro/alarm-fuer-cobra-11-die-autobahnpolizei/auf-eigene-faust-pilot/player',
 135         'only_matching': True,
 136     }, {
 137         # superrtl
 138         'url': 'https://www.tvnow.de/superrtl/die-lustigsten-schlamassel-der-welt/u-a-ketchup-effekt/player',
 139         'only_matching': True,
 140     }, {
 141         # ntv
 142         'url': 'https://www.tvnow.de/ntv/startup-news/goetter-in-weiss/player',
 143         'only_matching': True,
 144     }, {
 145         # vox
 146         'url': 'https://www.tvnow.de/vox/auto-mobil/neues-vom-automobilmarkt-2017-11-19-17-00-00/player',
 147         'only_matching': True,
 148     }, {
 149         # rtlplus
 150         'url': 'https://www.tvnow.de/rtlplus/op-ruft-dr-bruckner/die-vernaehte-frau/player',
 151         'only_matching': True,
 152     }, {
 153         'url': 'https://www.tvnow.de/rtl2/grip-das-motormagazin/der-neue-porsche-911-gt-3',
 154         'only_matching': True,
 155     }]
 156
 157     def _real_extract(self, url):
 158         mobj = re.match(self._VALID_URL, url)
 159         display_id = '%s/%s' % mobj.group(2, 3)
 160
 161         info = self._call_api(
 162             'movies/' + display_id, display_id, query={
 163                 'fields': ','.join(self._VIDEO_FIELDS),
 164                 'station': mobj.group(1),
 165             })
 166
 167         return self._extract_video(info, display_id)
 168
 169
 170 class TVNowListBaseIE(TVNowBaseIE):
 171     _SHOW_VALID_URL = r'''(?x)
 172                     (?P<base_url>
 173                         https?://
 174                             (?:www\.)?tvnow\.(?:de|at|ch)/[^/]+/
 175                             (?P<show_id>[^/]+)
 176                     )
 177                     '''
 178
 179     def _extract_list_info(self, display_id, show_id):
 180         fields = list(self._SHOW_FIELDS)
 181         fields.extend('formatTabs.%s' % field for field in self._SEASON_FIELDS)
 182         fields.extend(
 183             'formatTabs.formatTabPages.container.movies.%s' % field
 184             for field in self._VIDEO_FIELDS)
 185         return self._call_api(
 186             'formats/seo', display_id, query={
 187                 'fields': ','.join(fields),
 188                 'name': show_id + '.php'
 189             })
 190
 191
 192 class TVNowListIE(TVNowListBaseIE):
 193     _VALID_URL = r'%s/(?:list|jahr)/(?P<id>[^?\#&]+)' % TVNowListBaseIE._SHOW_VALID_URL
 194
 195     _SHOW_FIELDS = ('title', )
 196     _SEASON_FIELDS = ('id', 'headline', 'seoheadline', )
 197     _VIDEO_FIELDS = ('id', 'headline', 'seoUrl', )
 198
 199     _TESTS = [{
 200         'url': 'https://www.tvnow.de/rtl/30-minuten-deutschland/list/aktuell',
 201         'info_dict': {
 202             'id': '28296',
 203             'title': '30 Minuten Deutschland - Aktuell',
 204         },
 205         'playlist_mincount': 1,
 206     }, {
 207         'url': 'https://www.tvnow.de/vox/ab-ins-beet/list/staffel-14',
 208         'only_matching': True,
 209     }, {
 210         'url': 'https://www.tvnow.de/rtl2/grip-das-motormagazin/jahr/2018/3',
 211         'only_matching': True,
 212     }]
 213
 214     @classmethod
 215     def suitable(cls, url):
 216         return (False if TVNowIE.suitable(url)
 217                 else super(TVNowListIE, cls).suitable(url))
 218
 219     def _real_extract(self, url):
 220         base_url, show_id, season_id = re.match(self._VALID_URL, url).groups()
 221
 222         list_info = self._extract_list_info(season_id, show_id)
 223
 224         season = next(
 225             season for season in list_info['formatTabs']['items']
 226             if season.get('seoheadline') == season_id)
 227
 228         title = list_info.get('title')
 229         headline = season.get('headline')
 230         if title and headline:
 231             title = '%s - %s' % (title, headline)
 232         else:
 233             title = headline or title
 234
 235         entries = []
 236         for container in season['formatTabPages']['items']:
 237             items = try_get(
 238                 container, lambda x: x['container']['movies']['items'],
 239                 list) or []
 240             for info in items:
 241                 seo_url = info.get('seoUrl')
 242                 if not seo_url:
 243                     continue
 244                 video_id = info.get('id')
 245                 entries.append(self.url_result(
 246                     '%s/%s/player' % (base_url, seo_url), TVNowIE.ie_key(),
 247                     compat_str(video_id) if video_id else None))
 248
 249         return self.playlist_result(
 250             entries, compat_str(season.get('id') or season_id), title)
 251
 252
 253 class TVNowShowIE(TVNowListBaseIE):
 254     _VALID_URL = TVNowListBaseIE._SHOW_VALID_URL
 255
 256     _SHOW_FIELDS = ('id', 'title', )
 257     _SEASON_FIELDS = ('id', 'headline', 'seoheadline', )
 258     _VIDEO_FIELDS = ()
 259
 260     _TESTS = [{
 261         'url': 'https://www.tvnow.at/vox/ab-ins-beet',
 262         'info_dict': {
 263             'id': 'ab-ins-beet',
 264             'title': 'Ab ins Beet!',
 265         },
 266         'playlist_mincount': 7,
 267     }, {
 268         'url': 'https://www.tvnow.at/vox/ab-ins-beet/list',
 269         'only_matching': True,
 270     }, {
 271         'url': 'https://www.tvnow.de/rtl2/grip-das-motormagazin/jahr/',
 272         'only_matching': True,
 273     }]
 274
 275     @classmethod
 276     def suitable(cls, url):
 277         return (False if TVNowIE.suitable(url) or TVNowListIE.suitable(url)
 278                 else super(TVNowShowIE, cls).suitable(url))
 279
 280     def _real_extract(self, url):
 281         base_url, show_id = re.match(self._VALID_URL, url).groups()
 282
 283         list_info = self._extract_list_info(show_id, show_id)
 284
 285         entries = []
 286         for season_info in list_info['formatTabs']['items']:
 287             season_url = season_info.get('seoheadline')
 288             if not season_url:
 289                 continue
 290             season_id = season_info.get('id')
 291             entries.append(self.url_result(
 292                 '%s/list/%s' % (base_url, season_url), TVNowListIE.ie_key(),
 293                 compat_str(season_id) if season_id else None,
 294                 season_info.get('headline')))
 295
 296         return self.playlist_result(entries, show_id, list_info.get('title'))