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

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     determine_ext,
   7     js_to_json,
   8 )
   9
  10
  11 class RTPIE(InfoExtractor):
  12     _VALID_URL = r'https?://(?:www\.)?rtp\.pt/play/p(?P<program_id>[0-9]+)/(?P<id>[^/?#]+)/?'
  13     _TESTS = [{
  14         'url': 'http://www.rtp.pt/play/p405/e174042/paixoes-cruzadas',
  15         'md5': 'e736ce0c665e459ddb818546220b4ef8',
  16         'info_dict': {
  17             'id': 'e174042',
  18             'ext': 'mp3',
  19             'title': 'Paixões Cruzadas',
  20             'description': 'As paixões musicais de António Cartaxo e António Macedo',
  21             'thumbnail': r're:^https?://.*\.jpg',
  22         },
  23     }, {
  24         'url': 'http://www.rtp.pt/play/p831/a-quimica-das-coisas',
  25         'only_matching': True,
  26     }]
  27
  28     def _real_extract(self, url):
  29         video_id = self._match_id(url)
  30
  31         webpage = self._download_webpage(url, video_id)
  32         title = self._html_search_meta(
  33             'twitter:title', webpage, display_name='title', fatal=True)
  34
  35         config = self._parse_json(self._search_regex(
  36             r'(?s)RTPPlayer\(({.+?})\);', webpage,
  37             'player config'), video_id, js_to_json)
  38         file_url = config['file']
  39         ext = determine_ext(file_url)
  40         if ext == 'm3u8':
  41             file_key = config.get('fileKey')
  42             formats = self._extract_m3u8_formats(
  43                 file_url, video_id, 'mp4', 'm3u8_native',
  44                 m3u8_id='hls', fatal=file_key)
  45             if file_key:
  46                 formats.append({
  47                     'url': 'https://cdn-ondemand.rtp.pt' + file_key,
  48                     'preference': 1,
  49                 })
  50             self._sort_formats(formats)
  51         else:
  52             formats = [{
  53                 'url': file_url,
  54                 'ext': ext,
  55             }]
  56         if config.get('mediaType') == 'audio':
  57             for f in formats:
  58                 f['vcodec'] = 'none'
  59
  60         return {
  61             'id': video_id,
  62             'title': title,
  63             'formats': formats,
  64             'description': self._html_search_meta(['description', 'twitter:description'], webpage),
  65             'thumbnail': config.get('poster') or self._og_search_thumbnail(webpage),
  66         }