_ Git - youtube-dl/blob - youtube_dl/extractor/ina.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     strip_or_none,
   8     xpath_attr,
   9     xpath_text,
  10 )
  11
  12
  13 class InaIE(InfoExtractor):
  14     _VALID_URL = r'https?://(?:www\.)?ina\.fr/video/(?P<id>[A-Z0-9_]+)'
  15     _TESTS = [{
  16         'url': 'http://www.ina.fr/video/I12055569/francois-hollande-je-crois-que-c-est-clair-video.html',
  17         'md5': 'a667021bf2b41f8dc6049479d9bb38a3',
  18         'info_dict': {
  19             'id': 'I12055569',
  20             'ext': 'mp4',
  21             'title': 'François Hollande "Je crois que c\'est clair"',
  22             'description': 'md5:3f09eb072a06cb286b8f7e4f77109663',
  23         }
  24     }, {
  25         'url': 'https://www.ina.fr/video/S806544_001/don-d-organes-des-avancees-mais-d-importants-besoins-video.html',
  26         'only_matching': True,
  27     }]
  28
  29     def _real_extract(self, url):
  30         video_id = self._match_id(url)
  31         info_doc = self._download_xml(
  32             'http://player.ina.fr/notices/%s.mrss' % video_id, video_id)
  33         item = info_doc.find('channel/item')
  34         title = xpath_text(item, 'title', fatal=True)
  35         media_ns_xpath = lambda x: self._xpath_ns(x, 'http://search.yahoo.com/mrss/')
  36         content = item.find(media_ns_xpath('content'))
  37
  38         get_furl = lambda x: xpath_attr(content, media_ns_xpath(x), 'url')
  39         formats = []
  40         for q, w, h in (('bq', 400, 300), ('mq', 512, 384), ('hq', 768, 576)):
  41             q_url = get_furl(q)
  42             if not q_url:
  43                 continue
  44             formats.append({
  45                 'format_id': q,
  46                 'url': q_url,
  47                 'width': w,
  48                 'height': h,
  49             })
  50         if not formats:
  51             formats = [{
  52                 'url': get_furl('player') or content.attrib['url'],
  53             }]
  54
  55         thumbnails = []
  56         for thumbnail in content.findall(media_ns_xpath('thumbnail')):
  57             thumbnail_url = thumbnail.get('url')
  58             if not thumbnail_url:
  59                 continue
  60             thumbnails.append({
  61                 'url': thumbnail_url,
  62                 'height': int_or_none(thumbnail.get('height')),
  63                 'width': int_or_none(thumbnail.get('width')),
  64             })
  65
  66         return {
  67             'id': video_id,
  68             'formats': formats,
  69             'title': title,
  70             'description': strip_or_none(xpath_text(item, 'description')),
  71             'thumbnails': thumbnails,
  72         }