Raphaël G. Git Repositories - youtubedl/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': 'ee009bafcc794541570edd44b71cbea3',
  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                 format_list = [format_list]
  48             for format_url in format_list:
  49                 format_url = url_or_none(format_url)
  50                 if not format_url:
  51                     continue
  52                 if format_id == 'hls':
  53                     formats.extend(self._extract_m3u8_formats(
  54                         format_url, video_id, ext='mp4',
  55                         entry_protocol='m3u8_native', m3u8_id='hls',
  56                         fatal=False))
  57                     continue
  58                 f = {
  59                     'url': format_url,
  60                 }
  61                 f_id = format_id
  62                 for quality in QUALITIES:
  63                     if '%s.mp4' % quality in format_url:
  64                         f_id += '-%s' % quality
  65                         f.update({
  66                             'quality': quality_key(quality),
  67                             'format_note': quality.upper(),
  68                         })
  69                         break
  70                 f['format_id'] = f_id
  71                 formats.append(f)
  72         self._sort_formats(formats)
  73
  74         title = self._og_search_title(
  75             webpage, default=None) or self._search_regex(
  76             (r'<value>(?P<title>[^<]+)',
  77              r'videoTitle\s*:\s*(["\'])(?P<value>(?:(?!\1).)+)\1'), webpage,
  78             'title', group='value')
  79         thumbnail = self._og_search_thumbnail(
  80             webpage, default=None) or self._search_regex(
  81             r'poster\s*:\s*(["\'])(?P<value>(?:(?!\1).)+)\1', webpage,
  82             'thumbnail', fatal=False, group='value')
  83         duration = int_or_none(self._search_regex(
  84             r'videoDuration\s*:\s*(\d+)', webpage, 'duration', fatal=False))
  85
  86         return {
  87             'id': video_id,
  88             'title': title,
  89             'thumbnail': thumbnail,
  90             'duration': duration,
  91             'formats': formats,
  92         }
  93
  94
  95 class NovaIE(InfoExtractor):
  96     IE_DESC = 'TN.cz, Prásk.tv, Nova.cz, Novaplus.cz, FANDA.tv, Krásná.cz and Doma.cz'
  97     _VALID_URL = r'https?://(?:[^.]+\.)?(?P<site>tv(?:noviny)?|tn|novaplus|vymena|fanda|krasna|doma|prask)\.nova\.cz/(?:[^/]+/)+(?P<id>[^/]+?)(?:\.html|/|$)'
  98     _TESTS = [{
  99         'url': 'http://tn.nova.cz/clanek/tajemstvi-ukryte-v-podzemi-specialni-nemocnice-v-prazske-krci.html#player_13260',
 100         'md5': '249baab7d0104e186e78b0899c7d5f28',
 101         'info_dict': {
 102             'id': '1757139',
 103             'display_id': 'tajemstvi-ukryte-v-podzemi-specialni-nemocnice-v-prazske-krci',
 104             'ext': 'mp4',
 105             'title': 'Podzemní nemocnice v pražské Krči',
 106             'description': 'md5:f0a42dd239c26f61c28f19e62d20ef53',
 107             'thumbnail': r're:^https?://.*\.(?:jpg)',
 108         }
 109     }, {
 110         'url': 'http://fanda.nova.cz/clanek/fun-and-games/krvavy-epos-zaklinac-3-divoky-hon-vychazi-vyhrajte-ho-pro-sebe.html',
 111         'info_dict': {
 112             'id': '1753621',
 113             'ext': 'mp4',
 114             'title': 'Zaklínač 3: Divoký hon',
 115             'description': 're:.*Pokud se stejně jako my nemůžete.*',
 116             'thumbnail': r're:https?://.*\.jpg(\?.*)?',
 117             'upload_date': '20150521',
 118         },
 119         'params': {
 120             # rtmp download
 121             'skip_download': True,
 122         },
 123         'skip': 'gone',
 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         'skip': 'CHYBA 404: STRÁNKA NENALEZENA',
 139     }, {
 140         'url': 'http://sport.tn.nova.cz/clanek/sport/hokej/nhl/zivot-jde-dal-hodnotil-po-vyrazeni-z-playoff-jiri-sekac.html',
 141         'only_matching': True,
 142     }, {
 143         'url': 'http://fanda.nova.cz/clanek/fun-and-games/krvavy-epos-zaklinac-3-divoky-hon-vychazi-vyhrajte-ho-pro-sebe.html',
 144         'only_matching': True,
 145     }, {
 146         'url': 'http://doma.nova.cz/clanek/zdravi/prijdte-se-zapsat-do-registru-kostni-drene-jiz-ve-stredu-3-cervna.html',
 147         'only_matching': True,
 148     }, {
 149         'url': 'http://prask.nova.cz/clanek/novinky/co-si-na-sobe-nase-hvezdy-nechaly-pojistit.html',
 150         'only_matching': True,
 151     }, {
 152         'url': 'http://tv.nova.cz/clanek/novinky/zivot-je-zivot-bondovsky-trailer.html',
 153         'only_matching': True,
 154     }]
 155
 156     def _real_extract(self, url):
 157         mobj = re.match(self._VALID_URL, url)
 158         display_id = mobj.group('id')
 159         site = mobj.group('site')
 160
 161         webpage = self._download_webpage(url, display_id)
 162
 163         description = clean_html(self._og_search_description(webpage, default=None))
 164         if site == 'novaplus':
 165             upload_date = unified_strdate(self._search_regex(
 166                 r'(\d{1,2}-\d{1,2}-\d{4})$', display_id, 'upload date', default=None))
 167         elif site == 'fanda':
 168             upload_date = unified_strdate(self._search_regex(
 169                 r'<span class="date_time">(\d{1,2}\.\d{1,2}\.\d{4})', webpage, 'upload date', default=None))
 170         else:
 171             upload_date = None
 172
 173         # novaplus
 174         embed_id = self._search_regex(
 175             r'<iframe[^>]+\bsrc=["\'](?:https?:)?//media\.cms\.nova\.cz/embed/([^/?#&]+)',
 176             webpage, 'embed url', default=None)
 177         if embed_id:
 178             return {
 179                 '_type': 'url_transparent',
 180                 'url': 'https://media.cms.nova.cz/embed/%s' % embed_id,
 181                 'ie_key': NovaEmbedIE.ie_key(),
 182                 'id': embed_id,
 183                 'description': description,
 184                 'upload_date': upload_date
 185             }
 186
 187         video_id = self._search_regex(
 188             [r"(?:media|video_id)\s*:\s*'(\d+)'",
 189              r'media=(\d+)',
 190              r'id="article_video_(\d+)"',
 191              r'id="player_(\d+)"'],
 192             webpage, 'video id')
 193
 194         config_url = self._search_regex(
 195             r'src="(https?://(?:tn|api)\.nova\.cz/bin/player/videojs/config\.php\?[^"]+)"',
 196             webpage, 'config url', default=None)
 197         config_params = {}
 198
 199         if not config_url:
 200             player = self._parse_json(
 201                 self._search_regex(
 202                     r'(?s)Player\s*\(.+?\s*,\s*({.+?\bmedia\b["\']?\s*:\s*["\']?\d+.+?})\s*\)', webpage,
 203                     'player', default='{}'),
 204                 video_id, transform_source=js_to_json, fatal=False)
 205             if player:
 206                 config_url = url_or_none(player.get('configUrl'))
 207                 params = player.get('configParams')
 208                 if isinstance(params, dict):
 209                     config_params = params
 210
 211         if not config_url:
 212             DEFAULT_SITE_ID = '23000'
 213             SITES = {
 214                 'tvnoviny': DEFAULT_SITE_ID,
 215                 'novaplus': DEFAULT_SITE_ID,
 216                 'vymena': DEFAULT_SITE_ID,
 217                 'krasna': DEFAULT_SITE_ID,
 218                 'fanda': '30',
 219                 'tn': '30',
 220                 'doma': '30',
 221             }
 222
 223             site_id = self._search_regex(
 224                 r'site=(\d+)', webpage, 'site id', default=None) or SITES.get(
 225                 site, DEFAULT_SITE_ID)
 226
 227             config_url = 'https://api.nova.cz/bin/player/videojs/config.php'
 228             config_params = {
 229                 'site': site_id,
 230                 'media': video_id,
 231                 'quality': 3,
 232                 'version': 1,
 233             }
 234
 235         config = self._download_json(
 236             config_url, display_id,
 237             'Downloading config JSON', query=config_params,
 238             transform_source=lambda s: s[s.index('{'):s.rindex('}') + 1])
 239
 240         mediafile = config['mediafile']
 241         video_url = mediafile['src']
 242
 243         m = re.search(r'^(?P<url>rtmpe?://[^/]+/(?P<app>[^/]+?))/&*(?P<playpath>.+)$', video_url)
 244         if m:
 245             formats = [{
 246                 'url': m.group('url'),
 247                 'app': m.group('app'),
 248                 'play_path': m.group('playpath'),
 249                 'player_path': 'http://tvnoviny.nova.cz/static/shared/app/videojs/video-js.swf',
 250                 'ext': 'flv',
 251             }]
 252         else:
 253             formats = [{
 254                 'url': video_url,
 255             }]
 256         self._sort_formats(formats)
 257
 258         title = mediafile.get('meta', {}).get('title') or self._og_search_title(webpage)
 259         thumbnail = config.get('poster')
 260
 261         return {
 262             'id': video_id,
 263             'display_id': display_id,
 264             'title': title,
 265             'description': description,
 266             'upload_date': upload_date,
 267             'thumbnail': thumbnail,
 268             'formats': formats,
 269         }