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

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import itertools
   5 import re
   6
   7 from .common import InfoExtractor
   8 from ..compat import (
   9     compat_str,
  10     compat_urllib_parse_unquote,
  11     compat_urlparse
  12 )
  13 from ..utils import (
  14     extract_attributes,
  15     int_or_none,
  16     strip_or_none,
  17     unified_timestamp,
  18 )
  19
  20
  21 class PolskieRadioIE(InfoExtractor):
  22     _VALID_URL = r'https?://(?:www\.)?polskieradio\.pl/\d+/\d+/Artykul/(?P<id>[0-9]+)'
  23     _TESTS = [{
  24         'url': 'http://www.polskieradio.pl/7/5102/Artykul/1587943,Prof-Andrzej-Nowak-o-historii-nie-da-sie-myslec-beznamietnie',
  25         'info_dict': {
  26             'id': '1587943',
  27             'title': 'Prof. Andrzej Nowak: o historii nie da się myśleć beznamiętnie',
  28             'description': 'md5:12f954edbf3120c5e7075e17bf9fc5c5',
  29         },
  30         'playlist': [{
  31             'md5': '2984ee6ce9046d91fc233bc1a864a09a',
  32             'info_dict': {
  33                 'id': '1540576',
  34                 'ext': 'mp3',
  35                 'title': 'md5:d4623290d4ac983bf924061c75c23a0d',
  36                 'timestamp': 1456594200,
  37                 'upload_date': '20160227',
  38                 'duration': 2364,
  39                 'thumbnail': r're:^https?://static\.prsa\.pl/images/.*\.jpg$'
  40             },
  41         }],
  42     }, {
  43         'url': 'http://www.polskieradio.pl/265/5217/Artykul/1635803,Euro-2016-nie-ma-miejsca-na-blad-Polacy-graja-ze-Szwajcaria-o-cwiercfinal',
  44         'info_dict': {
  45             'id': '1635803',
  46             'title': 'Euro 2016: nie ma miejsca na błąd. Polacy grają ze Szwajcarią o ćwierćfinał',
  47             'description': 'md5:01cb7d0cad58664095d72b51a1ebada2',
  48         },
  49         'playlist_mincount': 12,
  50     }, {
  51         'url': 'http://polskieradio.pl/9/305/Artykul/1632955,Bardzo-popularne-slowo-remis',
  52         'only_matching': True,
  53     }, {
  54         'url': 'http://www.polskieradio.pl/7/5102/Artykul/1587943',
  55         'only_matching': True,
  56     }, {
  57         # with mp4 video
  58         'url': 'http://www.polskieradio.pl/9/299/Artykul/1634903,Brexit-Leszek-Miller-swiat-sie-nie-zawali-Europa-bedzie-trwac-dalej',
  59         'only_matching': True,
  60     }]
  61
  62     def _real_extract(self, url):
  63         playlist_id = self._match_id(url)
  64
  65         webpage = self._download_webpage(url, playlist_id)
  66
  67         content = self._search_regex(
  68             r'(?s)<div[^>]+class="\s*this-article\s*"[^>]*>(.+?)<div[^>]+class="tags"[^>]*>',
  69             webpage, 'content')
  70
  71         timestamp = unified_timestamp(self._html_search_regex(
  72             r'(?s)<span[^>]+id="datetime2"[^>]*>(.+?)</span>',
  73             webpage, 'timestamp', fatal=False))
  74
  75         thumbnail_url = self._og_search_thumbnail(webpage)
  76
  77         entries = []
  78
  79         media_urls = set()
  80
  81         for data_media in re.findall(r'<[^>]+data-media=({[^>]+})', content):
  82             media = self._parse_json(data_media, playlist_id, fatal=False)
  83             if not media.get('file') or not media.get('desc'):
  84                 continue
  85             media_url = self._proto_relative_url(media['file'], 'http:')
  86             if media_url in media_urls:
  87                 continue
  88             media_urls.add(media_url)
  89             entries.append({
  90                 'id': compat_str(media['id']),
  91                 'url': media_url,
  92                 'title': compat_urllib_parse_unquote(media['desc']),
  93                 'duration': int_or_none(media.get('length')),
  94                 'vcodec': 'none' if media.get('provider') == 'audio' else None,
  95                 'timestamp': timestamp,
  96                 'thumbnail': thumbnail_url
  97             })
  98
  99         title = self._og_search_title(webpage).strip()
 100         description = strip_or_none(self._og_search_description(webpage))
 101
 102         return self.playlist_result(entries, playlist_id, title, description)
 103
 104
 105 class PolskieRadioCategoryIE(InfoExtractor):
 106     _VALID_URL = r'https?://(?:www\.)?polskieradio\.pl/\d+(?:,[^/]+)?/(?P<id>\d+)'
 107     _TESTS = [{
 108         'url': 'http://www.polskieradio.pl/7/5102,HISTORIA-ZYWA',
 109         'info_dict': {
 110             'id': '5102',
 111             'title': 'HISTORIA ŻYWA',
 112         },
 113         'playlist_mincount': 38,
 114     }, {
 115         'url': 'http://www.polskieradio.pl/7/4807',
 116         'info_dict': {
 117             'id': '4807',
 118             'title': 'Vademecum 1050. rocznicy Chrztu Polski'
 119         },
 120         'playlist_mincount': 5
 121     }, {
 122         'url': 'http://www.polskieradio.pl/7/129,Sygnaly-dnia?ref=source',
 123         'only_matching': True
 124     }, {
 125         'url': 'http://www.polskieradio.pl/37,RedakcjaKatolicka/4143,Kierunek-Krakow',
 126         'info_dict': {
 127             'id': '4143',
 128             'title': 'Kierunek Kraków',
 129         },
 130         'playlist_mincount': 61
 131     }, {
 132         'url': 'http://www.polskieradio.pl/10,czworka/214,muzyka',
 133         'info_dict': {
 134             'id': '214',
 135             'title': 'Muzyka',
 136         },
 137         'playlist_mincount': 61
 138     }, {
 139         'url': 'http://www.polskieradio.pl/7,Jedynka/5102,HISTORIA-ZYWA',
 140         'only_matching': True,
 141     }, {
 142         'url': 'http://www.polskieradio.pl/8,Dwojka/196,Publicystyka',
 143         'only_matching': True,
 144     }]
 145
 146     @classmethod
 147     def suitable(cls, url):
 148         return False if PolskieRadioIE.suitable(url) else super(PolskieRadioCategoryIE, cls).suitable(url)
 149
 150     def _entries(self, url, page, category_id):
 151         content = page
 152         for page_num in itertools.count(2):
 153             for a_entry, entry_id in re.findall(
 154                     r'(?s)<article[^>]+>.*?(<a[^>]+href=["\']/\d+/\d+/Artykul/(\d+)[^>]+>).*?</article>',
 155                     content):
 156                 entry = extract_attributes(a_entry)
 157                 href = entry.get('href')
 158                 if not href:
 159                     continue
 160                 yield self.url_result(
 161                     compat_urlparse.urljoin(url, href), PolskieRadioIE.ie_key(),
 162                     entry_id, entry.get('title'))
 163             mobj = re.search(
 164                 r'<div[^>]+class=["\']next["\'][^>]*>\s*<a[^>]+href=(["\'])(?P<url>(?:(?!\1).)+)\1',
 165                 content)
 166             if not mobj:
 167                 break
 168             next_url = compat_urlparse.urljoin(url, mobj.group('url'))
 169             content = self._download_webpage(
 170                 next_url, category_id, 'Downloading page %s' % page_num)
 171
 172     def _real_extract(self, url):
 173         category_id = self._match_id(url)
 174         webpage = self._download_webpage(url, category_id)
 175         title = self._html_search_regex(
 176             r'<title>([^<]+) - [^<]+ - [^<]+</title>',
 177             webpage, 'title', fatal=False)
 178         return self.playlist_result(
 179             self._entries(url, webpage, category_id),
 180             category_id, title)