]> git.bitcoin.ninja Git - youtube-dl/blob - youtube_dl/extractor/expotv.py
[brightcove:new] extract subtitles and strip video title
[youtube-dl] / youtube_dl / extractor / expotv.py
1 from __future__ import unicode_literals
2
3 import re
4
5 from .common import InfoExtractor
6 from ..utils import (
7     int_or_none,
8     unified_strdate,
9 )
10
11
12 class ExpoTVIE(InfoExtractor):
13     _VALID_URL = r'https?://www\.expotv\.com/videos/[^?#]*/(?P<id>[0-9]+)($|[?#])'
14     _TEST = {
15         'url': 'http://www.expotv.com/videos/reviews/1/24/LinneCardscom/17561',
16         'md5': '2985e6d7a392b2f7a05e0ca350fe41d0',
17         'info_dict': {
18             'id': '17561',
19             'ext': 'mp4',
20             'upload_date': '20060212',
21             'title': 'My Favorite Online Scrapbook Store',
22             'view_count': int,
23             'description': 'You\'ll find most everything you need at this virtual store front.',
24             'uploader': 'Anna T.',
25             'thumbnail': 're:^https?://.*\.jpg$',
26         }
27     }
28
29     def _real_extract(self, url):
30         mobj = re.match(self._VALID_URL, url)
31         video_id = mobj.group('id')
32
33         webpage = self._download_webpage(url, video_id)
34         player_key = self._search_regex(
35             r'<param name="playerKey" value="([^"]+)"', webpage, 'player key')
36         config = self._download_json(
37             'http://client.expotv.com/video/config/%s/%s' % (video_id, player_key),
38             video_id, 'Downloading video configuration')
39
40         formats = []
41         for fcfg in config['sources']:
42             media_url = fcfg.get('file')
43             if not media_url:
44                 continue
45             if fcfg.get('type') == 'm3u8':
46                 formats.extend(self._extract_m3u8_formats(
47                     media_url, video_id, 'mp4', entry_protocol='m3u8_native', m3u8_id='hls'))
48             else:
49                 formats.append({
50                     'url': media_url,
51                     'height': int_or_none(fcfg.get('height')),
52                     'format_id': fcfg.get('label'),
53                     'ext': self._search_regex(
54                         r'filename=.*\.([a-z0-9_A-Z]+)&', media_url,
55                         'file extension', default=None) or fcfg.get('type'),
56                 })
57         self._sort_formats(formats)
58
59         title = self._og_search_title(webpage)
60         description = self._og_search_description(webpage)
61         thumbnail = config.get('image')
62         view_count = int_or_none(self._search_regex(
63             r'<h5>Plays: ([0-9]+)</h5>', webpage, 'view counts'))
64         uploader = self._search_regex(
65             r'<div class="reviewer">\s*<img alt="([^"]+)"', webpage, 'uploader',
66             fatal=False)
67         upload_date = unified_strdate(self._search_regex(
68             r'<h5>Reviewed on ([0-9/.]+)</h5>', webpage, 'upload date',
69             fatal=False))
70
71         return {
72             'id': video_id,
73             'formats': formats,
74             'title': title,
75             'description': description,
76             'view_count': view_count,
77             'thumbnail': thumbnail,
78             'uploader': uploader,
79             'upload_date': upload_date,
80         }