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

   1 # coding: utf-8
   2
   3 import re
   4 import xml.etree.ElementTree
   5 import json
   6
   7 from .common import InfoExtractor
   8 from ..utils import (
   9     compat_urlparse,
  10     ExtractorError,
  11     find_xpath_attr,
  12 )
  13
  14 class ORFIE(InfoExtractor):
  15     _VALID_URL = r'https?://tvthek.orf.at/(programs/.+?/episodes|topics/.+?)/(?P<id>\d+)'
  16
  17     _TEST = {
  18         u'url': u'http://tvthek.orf.at/programs/1171769-Wetter-ZIB/episodes/6557323-Wetter',
  19         u'file': u'6566957.flv',
  20         u'info_dict': {
  21             u'title': u'Wetter',
  22             u'description': u'Christa Kummer, Marcus Wadsak und Kollegen  präsentieren abwechselnd ihre täglichen Wetterprognosen für Österreich.\r \r Mehr Wetter unter wetter.ORF.at',
  23         },
  24         u'params': {
  25             # It uses rtmp
  26             u'skip_download': True,
  27         }
  28     }
  29
  30     def _real_extract(self, url):
  31         mobj = re.match(self._VALID_URL, url)
  32         playlist_id = mobj.group('id')
  33         webpage = self._download_webpage(url, playlist_id)
  34
  35         flash_xml = self._search_regex('ORF.flashXML = \'(.+?)\'', webpage, u'flash xml')
  36         flash_xml = compat_urlparse.parse_qs('xml='+flash_xml)['xml'][0]
  37         flash_config = xml.etree.ElementTree.fromstring(flash_xml.encode('utf-8'))
  38         playlist_json = self._search_regex(r'playlist\': \'(\[.*?\])\'', webpage, u'playlist').replace(r'\"','"')
  39         playlist = json.loads(playlist_json)
  40
  41         videos = []
  42         ns = '{http://tempuri.org/XMLSchema.xsd}'
  43         xpath = '%(ns)sPlaylist/%(ns)sItems/%(ns)sItem' % {'ns': ns}
  44         webpage_description = self._og_search_description(webpage)
  45         for (i, (item, info)) in enumerate(zip(flash_config.findall(xpath), playlist), 1):
  46             # Get best quality url
  47             rtmp_url = None
  48             for q in ['Q6A', 'Q4A', 'Q1A']:
  49                 video_url = find_xpath_attr(item, '%sVideoUrl' % ns, 'quality', q)
  50                 if video_url is not None:
  51                     rtmp_url = video_url.text
  52                     break
  53             if rtmp_url is None:
  54                 raise ExtractorError(u'Couldn\'t get video url: %s' % info['id'])
  55             description = self._html_search_regex(
  56                 r'id="playlist_entry_%s".*?<p>(.*?)</p>' % i, webpage,
  57                 u'description', default=webpage_description, flags=re.DOTALL)
  58             videos.append({
  59                 '_type': 'video',
  60                 'id': info['id'],
  61                 'title': info['title'],
  62                 'url': rtmp_url,
  63                 'ext': 'flv',
  64                 'description': description,
  65                 })
  66
  67         return videos