yt_dlp/extractor/nova.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     clean_html,
   9     determine_ext,
  10     int_or_none,
  11     js_to_json,
  12     qualities,
  13     traverse_obj,
  14     unified_strdate,
  15     url_or_none,
  16 )
  17
  18
  19 class NovaEmbedIE(InfoExtractor):
  20     _VALID_URL = r'https?://media\.cms\.nova\.cz/embed/(?P<id>[^/?#&]+)'
  21     _TESTS = [{
  22         'url': 'https://media.cms.nova.cz/embed/8o0n0r?autoplay=1',
  23         'info_dict': {
  24             'id': '8o0n0r',
  25             'title': '2180. díl',
  26             'thumbnail': r're:^https?://.*\.jpg',
  27             'duration': 2578,
  28         },
  29         'params': {
  30             'skip_download': True,
  31             'ignore_no_formats_error': True,
  32         },
  33         'expected_warnings': ['DRM protected', 'Requested format is not available'],
  34     }, {
  35         'url': 'https://media.cms.nova.cz/embed/KybpWYvcgOa',
  36         'info_dict': {
  37             'id': 'KybpWYvcgOa',
  38             'ext': 'mp4',
  39             'title': 'Borhyová oslavila 60? Soutěžící z pořadu odboural moderátora Ondřeje Sokola',
  40             'thumbnail': r're:^https?://.*\.jpg',
  41             'duration': 114,
  42         },
  43         'params': {'skip_download': 'm3u8'},
  44     }]
  45
  46     def _real_extract(self, url):
  47         video_id = self._match_id(url)
  48
  49         webpage = self._download_webpage(url, video_id)
  50
  51         has_drm = False
  52         duration = None
  53         formats = []
  54
  55         player = self._parse_json(
  56             self._search_regex(
  57                 (r'(?:(?:replacePlaceholders|processAdTagModifier).*?:\s*)?(?:replacePlaceholders|processAdTagModifier)\s*\(\s*(?P<json>{.*?})\s*\)(?:\s*\))?\s*,',
  58                     r'Player\.init\s*\([^,]+,(?P<cndn>\s*\w+\s*\?)?\s*(?P<json>{(?(cndn).+?|.+)})\s*(?(cndn):|,\s*{.+?}\s*\)\s*;)'),
  59                 webpage, 'player', default='{}', group='json'), video_id, fatal=False)
  60         if player:
  61             for format_id, format_list in player['tracks'].items():
  62                 if not isinstance(format_list, list):
  63                     format_list = [format_list]
  64                 for format_dict in format_list:
  65                     if not isinstance(format_dict, dict):
  66                         continue
  67                     if (not self.get_param('allow_unplayable_formats')
  68                             and traverse_obj(format_dict, ('drm', 'keySystem'))):
  69                         has_drm = True
  70                         continue
  71                     format_url = url_or_none(format_dict.get('src'))
  72                     format_type = format_dict.get('type')
  73                     ext = determine_ext(format_url)
  74                     if (format_type == 'application/x-mpegURL'
  75                             or format_id == 'HLS' or ext == 'm3u8'):
  76                         formats.extend(self._extract_m3u8_formats(
  77                             format_url, video_id, 'mp4',
  78                             entry_protocol='m3u8_native', m3u8_id='hls',
  79                             fatal=False))
  80                     elif (format_type == 'application/dash+xml'
  81                           or format_id == 'DASH' or ext == 'mpd'):
  82                         formats.extend(self._extract_mpd_formats(
  83                             format_url, video_id, mpd_id='dash', fatal=False))
  84                     else:
  85                         formats.append({
  86                             'url': format_url,
  87                         })
  88             duration = int_or_none(player.get('duration'))
  89         else:
  90             # Old path, not actual as of 08.04.2020
  91             bitrates = self._parse_json(
  92                 self._search_regex(
  93                     r'(?s)(?:src|bitrates)\s*=\s*({.+?})\s*;', webpage, 'formats'),
  94                 video_id, transform_source=js_to_json)
  95
  96             QUALITIES = ('lq', 'mq', 'hq', 'hd')
  97             quality_key = qualities(QUALITIES)
  98
  99             for format_id, format_list in bitrates.items():
 100                 if not isinstance(format_list, list):
 101                     format_list = [format_list]
 102                 for format_url in format_list:
 103                     format_url = url_or_none(format_url)
 104                     if not format_url:
 105                         continue
 106                     if format_id == 'hls':
 107                         formats.extend(self._extract_m3u8_formats(
 108                             format_url, video_id, ext='mp4',
 109                             entry_protocol='m3u8_native', m3u8_id='hls',
 110                             fatal=False))
 111                         continue
 112                     f = {
 113                         'url': format_url,
 114                     }
 115                     f_id = format_id
 116                     for quality in QUALITIES:
 117                         if '%s.mp4' % quality in format_url:
 118                             f_id += '-%s' % quality
 119                             f.update({
 120                                 'quality': quality_key(quality),
 121                                 'format_note': quality.upper(),
 122                             })
 123                             break
 124                     f['format_id'] = f_id
 125                     formats.append(f)
 126
 127         if not formats and has_drm:
 128             self.report_drm(video_id)
 129         self._sort_formats(formats)
 130
 131         title = self._og_search_title(
 132             webpage, default=None) or self._search_regex(
 133             (r'<value>(?P<title>[^<]+)',
 134              r'videoTitle\s*:\s*(["\'])(?P<value>(?:(?!\1).)+)\1'), webpage,
 135             'title', group='value')
 136         thumbnail = self._og_search_thumbnail(
 137             webpage, default=None) or self._search_regex(
 138             r'poster\s*:\s*(["\'])(?P<value>(?:(?!\1).)+)\1', webpage,
 139             'thumbnail', fatal=False, group='value')
 140         duration = int_or_none(self._search_regex(
 141             r'videoDuration\s*:\s*(\d+)', webpage, 'duration',
 142             default=duration))
 143
 144         return {
 145             'id': video_id,
 146             'title': title,
 147             'thumbnail': thumbnail,
 148             'duration': duration,
 149             'formats': formats,
 150         }
 151
 152
 153 class NovaIE(InfoExtractor):
 154     IE_DESC = 'TN.cz, Prásk.tv, Nova.cz, Novaplus.cz, FANDA.tv, Krásná.cz and Doma.cz'
 155     _VALID_URL = r'https?://(?:[^.]+\.)?(?P<site>tv(?:noviny)?|tn|novaplus|vymena|fanda|krasna|doma|prask)\.nova\.cz/(?:[^/]+/)+(?P<id>[^/]+?)(?:\.html|/|$)'
 156     _TESTS = [{
 157         'url': 'http://tn.nova.cz/clanek/tajemstvi-ukryte-v-podzemi-specialni-nemocnice-v-prazske-krci.html#player_13260',
 158         'md5': '249baab7d0104e186e78b0899c7d5f28',
 159         'info_dict': {
 160             'id': '1757139',
 161             'display_id': 'tajemstvi-ukryte-v-podzemi-specialni-nemocnice-v-prazske-krci',
 162             'ext': 'mp4',
 163             'title': 'Podzemní nemocnice v pražské Krči',
 164             'description': 'md5:f0a42dd239c26f61c28f19e62d20ef53',
 165             'thumbnail': r're:^https?://.*\.(?:jpg)',
 166         }
 167     }, {
 168         'url': 'http://fanda.nova.cz/clanek/fun-and-games/krvavy-epos-zaklinac-3-divoky-hon-vychazi-vyhrajte-ho-pro-sebe.html',
 169         'info_dict': {
 170             'id': '1753621',
 171             'ext': 'mp4',
 172             'title': 'Zaklínač 3: Divoký hon',
 173             'description': 're:.*Pokud se stejně jako my nemůžete.*',
 174             'thumbnail': r're:https?://.*\.jpg(\?.*)?',
 175             'upload_date': '20150521',
 176         },
 177         'params': {
 178             # rtmp download
 179             'skip_download': True,
 180         },
 181         'skip': 'gone',
 182     }, {
 183         # media.cms.nova.cz embed
 184         'url': 'https://novaplus.nova.cz/porad/ulice/epizoda/18760-2180-dil',
 185         'info_dict': {
 186             'id': '8o0n0r',
 187             'ext': 'mp4',
 188             'title': '2180. díl',
 189             'thumbnail': r're:^https?://.*\.jpg',
 190             'duration': 2578,
 191         },
 192         'params': {
 193             'skip_download': True,
 194         },
 195         'add_ie': [NovaEmbedIE.ie_key()],
 196         'skip': 'CHYBA 404: STRÁNKA NENALEZENA',
 197     }, {
 198         'url': 'http://sport.tn.nova.cz/clanek/sport/hokej/nhl/zivot-jde-dal-hodnotil-po-vyrazeni-z-playoff-jiri-sekac.html',
 199         'only_matching': True,
 200     }, {
 201         'url': 'http://fanda.nova.cz/clanek/fun-and-games/krvavy-epos-zaklinac-3-divoky-hon-vychazi-vyhrajte-ho-pro-sebe.html',
 202         'only_matching': True,
 203     }, {
 204         'url': 'http://doma.nova.cz/clanek/zdravi/prijdte-se-zapsat-do-registru-kostni-drene-jiz-ve-stredu-3-cervna.html',
 205         'only_matching': True,
 206     }, {
 207         'url': 'http://prask.nova.cz/clanek/novinky/co-si-na-sobe-nase-hvezdy-nechaly-pojistit.html',
 208         'only_matching': True,
 209     }, {
 210         'url': 'http://tv.nova.cz/clanek/novinky/zivot-je-zivot-bondovsky-trailer.html',
 211         'only_matching': True,
 212     }]
 213
 214     def _real_extract(self, url):
 215         mobj = self._match_valid_url(url)
 216         display_id = mobj.group('id')
 217         site = mobj.group('site')
 218
 219         webpage = self._download_webpage(url, display_id)
 220
 221         description = clean_html(self._og_search_description(webpage, default=None))
 222         if site == 'novaplus':
 223             upload_date = unified_strdate(self._search_regex(
 224                 r'(\d{1,2}-\d{1,2}-\d{4})$', display_id, 'upload date', default=None))
 225         elif site == 'fanda':
 226             upload_date = unified_strdate(self._search_regex(
 227                 r'<span class="date_time">(\d{1,2}\.\d{1,2}\.\d{4})', webpage, 'upload date', default=None))
 228         else:
 229             upload_date = None
 230
 231         # novaplus
 232         embed_id = self._search_regex(
 233             r'<iframe[^>]+\bsrc=["\'](?:https?:)?//media\.cms\.nova\.cz/embed/([^/?#&]+)',
 234             webpage, 'embed url', default=None)
 235         if embed_id:
 236             return {
 237                 '_type': 'url_transparent',
 238                 'url': 'https://media.cms.nova.cz/embed/%s' % embed_id,
 239                 'ie_key': NovaEmbedIE.ie_key(),
 240                 'id': embed_id,
 241                 'description': description,
 242                 'upload_date': upload_date
 243             }
 244
 245         video_id = self._search_regex(
 246             [r"(?:media|video_id)\s*:\s*'(\d+)'",
 247              r'media=(\d+)',
 248              r'id="article_video_(\d+)"',
 249              r'id="player_(\d+)"'],
 250             webpage, 'video id')
 251
 252         config_url = self._search_regex(
 253             r'src="(https?://(?:tn|api)\.nova\.cz/bin/player/videojs/config\.php\?[^"]+)"',
 254             webpage, 'config url', default=None)
 255         config_params = {}
 256
 257         if not config_url:
 258             player = self._parse_json(
 259                 self._search_regex(
 260                     r'(?s)Player\s*\(.+?\s*,\s*({.+?\bmedia\b["\']?\s*:\s*["\']?\d+.+?})\s*\)', webpage,
 261                     'player', default='{}'),
 262                 video_id, transform_source=js_to_json, fatal=False)
 263             if player:
 264                 config_url = url_or_none(player.get('configUrl'))
 265                 params = player.get('configParams')
 266                 if isinstance(params, dict):
 267                     config_params = params
 268
 269         if not config_url:
 270             DEFAULT_SITE_ID = '23000'
 271             SITES = {
 272                 'tvnoviny': DEFAULT_SITE_ID,
 273                 'novaplus': DEFAULT_SITE_ID,
 274                 'vymena': DEFAULT_SITE_ID,
 275                 'krasna': DEFAULT_SITE_ID,
 276                 'fanda': '30',
 277                 'tn': '30',
 278                 'doma': '30',
 279             }
 280
 281             site_id = self._search_regex(
 282                 r'site=(\d+)', webpage, 'site id', default=None) or SITES.get(
 283                 site, DEFAULT_SITE_ID)
 284
 285             config_url = 'https://api.nova.cz/bin/player/videojs/config.php'
 286             config_params = {
 287                 'site': site_id,
 288                 'media': video_id,
 289                 'quality': 3,
 290                 'version': 1,
 291             }
 292
 293         config = self._download_json(
 294             config_url, display_id,
 295             'Downloading config JSON', query=config_params,
 296             transform_source=lambda s: s[s.index('{'):s.rindex('}') + 1])
 297
 298         mediafile = config['mediafile']
 299         video_url = mediafile['src']
 300
 301         m = re.search(r'^(?P<url>rtmpe?://[^/]+/(?P<app>[^/]+?))/&*(?P<playpath>.+)$', video_url)
 302         if m:
 303             formats = [{
 304                 'url': m.group('url'),
 305                 'app': m.group('app'),
 306                 'play_path': m.group('playpath'),
 307                 'player_path': 'http://tvnoviny.nova.cz/static/shared/app/videojs/video-js.swf',
 308                 'ext': 'flv',
 309             }]
 310         else:
 311             formats = [{
 312                 'url': video_url,
 313             }]
 314         self._sort_formats(formats)
 315
 316         title = mediafile.get('meta', {}).get('title') or self._og_search_title(webpage)
 317         thumbnail = config.get('poster')
 318
 319         return {
 320             'id': video_id,
 321             'display_id': display_id,
 322             'title': title,
 323             'description': description,
 324             'upload_date': upload_date,
 325             'thumbnail': thumbnail,
 326             'formats': formats,
 327         }