[youtube] Fix extraction.
[youtube-dl] / youtube_dl / extractor / meipai.py
1 # coding: utf-8
2 from __future__ import unicode_literals
3
4 from .common import InfoExtractor
5 from ..utils import (
6     int_or_none,
7     parse_duration,
8     unified_timestamp,
9 )
10
11
12 class MeipaiIE(InfoExtractor):
13     IE_DESC = '美拍'
14     _VALID_URL = r'https?://(?:www\.)?meipai\.com/media/(?P<id>[0-9]+)'
15     _TESTS = [{
16         # regular uploaded video
17         'url': 'http://www.meipai.com/media/531697625',
18         'md5': 'e3e9600f9e55a302daecc90825854b4f',
19         'info_dict': {
20             'id': '531697625',
21             'ext': 'mp4',
22             'title': '#葉子##阿桑##余姿昀##超級女聲#',
23             'description': '#葉子##阿桑##余姿昀##超級女聲#',
24             'thumbnail': r're:^https?://.*\.jpg$',
25             'duration': 152,
26             'timestamp': 1465492420,
27             'upload_date': '20160609',
28             'view_count': 35511,
29             'creator': '她她-TATA',
30             'tags': ['葉子', '阿桑', '余姿昀', '超級女聲'],
31         }
32     }, {
33         # record of live streaming
34         'url': 'http://www.meipai.com/media/585526361',
35         'md5': 'ff7d6afdbc6143342408223d4f5fb99a',
36         'info_dict': {
37             'id': '585526361',
38             'ext': 'mp4',
39             'title': '姿昀和善願 練歌練琴啦😁😁😁',
40             'description': '姿昀和善願 練歌練琴啦😁😁😁',
41             'thumbnail': r're:^https?://.*\.jpg$',
42             'duration': 5975,
43             'timestamp': 1474311799,
44             'upload_date': '20160919',
45             'view_count': 1215,
46             'creator': '她她-TATA',
47         }
48     }]
49
50     def _real_extract(self, url):
51         video_id = self._match_id(url)
52         webpage = self._download_webpage(url, video_id)
53
54         title = self._og_search_title(
55             webpage, default=None) or self._html_search_regex(
56             r'<title[^>]*>([^<]+)</title>', webpage, 'title')
57
58         formats = []
59
60         # recorded playback of live streaming
61         m3u8_url = self._html_search_regex(
62             r'file:\s*encodeURIComponent\((["\'])(?P<url>(?:(?!\1).)+)\1\)',
63             webpage, 'm3u8 url', group='url', default=None)
64         if m3u8_url:
65             formats.extend(self._extract_m3u8_formats(
66                 m3u8_url, video_id, 'mp4', entry_protocol='m3u8_native',
67                 m3u8_id='hls', fatal=False))
68
69         if not formats:
70             # regular uploaded video
71             video_url = self._search_regex(
72                 r'data-video=(["\'])(?P<url>(?:(?!\1).)+)\1', webpage, 'video url',
73                 group='url', default=None)
74             if video_url:
75                 formats.append({
76                     'url': video_url,
77                     'format_id': 'http',
78                 })
79
80         timestamp = unified_timestamp(self._og_search_property(
81             'video:release_date', webpage, 'release date', fatal=False))
82
83         tags = self._og_search_property(
84             'video:tag', webpage, 'tags', default='').split(',')
85
86         view_count = int_or_none(self._html_search_meta(
87             'interactionCount', webpage, 'view count'))
88         duration = parse_duration(self._html_search_meta(
89             'duration', webpage, 'duration'))
90         creator = self._og_search_property(
91             'video:director', webpage, 'creator', fatal=False)
92
93         return {
94             'id': video_id,
95             'title': title,
96             'description': self._og_search_description(webpage),
97             'thumbnail': self._og_search_thumbnail(webpage),
98             'duration': duration,
99             'timestamp': timestamp,
100             'view_count': view_count,
101             'creator': creator,
102             'tags': tags,
103             'formats': formats,
104         }