youtube_dl/extractor/nova.py

   1 # encoding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import determine_ext
   8
   9
  10 class NovaIE(InfoExtractor):
  11     IE_DESC = 'TN.cz, Prásk.tv, Nova.cz, Novaplus.cz, FANDA.tv, Krásná.cz and Doma.cz'
  12     _VALID_URL = 'http://(?:[^.]+\.)?(?P<site>tv(?:noviny)?|tn|novaplus|vymena|fanda|krasna|doma|prask)\.nova\.cz/(?:[^/]+/)+(?P<id>[^/]+?)(?:\.html|/?)$'
  13     _TESTS = [{
  14         'url': 'http://tvnoviny.nova.cz/clanek/novinky/co-na-sebe-sportaci-praskli-vime-jestli-pujde-hrdlicka-na-materskou.html',
  15         'info_dict': {
  16             'id': '1608920',
  17             'display_id': 'co-na-sebe-sportaci-praskli-vime-jestli-pujde-hrdlicka-na-materskou',
  18             'ext': 'flv',
  19             'title': 'Duel: Michal Hrdlička a Petr Suchoň',
  20             'description': 'md5:d0cc509858eee1b1374111c588c6f5d5',
  21             'thumbnail': 're:^https?://.*\.(?:jpg)',
  22         },
  23         'params': {
  24             # rtmp download
  25             'skip_download': True,
  26         }
  27     }, {
  28         'url': 'http://tn.nova.cz/clanek/tajemstvi-ukryte-v-podzemi-specialni-nemocnice-v-prazske-krci.html',
  29         'md5': '1dd7b9d5ea27bc361f110cd855a19bd3',
  30         'info_dict': {
  31             'id': '1757139',
  32             'display_id': 'tajemstvi-ukryte-v-podzemi-specialni-nemocnice-v-prazske-krci',
  33             'ext': 'mp4',
  34             'title': 'Podzemní nemocnice v pražské Krči',
  35             'description': 'md5:f0a42dd239c26f61c28f19e62d20ef53',
  36             'thumbnail': 're:^https?://.*\.(?:jpg)',
  37         }
  38     }, {
  39         'url': 'http://novaplus.nova.cz/porad/policie-modrava/video/5591-policie-modrava-15-dil-blondynka-na-hrbitove/',
  40         'info_dict': {
  41             'id': '1756825',
  42             'display_id': '5591-policie-modrava-15-dil-blondynka-na-hrbitove',
  43             'ext': 'mp4',
  44             'title': 'Policie Modrava - 15. díl - Blondýnka na hřbitově',
  45             'description': 'md5:d804ba6b30bc7da2705b1fea961bddfe',
  46             'thumbnail': 're:^https?://.*\.(?:jpg)',
  47         },
  48         'params': {
  49             # rtmp download
  50             'skip_download': True,
  51         }
  52     }, {
  53         'url': 'http://novaplus.nova.cz/porad/televizni-noviny/video/5585-televizni-noviny-30-5-2015/',
  54         'info_dict': {
  55             'id': '1756858',
  56             'ext': 'mp4',
  57             'title': 'Televizní noviny - 30. 5. 2015',
  58             'thumbnail': 're:^https?://.*\.(?:jpg)',
  59             'upload_date': '20150530',
  60         },
  61         'params': {
  62             # rtmp download
  63             'skip_download': True,
  64         }
  65     }, {
  66         'url': 'http://fanda.nova.cz/clanek/fun-and-games/krvavy-epos-zaklinac-3-divoky-hon-vychazi-vyhrajte-ho-pro-sebe.html',
  67         'info_dict': {
  68             'id': '1753621',
  69             'ext': 'mp4',
  70             'title': 'Zaklínač 3: Divoký hon',
  71             'description': 're:.*Pokud se stejně jako my nemůžete.*',
  72             'thumbnail': 're:https?://.*\.jpg(\?.*)?',
  73             'upload_date': '20150521',
  74         },
  75         'params': {
  76             # rtmp download
  77             'skip_download': True,
  78         }
  79     }, {
  80         'url': 'http://sport.tn.nova.cz/clanek/sport/hokej/nhl/zivot-jde-dal-hodnotil-po-vyrazeni-z-playoff-jiri-sekac.html',
  81         'only_matching': True,
  82     }, {
  83         'url': 'http://fanda.nova.cz/clanek/fun-and-games/krvavy-epos-zaklinac-3-divoky-hon-vychazi-vyhrajte-ho-pro-sebe.html',
  84         'only_matching': True,
  85     }, {
  86         'url': 'http://doma.nova.cz/clanek/zdravi/prijdte-se-zapsat-do-registru-kostni-drene-jiz-ve-stredu-3-cervna.html',
  87         'only_matching': True,
  88     }, {
  89         'url': 'http://prask.nova.cz/clanek/novinky/co-si-na-sobe-nase-hvezdy-nechaly-pojistit.html',
  90         'only_matching': True,
  91     }, {
  92         'url': 'http://tv.nova.cz/clanek/novinky/zivot-je-zivot-bondovsky-trailer.html',
  93         'only_matching': True,
  94     }]
  95
  96     def _real_extract(self, url):
  97         mobj = re.match(self._VALID_URL, url)
  98         display_id = mobj.group('id')
  99         site = mobj.group('site')
 100
 101         webpage = self._download_webpage(url, display_id)
 102
 103         video_id = self._search_regex(
 104             [r"(?:media|video_id)\s*:\s*'(\d+)'",
 105              r'media=(\d+)',
 106              r'id="article_video_(\d+)"',
 107              r'id="player_(\d+)"'],
 108             webpage, 'video id')
 109
 110         config_url = self._search_regex(
 111             r'src="(http://tn\.nova\.cz/bin/player/videojs/config\.php\?[^"]+)"',
 112             webpage, 'config url', default=None)
 113
 114         if not config_url:
 115             DEFAULT_SITE_ID = '23000'
 116             SITES = {
 117                 'tvnoviny': DEFAULT_SITE_ID,
 118                 'novaplus': DEFAULT_SITE_ID,
 119                 'vymena': DEFAULT_SITE_ID,
 120                 'krasna': DEFAULT_SITE_ID,
 121                 'fanda': '30',
 122                 'tn': '30',
 123                 'doma': '30',
 124             }
 125
 126             site_id = self._search_regex(
 127                 r'site=(\d+)', webpage, 'site id', default=None) or SITES.get(site, DEFAULT_SITE_ID)
 128
 129             config_url = ('http://tn.nova.cz/bin/player/videojs/config.php?site=%s&media=%s&jsVar=vjsconfig'
 130                           % (site_id, video_id))
 131
 132         config = self._download_json(
 133             config_url, display_id,
 134             'Downloading config JSON',
 135             transform_source=lambda s: re.sub(r'var\s+[\da-zA-Z_]+\s*=\s*({.+?});', r'\1', s))
 136
 137         mediafile = config['mediafile']
 138         video_url = mediafile['src']
 139         ext = determine_ext(video_url)
 140         video_url = video_url.replace('&{}:'.format(ext), '')
 141
 142         title = mediafile.get('meta', {}).get('title') or self._og_search_title(webpage)
 143         description = self._og_search_description(webpage)
 144         thumbnail = config.get('poster')
 145
 146         mobj = None
 147         if site == 'novaplus':
 148             mobj = re.search(r'(?P<day>\d{1,2})-(?P<month>\d{1,2})-(?P<year>\d{4})$', display_id)
 149         if site == 'fanda':
 150             mobj = re.search(
 151                 r'<span class="date_time">(?P<day>\d{1,2})\.(?P<month>\d{1,2})\.(?P<year>\d{4})\b', webpage)
 152         if mobj:
 153             upload_date = '{}{:02d}{:02d}'.format(mobj.group('year'), int(mobj.group('month')), int(mobj.group('day')))
 154         else:
 155             upload_date = None
 156
 157         return {
 158             'id': video_id,
 159             'display_id': display_id,
 160             'title': title,
 161             'description': description,
 162             'upload_date': upload_date,
 163             'thumbnail': thumbnail,
 164             'url': video_url,
 165             'ext': ext,
 166         }