Add an extractor for Clipsyndicate (closes #1744)
[youtube-dl] / youtube_dl / extractor / clipsyndicate.py
1 import re
2 import xml.etree.ElementTree
3
4 from .common import InfoExtractor
5 from ..utils import (
6     find_xpath_attr,
7 )
8
9
10 class ClipsyndicateIE(InfoExtractor):
11     _VALID_URL = r'http://www\.clipsyndicate\.com/video/play(list/\d+)?/(?P<id>\d+)'
12
13     _TEST = {
14         u'url': u'http://www.clipsyndicate.com/video/play/4629301/brick_briscoe',
15         u'md5': u'4d7d549451bad625e0ff3d7bd56d776c',
16         u'info_dict': {
17             u'id': u'4629301',
18             u'ext': u'mp4',
19             u'title': u'Brick Briscoe',
20             u'duration': 612,
21         },
22     }
23
24     def _real_extract(self, url):
25         mobj = re.match(self._VALID_URL, url)
26         video_id = mobj.group('id')
27         js_player = self._download_webpage(
28             'http://eplayer.clipsyndicate.com/embed/player.js?va_id=%s' % video_id,
29             video_id, u'Downlaoding player')
30         # it includes a required token
31         flvars = self._search_regex(r'flvars: "(.*?)"', js_player, u'flvars')
32
33         playlist_page = self._download_webpage(
34             'http://eplayer.clipsyndicate.com/osmf/playlist?%s' % flvars,
35             video_id, u'Downloading video info') 
36         # Fix broken xml
37         playlist_page = re.sub('&', '&amp;', playlist_page)
38         pdoc = xml.etree.ElementTree.fromstring(playlist_page.encode('utf-8'))
39
40         track_doc = pdoc.find('trackList/track')
41         def find_param(name):
42             node = find_xpath_attr(track_doc, './/param', 'name', name)
43             if node is not None:
44                 return node.attrib['value']
45
46         return {
47             'id': video_id,
48             'title': find_param('title'),
49             'url': track_doc.find('location').text,
50             'thumbnail': find_param('thumbnail'),
51             'duration': int(find_param('duration')),
52         }