]> Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/extractor/gamespot.py
d/p/remove-autoupdate-mechanism.patch: Update patch with metadata at the top.
[youtubedl] / 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/(?:video|article)s/(?:[^/]+/\d+-|embed/)(?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 'url': 'https://www.gamespot.com/videos/embed/6439218/',
40 'only_matching': True,
41 }, {
42 'url': 'https://www.gamespot.com/articles/the-last-of-us-2-receives-new-ps4-trailer/1100-6454469/',
43 'only_matching': True,
44 }]
45
46 def _real_extract(self, url):
47 page_id = self._match_id(url)
48 webpage = self._download_webpage(url, page_id)
49 data_video_json = self._search_regex(
50 r'data-video=["\'](.*?)["\']', webpage, 'data video')
51 data_video = self._parse_json(unescapeHTML(data_video_json), page_id)
52 streams = data_video['videoStreams']
53
54 manifest_url = None
55 formats = []
56 f4m_url = streams.get('f4m_stream')
57 if f4m_url:
58 manifest_url = f4m_url
59 formats.extend(self._extract_f4m_formats(
60 f4m_url + '?hdcore=3.7.0', page_id, f4m_id='hds', fatal=False))
61 m3u8_url = dict_get(streams, ('m3u8_stream', 'adaptive_stream'))
62 if m3u8_url:
63 manifest_url = m3u8_url
64 m3u8_formats = self._extract_m3u8_formats(
65 m3u8_url, page_id, 'mp4', 'm3u8_native',
66 m3u8_id='hls', fatal=False)
67 formats.extend(m3u8_formats)
68 progressive_url = dict_get(
69 streams, ('progressive_hd', 'progressive_high', 'progressive_low', 'other_lr'))
70 if progressive_url and manifest_url:
71 qualities_basename = self._search_regex(
72 r'/([^/]+)\.csmil/',
73 manifest_url, 'qualities basename', default=None)
74 if qualities_basename:
75 QUALITIES_RE = r'((,\d+)+,?)'
76 qualities = self._search_regex(
77 QUALITIES_RE, qualities_basename,
78 'qualities', default=None)
79 if qualities:
80 qualities = list(map(lambda q: int(q), qualities.strip(',').split(',')))
81 qualities.sort()
82 http_template = re.sub(QUALITIES_RE, r'%d', qualities_basename)
83 http_url_basename = url_basename(progressive_url)
84 if m3u8_formats:
85 self._sort_formats(m3u8_formats)
86 m3u8_formats = list(filter(
87 lambda f: f.get('vcodec') != 'none', m3u8_formats))
88 if len(qualities) == len(m3u8_formats):
89 for q, m3u8_format in zip(qualities, m3u8_formats):
90 f = m3u8_format.copy()
91 f.update({
92 'url': progressive_url.replace(
93 http_url_basename, http_template % q),
94 'format_id': f['format_id'].replace('hls', 'http'),
95 'protocol': 'http',
96 })
97 formats.append(f)
98 else:
99 for q in qualities:
100 formats.append({
101 'url': progressive_url.replace(
102 http_url_basename, http_template % q),
103 'ext': 'mp4',
104 'format_id': 'http-%d' % q,
105 'tbr': q,
106 })
107
108 onceux_json = self._search_regex(
109 r'data-onceux-options=["\'](.*?)["\']', webpage, 'data video', default=None)
110 if onceux_json:
111 onceux_url = self._parse_json(unescapeHTML(onceux_json), page_id).get('metadataUri')
112 if onceux_url:
113 formats.extend(self._extract_once_formats(re.sub(
114 r'https?://[^/]+', 'http://once.unicornmedia.com', onceux_url),
115 http_formats_preference=-1))
116
117 if not formats:
118 for quality in ['sd', 'hd']:
119 # It's actually a link to a flv file
120 flv_url = streams.get('f4m_{0}'.format(quality))
121 if flv_url is not None:
122 formats.append({
123 'url': flv_url,
124 'ext': 'flv',
125 'format_id': quality,
126 })
127 self._sort_formats(formats)
128
129 return {
130 'id': data_video['guid'],
131 'display_id': page_id,
132 'title': compat_urllib_parse_unquote(data_video['title']),
133 'formats': formats,
134 'description': self._html_search_meta('description', webpage),
135 'thumbnail': self._og_search_thumbnail(webpage),
136 }