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