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