]> git.bitcoin.ninja Git - youtube-dl/blob - youtube_dl/extractor/gamespot.py
[pornhub] Extract categories and tags (closes #10499)
[youtube-dl] / youtube_dl / extractor / gamespot.py
1 from __future__ import unicode_literals
2
3 import re
4
5 from .once import OnceIE
6 from ..compat import (
7     compat_urllib_parse_unquote,
8 )
9 from ..utils import (
10     unescapeHTML,
11     url_basename,
12     dict_get,
13 )
14
15
16 class GameSpotIE(OnceIE):
17     _VALID_URL = r'https?://(?:www\.)?gamespot\.com/.*-(?P<id>\d+)/?'
18     _TESTS = [{
19         'url': 'http://www.gamespot.com/videos/arma-3-community-guide-sitrep-i/2300-6410818/',
20         'md5': 'b2a30deaa8654fcccd43713a6b6a4825',
21         'info_dict': {
22             'id': 'gs-2300-6410818',
23             'ext': 'mp4',
24             'title': 'Arma 3 - Community Guide: SITREP I',
25             'description': 'Check out this video where some of the basics of Arma 3 is explained.',
26         },
27     }, {
28         'url': 'http://www.gamespot.com/videos/the-witcher-3-wild-hunt-xbox-one-now-playing/2300-6424837/',
29         'info_dict': {
30             'id': 'gs-2300-6424837',
31             'ext': 'mp4',
32             'title': 'Now Playing - The Witcher 3: Wild Hunt',
33             'description': 'Join us as we take a look at the early hours of The Witcher 3: Wild Hunt and more.',
34         },
35         'params': {
36             'skip_download': True,  # m3u8 downloads
37         },
38     }]
39
40     def _real_extract(self, url):
41         page_id = self._match_id(url)
42         webpage = self._download_webpage(url, page_id)
43         data_video_json = self._search_regex(
44             r'data-video=["\'](.*?)["\']', webpage, 'data video')
45         data_video = self._parse_json(unescapeHTML(data_video_json), page_id)
46         streams = data_video['videoStreams']
47
48         manifest_url = None
49         formats = []
50         f4m_url = streams.get('f4m_stream')
51         if f4m_url:
52             manifest_url = f4m_url
53             formats.extend(self._extract_f4m_formats(
54                 f4m_url + '?hdcore=3.7.0', page_id, f4m_id='hds', fatal=False))
55         m3u8_url = streams.get('m3u8_stream')
56         if m3u8_url:
57             manifest_url = m3u8_url
58             m3u8_formats = self._extract_m3u8_formats(
59                 m3u8_url, page_id, 'mp4', 'm3u8_native',
60                 m3u8_id='hls', fatal=False)
61             formats.extend(m3u8_formats)
62         progressive_url = dict_get(
63             streams, ('progressive_hd', 'progressive_high', 'progressive_low'))
64         if progressive_url and manifest_url:
65             qualities_basename = self._search_regex(
66                 '/([^/]+)\.csmil/',
67                 manifest_url, 'qualities basename', default=None)
68             if qualities_basename:
69                 QUALITIES_RE = r'((,\d+)+,?)'
70                 qualities = self._search_regex(
71                     QUALITIES_RE, qualities_basename,
72                     'qualities', default=None)
73                 if qualities:
74                     qualities = list(map(lambda q: int(q), qualities.strip(',').split(',')))
75                     qualities.sort()
76                     http_template = re.sub(QUALITIES_RE, r'%d', qualities_basename)
77                     http_url_basename = url_basename(progressive_url)
78                     if m3u8_formats:
79                         self._sort_formats(m3u8_formats)
80                         m3u8_formats = list(filter(
81                             lambda f: f.get('vcodec') != 'none' and f.get('resolution') != 'multiple',
82                             m3u8_formats))
83                     if len(qualities) == len(m3u8_formats):
84                         for q, m3u8_format in zip(qualities, m3u8_formats):
85                             f = m3u8_format.copy()
86                             f.update({
87                                 'url': progressive_url.replace(
88                                     http_url_basename, http_template % q),
89                                 'format_id': f['format_id'].replace('hls', 'http'),
90                                 'protocol': 'http',
91                             })
92                             formats.append(f)
93                     else:
94                         for q in qualities:
95                             formats.append({
96                                 'url': progressive_url.replace(
97                                     http_url_basename, http_template % q),
98                                 'ext': 'mp4',
99                                 'format_id': 'http-%d' % q,
100                                 'tbr': q,
101                             })
102
103         onceux_json = self._search_regex(
104             r'data-onceux-options=["\'](.*?)["\']', webpage, 'data video', default=None)
105         if onceux_json:
106             onceux_url = self._parse_json(unescapeHTML(onceux_json), page_id).get('metadataUri')
107             if onceux_url:
108                 formats.extend(self._extract_once_formats(re.sub(
109                     r'https?://[^/]+', 'http://once.unicornmedia.com', onceux_url).replace('ads/vmap/', '')))
110
111         if not formats:
112             for quality in ['sd', 'hd']:
113                 # It's actually a link to a flv file
114                 flv_url = streams.get('f4m_{0}'.format(quality))
115                 if flv_url is not None:
116                     formats.append({
117                         'url': flv_url,
118                         'ext': 'flv',
119                         'format_id': quality,
120                     })
121         self._sort_formats(formats)
122
123         return {
124             'id': data_video['guid'],
125             'display_id': page_id,
126             'title': compat_urllib_parse_unquote(data_video['title']),
127             'formats': formats,
128             'description': self._html_search_meta('description', webpage),
129             'thumbnail': self._og_search_thumbnail(webpage),
130         }