youtube_dl/extractor/franceculture.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     extract_attributes,
   8     int_or_none,
   9 )
  10
  11
  12 class FranceCultureIE(InfoExtractor):
  13     _VALID_URL = r'https?://(?:www\.)?franceculture\.fr/emissions/(?:[^/]+/)*(?P<id>[^/?#&]+)'
  14     _TESTS = [{
  15         'url': 'http://www.franceculture.fr/emissions/carnet-nomade/rendez-vous-au-pays-des-geeks',
  16         'info_dict': {
  17             'id': 'rendez-vous-au-pays-des-geeks',
  18             'display_id': 'rendez-vous-au-pays-des-geeks',
  19             'ext': 'mp3',
  20             'title': 'Rendez-vous au pays des geeks',
  21             'thumbnail': r're:^https?://.*\.jpg$',
  22             'upload_date': '20140301',
  23             'timestamp': 1393700400,
  24             'vcodec': 'none',
  25         }
  26     }, {
  27         # no thumbnail
  28         'url': 'https://www.franceculture.fr/emissions/la-recherche-montre-en-main/la-recherche-montre-en-main-du-mercredi-10-octobre-2018',
  29         'only_matching': True,
  30     }]
  31
  32     def _real_extract(self, url):
  33         display_id = self._match_id(url)
  34
  35         webpage = self._download_webpage(url, display_id)
  36
  37         video_data = extract_attributes(self._search_regex(
  38             r'''(?sx)
  39                 (?:
  40                     </h1>|
  41                     <div[^>]+class="[^"]*?(?:title-zone-diffusion|heading-zone-(?:wrapper|player-button))[^"]*?"[^>]*>
  42                 ).*?
  43                 (<button[^>]+data-(?:url|asset-source)="[^"]+"[^>]+>)
  44             ''',
  45             webpage, 'video data'))
  46
  47         video_url = video_data.get('data-url') or video_data['data-asset-source']
  48         title = video_data.get('data-asset-title') or video_data.get('data-diffusion-title') or self._og_search_title(webpage)
  49
  50         description = self._html_search_regex(
  51             r'(?s)<div[^>]+class="intro"[^>]*>.*?<h2>(.+?)</h2>',
  52             webpage, 'description', default=None)
  53         thumbnail = self._search_regex(
  54             r'(?s)<figure[^>]+itemtype="https://schema.org/ImageObject"[^>]*>.*?<img[^>]+(?:data-dejavu-)?src="([^"]+)"',
  55             webpage, 'thumbnail', default=None)
  56         uploader = self._html_search_regex(
  57             r'(?s)<span class="author">(.*?)</span>',
  58             webpage, 'uploader', default=None)
  59         ext = determine_ext(video_url.lower())
  60
  61         return {
  62             'id': display_id,
  63             'display_id': display_id,
  64             'url': video_url,
  65             'title': title,
  66             'description': description,
  67             'thumbnail': thumbnail,
  68             'ext': ext,
  69             'vcodec': 'none' if ext == 'mp3' else None,
  70             'uploader': uploader,
  71             'timestamp': int_or_none(video_data.get('data-start-time')) or int_or_none(video_data.get('data-asset-created-date')),
  72             'duration': int_or_none(video_data.get('data-duration')),
  73         }