_ Git - youtube-dl/blob - youtube_dl/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     int_or_none,
  10     js_to_json,
  11     qualities,
  12     unified_strdate,
  13     url_or_none,
  14 )
  15
  16
  17 class NovaEmbedIE(InfoExtractor):
  18     _VALID_URL = r'https?://media\.cms\.nova\.cz/embed/(?P<id>[^/?#&]+)'
  19     _TEST = {
  20         'url': 'https://media.cms.nova.cz/embed/8o0n0r?autoplay=1',
  21         'md5': 'b3834f6de5401baabf31ed57456463f7',
  22         'info_dict': {
  23             'id': '8o0n0r',
  24             'ext': 'mp4',
  25             'title': '2180. díl',
  26             'thumbnail': r're:^https?://.*\.jpg',
  27             'duration': 2578,
  28         },
  29     }
  30
  31     def _real_extract(self, url):
  32         video_id = self._match_id(url)
  33
  34         webpage = self._download_webpage(url, video_id)
  35
  36         bitrates = self._parse_json(
  37             self._search_regex(
  38                 r'(?s)(?:src|bitrates)\s*=\s*({.+?})\s*;', webpage, 'formats'),
  39             video_id, transform_source=js_to_json)
  40
  41         QUALITIES = ('lq', 'mq', 'hq', 'hd')
  42         quality_key = qualities(QUALITIES)
  43
  44         formats = []
  45         for format_id, format_list in bitrates.items():
  46             if format_id == 'hls':
  47                 m3u8_url = url_or_none(format_list)
  48                 if not m3u8_url:
  49                     continue
  50                 formats.extend(self._extract_m3u8_formats(
  51                     m3u8_url, video_id, ext='mp4', m3u8_id='hls', fatal=False))
  52
  53             if not isinstance(format_list, list):
  54                 continue
  55             for format_url in format_list:
  56                 format_url = url_or_none(format_url)
  57                 if not format_url:
  58                     continue
  59                 f = {
  60                     'url': format_url,
  61                 }
  62                 f_id = format_id
  63                 for quality in QUALITIES:
  64                     if '%s.mp4' % quality in format_url:
  65                         f_id += '-%s' % quality
  66                         f.update({
  67                             'quality': quality_key(quality),
  68                             'format_note': quality.upper(),
  69                         })
  70                         break
  71                 f['format_id'] = f_id
  72                 formats.append(f)
  73         self._sort_formats(formats)
  74
  75         title = self._og_search_title(
  76             webpage, default=None) or self._search_regex(
  77             (r'<value>(?P<title>[^<]+)',
  78              r'videoTitle\s*:\s*(["\'])(?P<value>(?:(?!\1).)+)\1'), webpage,
  79             'title', group='value')
  80         thumbnail = self._og_search_thumbnail(
  81             webpage, default=None) or self._search_regex(
  82             r'poster\s*:\s*(["\'])(?P<value>(?:(?!\1).)+)\1', webpage,
  83             'thumbnail', fatal=False, group='value')
  84         duration = int_or_none(self._search_regex(
  85             r'videoDuration\s*:\s*(\d+)', webpage, 'duration', fatal=False))
  86
  87         return {
  88             'id': video_id,
  89             'title': title,
  90             'thumbnail': thumbnail,
  91             'duration': duration,
  92             'formats': formats,
  93         }
  94
  95
  96 class NovaIE(InfoExtractor):
  97     IE_DESC = 'TN.cz, Prásk.tv, Nova.cz, Novaplus.cz, FANDA.tv, Krásná.cz and Doma.cz'
  98     _VALID_URL = r'https?://(?:[^.]+\.)?(?P<site>tv(?:noviny)?|tn|novaplus|vymena|fanda|krasna|doma|prask)\.nova\.cz/(?:[^/]+/)+(?P<id>[^/]+?)(?:\.html|/|$)'
  99     _TESTS = [{
 100         'url': 'http://tn.nova.cz/clanek/tajemstvi-ukryte-v-podzemi-specialni-nemocnice-v-prazske-krci.html#player_13260',
 101         'md5': '1dd7b9d5ea27bc361f110cd855a19bd3',
 102         'info_dict': {
 103             'id': '1757139',
 104             'display_id': 'tajemstvi-ukryte-v-podzemi-specialni-nemocnice-v-prazske-krci',
 105             'ext': 'mp4',
 106             'title': 'Podzemní nemocnice v pražské Krči',
 107             'description': 'md5:f0a42dd239c26f61c28f19e62d20ef53',
 108             'thumbnail': r're:^https?://.*\.(?:jpg)',
 109         }
 110     }, {
 111         'url': 'http://fanda.nova.cz/clanek/fun-and-games/krvavy-epos-zaklinac-3-divoky-hon-vychazi-vyhrajte-ho-pro-sebe.html',
 112         'info_dict': {
 113             'id': '1753621',
 114             'ext': 'mp4',
 115             'title': 'Zaklínač 3: Divoký hon',
 116             'description': 're:.*Pokud se stejně jako my nemůžete.*',
 117             'thumbnail': r're:https?://.*\.jpg(\?.*)?',
 118             'upload_date': '20150521',
 119         },
 120         'params': {
 121             # rtmp download
 122             'skip_download': True,
 123         }
 124     }, {
 125         # media.cms.nova.cz embed
 126         'url': 'https://novaplus.nova.cz/porad/ulice/epizoda/18760-2180-dil',
 127         'info_dict': {
 128             'id': '8o0n0r',
 129             'ext': 'mp4',
 130             'title': '2180. díl',
 131             'thumbnail': r're:^https?://.*\.jpg',
 132             'duration': 2578,
 133         },
 134         'params': {
 135             'skip_download': True,
 136         },
 137         'add_ie': [NovaEmbedIE.ie_key()],
 138     }, {
 139         'url': 'http://sport.tn.nova.cz/clanek/sport/hokej/nhl/zivot-jde-dal-hodnotil-po-vyrazeni-z-playoff-jiri-sekac.html',
 140         'only_matching': True,
 141     }, {
 142         'url': 'http://fanda.nova.cz/clanek/fun-and-games/krvavy-epos-zaklinac-3-divoky-hon-vychazi-vyhrajte-ho-pro-sebe.html',
 143         'only_matching': True,
 144     }, {
 145         'url': 'http://doma.nova.cz/clanek/zdravi/prijdte-se-zapsat-do-registru-kostni-drene-jiz-ve-stredu-3-cervna.html',
 146         'only_matching': True,
 147     }, {
 148         'url': 'http://prask.nova.cz/clanek/novinky/co-si-na-sobe-nase-hvezdy-nechaly-pojistit.html',
 149         'only_matching': True,
 150     }, {
 151         'url': 'http://tv.nova.cz/clanek/novinky/zivot-je-zivot-bondovsky-trailer.html',
 152         'only_matching': True,
 153     }]
 154
 155     def _real_extract(self, url):
 156         mobj = re.match(self._VALID_URL, url)
 157         display_id = mobj.group('id')
 158         site = mobj.group('site')
 159
 160         webpage = self._download_webpage(url, display_id)
 161
 162         # novaplus
 163         embed_id = self._search_regex(
 164             r'<iframe[^>]+\bsrc=["\'](?:https?:)?//media\.cms\.nova\.cz/embed/([^/?#&]+)',
 165             webpage, 'embed url', default=None)
 166         if embed_id:
 167             return self.url_result(
 168                 'https://media.cms.nova.cz/embed/%s' % embed_id,
 169                 ie=NovaEmbedIE.ie_key(), video_id=embed_id)
 170
 171         video_id = self._search_regex(
 172             [r"(?:media|video_id)\s*:\s*'(\d+)'",
 173              r'media=(\d+)',
 174              r'id="article_video_(\d+)"',
 175              r'id="player_(\d+)"'],
 176             webpage, 'video id')
 177
 178         config_url = self._search_regex(
 179             r'src="(https?://(?:tn|api)\.nova\.cz/bin/player/videojs/config\.php\?[^"]+)"',
 180             webpage, 'config url', default=None)
 181         config_params = {}
 182
 183         if not config_url:
 184             player = self._parse_json(
 185                 self._search_regex(
 186                     r'(?s)Player\s*\(.+?\s*,\s*({.+?\bmedia\b["\']?\s*:\s*["\']?\d+.+?})\s*\)', webpage,
 187                     'player', default='{}'),
 188                 video_id, transform_source=js_to_json, fatal=False)
 189             if player:
 190                 config_url = url_or_none(player.get('configUrl'))
 191                 params = player.get('configParams')
 192                 if isinstance(params, dict):
 193                     config_params = params
 194
 195         if not config_url:
 196             DEFAULT_SITE_ID = '23000'
 197             SITES = {
 198                 'tvnoviny': DEFAULT_SITE_ID,
 199                 'novaplus': DEFAULT_SITE_ID,
 200                 'vymena': DEFAULT_SITE_ID,
 201                 'krasna': DEFAULT_SITE_ID,
 202                 'fanda': '30',
 203                 'tn': '30',
 204                 'doma': '30',
 205             }
 206
 207             site_id = self._search_regex(
 208                 r'site=(\d+)', webpage, 'site id', default=None) or SITES.get(
 209                 site, DEFAULT_SITE_ID)
 210
 211             config_url = 'https://api.nova.cz/bin/player/videojs/config.php'
 212             config_params = {
 213                 'site': site_id,
 214                 'media': video_id,
 215                 'quality': 3,
 216                 'version': 1,
 217             }
 218
 219         config = self._download_json(
 220             config_url, display_id,
 221             'Downloading config JSON', query=config_params,
 222             transform_source=lambda s: s[s.index('{'):s.rindex('}') + 1])
 223
 224         mediafile = config['mediafile']
 225         video_url = mediafile['src']
 226
 227         m = re.search(r'^(?P<url>rtmpe?://[^/]+/(?P<app>[^/]+?))/&*(?P<playpath>.+)$', video_url)
 228         if m:
 229             formats = [{
 230                 'url': m.group('url'),
 231                 'app': m.group('app'),
 232                 'play_path': m.group('playpath'),
 233                 'player_path': 'http://tvnoviny.nova.cz/static/shared/app/videojs/video-js.swf',
 234                 'ext': 'flv',
 235             }]
 236         else:
 237             formats = [{
 238                 'url': video_url,
 239             }]
 240         self._sort_formats(formats)
 241
 242         title = mediafile.get('meta', {}).get('title') or self._og_search_title(webpage)
 243         description = clean_html(self._og_search_description(webpage, default=None))
 244         thumbnail = config.get('poster')
 245
 246         if site == 'novaplus':
 247             upload_date = unified_strdate(self._search_regex(
 248                 r'(\d{1,2}-\d{1,2}-\d{4})$', display_id, 'upload date', default=None))
 249         elif site == 'fanda':
 250             upload_date = unified_strdate(self._search_regex(
 251                 r'<span class="date_time">(\d{1,2}\.\d{1,2}\.\d{4})', webpage, 'upload date', default=None))
 252         else:
 253             upload_date = None
 254
 255         return {
 256             'id': video_id,
 257             'display_id': display_id,
 258             'title': title,
 259             'description': description,
 260             'upload_date': upload_date,
 261             'thumbnail': thumbnail,
 262             'formats': formats,
 263         }