]> gitweb @ CieloNegro.org - youtube-dl.git/blob - youtube_dl/extractor/bokecc.py
[cbs] extract smpte and vtt subtitles
[youtube-dl.git] / youtube_dl / extractor / bokecc.py
1 # coding: utf-8
2 from __future__ import unicode_literals
3
4 import re
5
6 from .common import InfoExtractor
7 from ..compat import compat_parse_qs
8 from ..utils import ExtractorError
9
10
11 class BokeCCBaseIE(InfoExtractor):
12     def _extract_bokecc_formats(self, webpage, video_id, format_id=None):
13         player_params_str = self._html_search_regex(
14             r'<(?:script|embed)[^>]+src="http://p\.bokecc\.com/player\?([^"]+)',
15             webpage, 'player params')
16
17         player_params = compat_parse_qs(player_params_str)
18
19         info_xml = self._download_xml(
20             'http://p.bokecc.com/servlet/playinfo?uid=%s&vid=%s&m=1' % (
21                 player_params['siteid'][0], player_params['vid'][0]), video_id)
22
23         formats = [{
24             'format_id': format_id,
25             'url': quality.find('./copy').attrib['playurl'],
26             'preference': int(quality.attrib['value']),
27         } for quality in info_xml.findall('./video/quality')]
28
29         self._sort_formats(formats)
30
31         return formats
32
33
34 class BokeCCIE(BokeCCBaseIE):
35     _IE_DESC = 'CC视频'
36     _VALID_URL = r'https?://union\.bokecc\.com/playvideo\.bo\?(?P<query>.*)'
37
38     _TESTS = [{
39         'url': 'http://union.bokecc.com/playvideo.bo?vid=E44D40C15E65EA30&uid=CD0C5D3C8614B28B',
40         'info_dict': {
41             'id': 'CD0C5D3C8614B28B_E44D40C15E65EA30',
42             'ext': 'flv',
43             'title': 'BokeCC Video',
44         },
45     }]
46
47     def _real_extract(self, url):
48         qs = compat_parse_qs(re.match(self._VALID_URL, url).group('query'))
49         if not qs.get('vid') or not qs.get('uid'):
50             raise ExtractorError('Invalid URL', expected=True)
51
52         video_id = '%s_%s' % (qs['uid'][0], qs['vid'][0])
53
54         webpage = self._download_webpage(url, video_id)
55
56         return {
57             'id': video_id,
58             'title': 'BokeCC Video',  # no title provided in the webpage
59             'formats': self._extract_bokecc_formats(webpage, video_id),
60         }