[youtube] Fix extraction.
[youtube-dl] / youtube_dl / extractor / mwave.py
1 from __future__ import unicode_literals
2
3 from .common import InfoExtractor
4 from ..compat import compat_str
5 from ..utils import (
6     int_or_none,
7     parse_duration,
8 )
9
10
11 class MwaveIE(InfoExtractor):
12     _VALID_URL = r'https?://mwave\.interest\.me/(?:[^/]+/)?mnettv/videodetail\.m\?searchVideoDetailVO\.clip_id=(?P<id>[0-9]+)'
13     _URL_TEMPLATE = 'http://mwave.interest.me/mnettv/videodetail.m?searchVideoDetailVO.clip_id=%s'
14     _TESTS = [{
15         'url': 'http://mwave.interest.me/mnettv/videodetail.m?searchVideoDetailVO.clip_id=168859',
16         # md5 is unstable
17         'info_dict': {
18             'id': '168859',
19             'ext': 'flv',
20             'title': '[M COUNTDOWN] SISTAR - SHAKE IT',
21             'thumbnail': r're:^https?://.*\.jpg$',
22             'uploader': 'M COUNTDOWN',
23             'duration': 206,
24             'view_count': int,
25         }
26     }, {
27         'url': 'http://mwave.interest.me/en/mnettv/videodetail.m?searchVideoDetailVO.clip_id=176199',
28         'only_matching': True,
29     }]
30
31     def _real_extract(self, url):
32         video_id = self._match_id(url)
33
34         vod_info = self._download_json(
35             'http://mwave.interest.me/onair/vod_info.m?vodtype=CL&sectorid=&endinfo=Y&id=%s' % video_id,
36             video_id, 'Download vod JSON')
37
38         formats = []
39         for num, cdn_info in enumerate(vod_info['cdn']):
40             stream_url = cdn_info.get('url')
41             if not stream_url:
42                 continue
43             stream_name = cdn_info.get('name') or compat_str(num)
44             f4m_stream = self._download_json(
45                 stream_url, video_id,
46                 'Download %s stream JSON' % stream_name)
47             f4m_url = f4m_stream.get('fileurl')
48             if not f4m_url:
49                 continue
50             formats.extend(
51                 self._extract_f4m_formats(f4m_url + '&hdcore=3.0.3', video_id, f4m_id=stream_name))
52         self._sort_formats(formats)
53
54         return {
55             'id': video_id,
56             'title': vod_info['title'],
57             'thumbnail': vod_info.get('cover'),
58             'uploader': vod_info.get('program_title'),
59             'duration': parse_duration(vod_info.get('time')),
60             'view_count': int_or_none(vod_info.get('hit')),
61             'formats': formats,
62         }
63
64
65 class MwaveMeetGreetIE(InfoExtractor):
66     _VALID_URL = r'https?://mwave\.interest\.me/(?:[^/]+/)?meetgreet/view/(?P<id>\d+)'
67     _TESTS = [{
68         'url': 'http://mwave.interest.me/meetgreet/view/256',
69         'info_dict': {
70             'id': '173294',
71             'ext': 'flv',
72             'title': '[MEET&GREET] Park BoRam',
73             'thumbnail': r're:^https?://.*\.jpg$',
74             'uploader': 'Mwave',
75             'duration': 3634,
76             'view_count': int,
77         }
78     }, {
79         'url': 'http://mwave.interest.me/en/meetgreet/view/256',
80         'only_matching': True,
81     }]
82
83     def _real_extract(self, url):
84         video_id = self._match_id(url)
85         webpage = self._download_webpage(url, video_id)
86         clip_id = self._html_search_regex(
87             r'<iframe[^>]+src="/mnettv/ifr_clip\.m\?searchVideoDetailVO\.clip_id=(\d+)',
88             webpage, 'clip ID')
89         clip_url = MwaveIE._URL_TEMPLATE % clip_id
90         return self.url_result(clip_url, 'Mwave', clip_id)