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

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..compat import (
   8     compat_str,
   9     compat_urllib_parse_unquote,
  10 )
  11 from ..utils import (
  12     int_or_none,
  13     strip_or_none,
  14     unified_timestamp,
  15 )
  16
  17
  18 class PolskieRadioIE(InfoExtractor):
  19     _VALID_URL = r'https?://(?:www\.)?polskieradio\.pl/\d+/\d+/Artykul/(?P<id>[0-9]+)'
  20     _TESTS = [{
  21         'url': 'http://www.polskieradio.pl/7/5102/Artykul/1587943,Prof-Andrzej-Nowak-o-historii-nie-da-sie-myslec-beznamietnie',
  22         'info_dict': {
  23             'id': '1587943',
  24             'title': 'Prof. Andrzej Nowak: o historii nie da się myśleć beznamiętnie',
  25             'description': 'md5:12f954edbf3120c5e7075e17bf9fc5c5',
  26         },
  27         'playlist': [{
  28             'md5': '2984ee6ce9046d91fc233bc1a864a09a',
  29             'info_dict': {
  30                 'id': '1540576',
  31                 'ext': 'mp3',
  32                 'title': 'md5:d4623290d4ac983bf924061c75c23a0d',
  33                 'timestamp': 1456594200,
  34                 'upload_date': '20160227',
  35                 'duration': 2364,
  36                 'thumbnail': 're:^https?://static\.prsa\.pl/images/.*\.jpg$'
  37             },
  38         }],
  39     }, {
  40         'url': 'http://www.polskieradio.pl/265/5217/Artykul/1635803,Euro-2016-nie-ma-miejsca-na-blad-Polacy-graja-ze-Szwajcaria-o-cwiercfinal',
  41         'info_dict': {
  42             'id': '1635803',
  43             'title': 'Euro 2016: nie ma miejsca na błąd. Polacy grają ze Szwajcarią o ćwierćfinał',
  44             'description': 'md5:01cb7d0cad58664095d72b51a1ebada2',
  45         },
  46         'playlist_mincount': 12,
  47     }, {
  48         'url': 'http://polskieradio.pl/9/305/Artykul/1632955,Bardzo-popularne-slowo-remis',
  49         'only_matching': True,
  50     }, {
  51         'url': 'http://www.polskieradio.pl/7/5102/Artykul/1587943',
  52         'only_matching': True,
  53     }, {
  54         # with mp4 video
  55         'url': 'http://www.polskieradio.pl/9/299/Artykul/1634903,Brexit-Leszek-Miller-swiat-sie-nie-zawali-Europa-bedzie-trwac-dalej',
  56         'only_matching': True,
  57     }]
  58
  59     def _real_extract(self, url):
  60         playlist_id = self._match_id(url)
  61
  62         webpage = self._download_webpage(url, playlist_id)
  63
  64         content = self._search_regex(
  65             r'(?s)<div[^>]+class="audio atarticle"[^>]*>(.+?)<script>',
  66             webpage, 'content')
  67
  68         timestamp = unified_timestamp(self._html_search_regex(
  69             r'(?s)<span[^>]+id="datetime2"[^>]*>(.+?)</span>',
  70             webpage, 'timestamp', fatal=False))
  71
  72         thumbnail_url = self._og_search_thumbnail(webpage)
  73
  74         entries = []
  75
  76         media_urls = set()
  77
  78         for data_media in re.findall(r'<[^>]+data-media=({[^>]+})', content):
  79             media = self._parse_json(data_media, playlist_id, fatal=False)
  80             if not media.get('file') or not media.get('desc'):
  81                 continue
  82             media_url = self._proto_relative_url(media['file'], 'http:')
  83             if media_url in media_urls:
  84                 continue
  85             media_urls.add(media_url)
  86             entries.append({
  87                 'id': compat_str(media['id']),
  88                 'url': media_url,
  89                 'title': compat_urllib_parse_unquote(media['desc']),
  90                 'duration': int_or_none(media.get('length')),
  91                 'vcodec': 'none' if media.get('provider') == 'audio' else None,
  92                 'timestamp': timestamp,
  93                 'thumbnail': thumbnail_url
  94             })
  95
  96         title = self._og_search_title(webpage).strip()
  97         description = strip_or_none(self._og_search_description(webpage))
  98
  99         return self.playlist_result(entries, playlist_id, title, description)