]> git.bitcoin.ninja Git - youtube-dl/blob - youtube_dl/extractor/ministrygrid.py
[brightcove:new] extract subtitles and strip video title
[youtube-dl] / youtube_dl / extractor / ministrygrid.py
1 from __future__ import unicode_literals
2
3 import json
4 import re
5
6 from .common import InfoExtractor
7 from ..utils import (
8     ExtractorError,
9     smuggle_url,
10 )
11
12
13 class MinistryGridIE(InfoExtractor):
14     _VALID_URL = r'https?://www\.ministrygrid.com/([^/?#]*/)*(?P<id>[^/#?]+)/?(?:$|[?#])'
15
16     _TEST = {
17         'url': 'http://www.ministrygrid.com/training-viewer/-/training/t4g-2014-conference/the-gospel-by-numbers-4/the-gospel-by-numbers',
18         'md5': '844be0d2a1340422759c2a9101bab017',
19         'info_dict': {
20             'id': '3453494717001',
21             'ext': 'mp4',
22             'title': 'The Gospel by Numbers',
23             'description': 'Coming soon from T4G 2014!',
24             'uploader': 'LifeWay Christian Resources (MG)',
25         },
26     }
27
28     def _real_extract(self, url):
29         mobj = re.match(self._VALID_URL, url)
30         video_id = mobj.group('id')
31
32         webpage = self._download_webpage(url, video_id)
33         portlets_json = self._search_regex(
34             r'Liferay\.Portlet\.list=(\[.+?\])', webpage, 'portlet list')
35         portlets = json.loads(portlets_json)
36         pl_id = self._search_regex(
37             r'<!--\s*p_l_id - ([0-9]+)<br>', webpage, 'p_l_id')
38
39         for i, portlet in enumerate(portlets):
40             portlet_url = 'http://www.ministrygrid.com/c/portal/render_portlet?p_l_id=%s&p_p_id=%s' % (pl_id, portlet)
41             portlet_code = self._download_webpage(
42                 portlet_url, video_id,
43                 note='Looking in portlet %s (%d/%d)' % (portlet, i + 1, len(portlets)),
44                 fatal=False)
45             video_iframe_url = self._search_regex(
46                 r'<iframe.*?src="([^"]+)"', portlet_code, 'video iframe',
47                 default=None)
48             if video_iframe_url:
49                 surl = smuggle_url(
50                     video_iframe_url, {'force_videoid': video_id})
51                 return {
52                     '_type': 'url',
53                     'id': video_id,
54                     'url': surl,
55                 }
56
57         raise ExtractorError('Could not find video iframe in any portlets')