_ Git - youtube-dl/blob - youtube_dl/extractor/orf.py

   1 import re
   2 import xml.etree.ElementTree
   3 import json
   4
   5 from .common import InfoExtractor
   6 from ..utils import (
   7     compat_urlparse,
   8     ExtractorError,
   9     find_xpath_attr,
  10 )
  11
  12 class ORFIE(InfoExtractor):
  13     _VALID_URL = r'https?://tvthek.orf.at/(programs/.+?/episodes|topics/.+?)/(?P<id>\d+)'
  14
  15     _TEST = {
  16         u'url': u'http://tvthek.orf.at/programs/1171769-Wetter-ZIB/episodes/6557323-Wetter',
  17         u'file': u'6566957.flv',
  18         u'info_dict': {
  19             u'title': u'Wetter',
  20             u'description': u'Christa Kummer, Marcus Wadsak und Kollegen  präsentieren abwechselnd ihre täglichen Wetterprognosen für Österreich.\r \r Mehr Wetter unter wetter.ORF.at',
  21         },
  22         u'params': {
  23             # It uses rtmp
  24             u'skip_download': True,
  25         }
  26     }
  27
  28     def _real_extract(self, url):
  29         mobj = re.match(self._VALID_URL, url)
  30         playlist_id = mobj.group('id')
  31         webpage = self._download_webpage(url, playlist_id)
  32
  33         flash_xml = self._search_regex('ORF.flashXML = \'(.+?)\'', webpage, u'flash xml')
  34         flash_xml = compat_urlparse.parse_qs('xml='+flash_xml)['xml'][0]
  35         flash_config = xml.etree.ElementTree.fromstring(flash_xml.encode('utf-8'))
  36         playlist_json = self._search_regex(r'playlist\': \'(\[.*?\])\'', webpage, u'playlist').replace(r'\"','"')
  37         playlist = json.loads(playlist_json)
  38
  39         videos = []
  40         ns = '{http://tempuri.org/XMLSchema.xsd}'
  41         xpath = '%(ns)sPlaylist/%(ns)sItems/%(ns)sItem' % {'ns': ns}
  42         webpage_description = self._og_search_description(webpage)
  43         for (i, (item, info)) in enumerate(zip(flash_config.findall(xpath), playlist), 1):
  44             # Get best quality url
  45             rtmp_url = None
  46             for q in ['Q6A', 'Q4A', 'Q1A']:
  47                 video_url = find_xpath_attr(item, '%sVideoUrl' % ns, 'quality', q)
  48                 if video_url is not None:
  49                     rtmp_url = video_url.text
  50                     break
  51             if rtmp_url is None:
  52                 raise ExtractorError(u'Couldn\'t get video url: %s' % info['id'])
  53             description = self._html_search_regex(
  54                 r'id="playlist_entry_%s".*?<p>(.*?)</p>' % i, webpage,
  55                 u'description', default=webpage_description, flags=re.DOTALL)
  56             videos.append({
  57                 '_type': 'video',
  58                 'id': info['id'],
  59                 'title': info['title'],
  60                 'url': rtmp_url,
  61                 'ext': 'flv',
  62                 'description': description,
  63                 })
  64
  65         return videos