]> git.bitcoin.ninja Git - youtube-dl/blob - youtube_dl/extractor/mgtv.py
[pornhub] Extract categories and tags (closes #10499)
[youtube-dl] / youtube_dl / extractor / mgtv.py
1 # coding: utf-8
2 from __future__ import unicode_literals
3
4 from .common import InfoExtractor
5 from ..utils import int_or_none
6
7
8 class MGTVIE(InfoExtractor):
9     _VALID_URL = r'https?://www\.mgtv\.com/v/(?:[^/]+/)*(?P<id>\d+)\.html'
10     IE_DESC = '芒果TV'
11
12     _TESTS = [{
13         'url': 'http://www.mgtv.com/v/1/290525/f/3116640.html',
14         'md5': '1bdadcf760a0b90946ca68ee9a2db41a',
15         'info_dict': {
16             'id': '3116640',
17             'ext': 'mp4',
18             'title': '我是歌手第四季双年巅峰会:韩红李玟“双王”领军对抗',
19             'description': '我是歌手第四季双年巅峰会',
20             'duration': 7461,
21             'thumbnail': 're:^https?://.*\.jpg$',
22         },
23     }, {
24         # no tbr extracted from stream_url
25         'url': 'http://www.mgtv.com/v/1/1/f/3324755.html',
26         'only_matching': True,
27     }]
28
29     def _real_extract(self, url):
30         video_id = self._match_id(url)
31         api_data = self._download_json(
32             'http://v.api.mgtv.com/player/video', video_id,
33             query={'video_id': video_id},
34             headers=self.geo_verification_headers())['data']
35         info = api_data['info']
36
37         formats = []
38         for idx, stream in enumerate(api_data['stream']):
39             stream_url = stream.get('url')
40             if not stream_url:
41                 continue
42             tbr = int_or_none(self._search_regex(
43                 r'(\d+)\.mp4', stream_url, 'tbr', default=None))
44
45             def extract_format(stream_url, format_id, idx, query={}):
46                 format_info = self._download_json(
47                     stream_url, video_id,
48                     note='Download video info for format %s' % (format_id or '#%d' % idx),
49                     query=query)
50                 return {
51                     'format_id': format_id,
52                     'url': format_info['info'],
53                     'ext': 'mp4',
54                     'tbr': tbr,
55                 }
56
57             formats.append(extract_format(
58                 stream_url, 'hls-%d' % tbr if tbr else None, idx * 2))
59             formats.append(extract_format(stream_url.replace(
60                 '/playlist.m3u8', ''), 'http-%d' % tbr if tbr else None, idx * 2 + 1, {'pno': 1031}))
61         self._sort_formats(formats)
62
63         return {
64             'id': video_id,
65             'title': info['title'].strip(),
66             'formats': formats,
67             'description': info.get('desc'),
68             'duration': int_or_none(info.get('duration')),
69             'thumbnail': info.get('thumb'),
70         }