]> Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/extractor/orf.py
b42eae89aca1bdc894e29a876d06e4c5d49564a0
[youtubedl] / 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 def _real_extract(self, url):
18 mobj = re.match(self._VALID_URL, url)
19 playlist_id = mobj.group('id')
20 webpage = self._download_webpage(url, playlist_id)
21
22 flash_xml = self._search_regex('ORF.flashXML = \'(.+?)\'', webpage, u'flash xml')
23 flash_xml = compat_urlparse.parse_qs('xml='+flash_xml)['xml'][0]
24 flash_config = xml.etree.ElementTree.fromstring(flash_xml.encode('utf-8'))
25 playlist_json = self._search_regex(r'playlist\': \'(\[.*?\])\'', webpage, u'playlist').replace(r'\"','"')
26 playlist = json.loads(playlist_json)
27
28 videos = []
29 ns = '{http://tempuri.org/XMLSchema.xsd}'
30 xpath = '%(ns)sPlaylist/%(ns)sItems/%(ns)sItem' % {'ns': ns}
31 webpage_description = self._og_search_description(webpage)
32 for (i, (item, info)) in enumerate(zip(flash_config.findall(xpath), playlist), 1):
33 # Get best quality url
34 rtmp_url = None
35 for q in ['Q6A', 'Q4A', 'Q1A']:
36 video_url = find_xpath_attr(item, '%sVideoUrl' % ns, 'quality', q)
37 if video_url is not None:
38 rtmp_url = video_url.text
39 break
40 if rtmp_url is None:
41 raise ExtractorError(u'Couldn\'t get video url: %s' % info['id'])
42 description = self._html_search_regex(
43 r'id="playlist_entry_%s".*?<p>(.*?)</p>' % i, webpage,
44 u'description', default=webpage_description, flags=re.DOTALL)
45 videos.append({
46 '_type': 'video',
47 'id': info['id'],
48 'title': info['title'],
49 'url': rtmp_url,
50 'ext': 'flv',
51 'description': description,
52 })
53
54 return videos