[youtube] Fix extraction.
[youtube-dl] / youtube_dl / extractor / godtube.py
1 from __future__ import unicode_literals
2
3 import re
4
5 from .common import InfoExtractor
6 from ..utils import (
7     parse_duration,
8     parse_iso8601,
9 )
10
11
12 class GodTubeIE(InfoExtractor):
13     _VALID_URL = r'https?://(?:www\.)?godtube\.com/watch/\?v=(?P<id>[\da-zA-Z]+)'
14     _TESTS = [
15         {
16             'url': 'https://www.godtube.com/watch/?v=0C0CNNNU',
17             'md5': '77108c1e4ab58f48031101a1a2119789',
18             'info_dict': {
19                 'id': '0C0CNNNU',
20                 'ext': 'mp4',
21                 'title': 'Woman at the well.',
22                 'duration': 159,
23                 'timestamp': 1205712000,
24                 'uploader': 'beverlybmusic',
25                 'upload_date': '20080317',
26                 'thumbnail': r're:^https?://.*\.jpg$',
27             },
28         },
29     ]
30
31     def _real_extract(self, url):
32         mobj = re.match(self._VALID_URL, url)
33         video_id = mobj.group('id')
34
35         config = self._download_xml(
36             'http://www.godtube.com/resource/mediaplayer/%s.xml' % video_id.lower(),
37             video_id, 'Downloading player config XML')
38
39         video_url = config.find('file').text
40         uploader = config.find('author').text
41         timestamp = parse_iso8601(config.find('date').text)
42         duration = parse_duration(config.find('duration').text)
43         thumbnail = config.find('image').text
44
45         media = self._download_xml(
46             'http://www.godtube.com/media/xml/?v=%s' % video_id, video_id, 'Downloading media XML')
47
48         title = media.find('title').text
49
50         return {
51             'id': video_id,
52             'url': video_url,
53             'title': title,
54             'thumbnail': thumbnail,
55             'timestamp': timestamp,
56             'uploader': uploader,
57             'duration': duration,
58         }