Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/tvn24.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     int_or_none,
   7     NO_DEFAULT,
   8     unescapeHTML,
   9 )
  10
  11
  12 class TVN24IE(InfoExtractor):
  13     _VALID_URL = r'https?://(?:(?:[^/]+)\.)?tvn24(?:bis)?\.pl/(?:[^/]+/)*(?P<id>[^/]+)'
  14     _TESTS = [{
  15         'url': 'http://www.tvn24.pl/wiadomosci-z-kraju,3/oredzie-artura-andrusa,702428.html',
  16         'md5': 'fbdec753d7bc29d96036808275f2130c',
  17         'info_dict': {
  18             'id': '1584444',
  19             'ext': 'mp4',
  20             'title': '"Święta mają być wesołe, dlatego, ludziska, wszyscy pod jemiołę"',
  21             'description': 'Wyjątkowe orędzie Artura Andrusa, jednego z gości Szkła kontaktowego.',
  22             'thumbnail': 're:https?://.*[.]jpeg',
  23         }
  24     }, {
  25         # different layout
  26         'url': 'https://tvnmeteo.tvn24.pl/magazyny/maja-w-ogrodzie,13/odcinki-online,1,4,1,0/pnacza-ptaki-i-iglaki-odc-691-hgtv-odc-29,1771763.html',
  27         'info_dict': {
  28             'id': '1771763',
  29             'ext': 'mp4',
  30             'title': 'Pnącza, ptaki i iglaki (odc. 691 /HGTV odc. 29)',
  31             'thumbnail': 're:https?://.*',
  32         },
  33         'params': {
  34             'skip_download': True,
  35         },
  36     }, {
  37         'url': 'http://fakty.tvn24.pl/ogladaj-online,60/53-konferencja-bezpieczenstwa-w-monachium,716431.html',
  38         'only_matching': True,
  39     }, {
  40         'url': 'http://sport.tvn24.pl/pilka-nozna,105/ligue-1-kamil-glik-rozcial-glowe-monaco-tylko-remisuje-z-bastia,716522.html',
  41         'only_matching': True,
  42     }, {
  43         'url': 'http://tvn24bis.pl/poranek,146,m/gen-koziej-w-tvn24-bis-wracamy-do-czasow-zimnej-wojny,715660.html',
  44         'only_matching': True,
  45     }, {
  46         'url': 'https://www.tvn24.pl/magazyn-tvn24/angie-w-jednej-czwartej-polka-od-szarej-myszki-do-cesarzowej-europy,119,2158',
  47         'only_matching': True,
  48     }]
  49
  50     def _real_extract(self, url):
  51         display_id = self._match_id(url)
  52
  53         webpage = self._download_webpage(url, display_id)
  54
  55         title = self._og_search_title(
  56             webpage, default=None) or self._search_regex(
  57             r'<h\d+[^>]+class=["\']magazineItemHeader[^>]+>(.+?)</h',
  58             webpage, 'title')
  59
  60         def extract_json(attr, name, default=NO_DEFAULT, fatal=True):
  61             return self._parse_json(
  62                 self._search_regex(
  63                     r'\b%s=(["\'])(?P<json>(?!\1).+?)\1' % attr, webpage,
  64                     name, group='json', default=default, fatal=fatal) or '{}',
  65                 display_id, transform_source=unescapeHTML, fatal=fatal)
  66
  67         quality_data = extract_json('data-quality', 'formats')
  68
  69         formats = []
  70         for format_id, url in quality_data.items():
  71             formats.append({
  72                 'url': url,
  73                 'format_id': format_id,
  74                 'height': int_or_none(format_id.rstrip('p')),
  75             })
  76         self._sort_formats(formats)
  77
  78         description = self._og_search_description(webpage, default=None)
  79         thumbnail = self._og_search_thumbnail(
  80             webpage, default=None) or self._html_search_regex(
  81             r'\bdata-poster=(["\'])(?P<url>(?!\1).+?)\1', webpage,
  82             'thumbnail', group='url')
  83
  84         video_id = None
  85
  86         share_params = extract_json(
  87             'data-share-params', 'share params', default=None)
  88         if isinstance(share_params, dict):
  89             video_id = share_params.get('id')
  90
  91         if not video_id:
  92             video_id = self._search_regex(
  93                 r'data-vid-id=["\'](\d+)', webpage, 'video id',
  94                 default=None) or self._search_regex(
  95                 r',(\d+)\.html', url, 'video id', default=display_id)
  96
  97         return {
  98             'id': video_id,
  99             'title': title,
 100             'description': description,
 101             'thumbnail': thumbnail,
 102             'formats': formats,
 103         }