]> Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/extractor/canvas.py
New upstream version 2017.10.15.1
[youtubedl] / youtube_dl / extractor / canvas.py
1 from __future__ import unicode_literals
2
3 import re
4
5 from .common import InfoExtractor
6 from ..utils import (
7 float_or_none,
8 strip_or_none,
9 )
10
11
12 class CanvasIE(InfoExtractor):
13 _VALID_URL = r'https?://mediazone\.vrt\.be/api/v1/(?P<site_id>canvas|een|ketnet)/assets/(?P<id>m[dz]-ast-[^/?#&]+)'
14 _TESTS = [{
15 'url': 'https://mediazone.vrt.be/api/v1/ketnet/assets/md-ast-4ac54990-ce66-4d00-a8ca-9eac86f4c475',
16 'md5': '90139b746a0a9bd7bb631283f6e2a64e',
17 'info_dict': {
18 'id': 'md-ast-4ac54990-ce66-4d00-a8ca-9eac86f4c475',
19 'display_id': 'md-ast-4ac54990-ce66-4d00-a8ca-9eac86f4c475',
20 'ext': 'flv',
21 'title': 'Nachtwacht: De Greystook',
22 'description': 'md5:1db3f5dc4c7109c821261e7512975be7',
23 'thumbnail': r're:^https?://.*\.jpg$',
24 'duration': 1468.03,
25 },
26 'expected_warnings': ['is not a supported codec', 'Unknown MIME type'],
27 }, {
28 'url': 'https://mediazone.vrt.be/api/v1/canvas/assets/mz-ast-5e5f90b6-2d72-4c40-82c2-e134f884e93e',
29 'only_matching': True,
30 }]
31
32 def _real_extract(self, url):
33 mobj = re.match(self._VALID_URL, url)
34 site_id, video_id = mobj.group('site_id'), mobj.group('id')
35
36 data = self._download_json(
37 'https://mediazone.vrt.be/api/v1/%s/assets/%s'
38 % (site_id, video_id), video_id)
39
40 title = data['title']
41 description = data.get('description')
42
43 formats = []
44 for target in data['targetUrls']:
45 format_url, format_type = target.get('url'), target.get('type')
46 if not format_url or not format_type:
47 continue
48 if format_type == 'HLS':
49 formats.extend(self._extract_m3u8_formats(
50 format_url, video_id, 'mp4', entry_protocol='m3u8_native',
51 m3u8_id=format_type, fatal=False))
52 elif format_type == 'HDS':
53 formats.extend(self._extract_f4m_formats(
54 format_url, video_id, f4m_id=format_type, fatal=False))
55 elif format_type == 'MPEG_DASH':
56 formats.extend(self._extract_mpd_formats(
57 format_url, video_id, mpd_id=format_type, fatal=False))
58 elif format_type == 'HSS':
59 formats.extend(self._extract_ism_formats(
60 format_url, video_id, ism_id='mss', fatal=False))
61 else:
62 formats.append({
63 'format_id': format_type,
64 'url': format_url,
65 })
66 self._sort_formats(formats)
67
68 subtitles = {}
69 subtitle_urls = data.get('subtitleUrls')
70 if isinstance(subtitle_urls, list):
71 for subtitle in subtitle_urls:
72 subtitle_url = subtitle.get('url')
73 if subtitle_url and subtitle.get('type') == 'CLOSED':
74 subtitles.setdefault('nl', []).append({'url': subtitle_url})
75
76 return {
77 'id': video_id,
78 'display_id': video_id,
79 'title': title,
80 'description': description,
81 'formats': formats,
82 'duration': float_or_none(data.get('duration'), 1000),
83 'thumbnail': data.get('posterImageUrl'),
84 'subtitles': subtitles,
85 }
86
87
88 class CanvasEenIE(InfoExtractor):
89 IE_DESC = 'canvas.be and een.be'
90 _VALID_URL = r'https?://(?:www\.)?(?P<site_id>canvas|een)\.be/(?:[^/]+/)*(?P<id>[^/?#&]+)'
91 _TESTS = [{
92 'url': 'http://www.canvas.be/video/de-afspraak/najaar-2015/de-afspraak-veilt-voor-de-warmste-week',
93 'md5': 'ed66976748d12350b118455979cca293',
94 'info_dict': {
95 'id': 'mz-ast-5e5f90b6-2d72-4c40-82c2-e134f884e93e',
96 'display_id': 'de-afspraak-veilt-voor-de-warmste-week',
97 'ext': 'flv',
98 'title': 'De afspraak veilt voor de Warmste Week',
99 'description': 'md5:24cb860c320dc2be7358e0e5aa317ba6',
100 'thumbnail': r're:^https?://.*\.jpg$',
101 'duration': 49.02,
102 },
103 'expected_warnings': ['is not a supported codec'],
104 }, {
105 # with subtitles
106 'url': 'http://www.canvas.be/video/panorama/2016/pieter-0167',
107 'info_dict': {
108 'id': 'mz-ast-5240ff21-2d30-4101-bba6-92b5ec67c625',
109 'display_id': 'pieter-0167',
110 'ext': 'mp4',
111 'title': 'Pieter 0167',
112 'description': 'md5:943cd30f48a5d29ba02c3a104dc4ec4e',
113 'thumbnail': r're:^https?://.*\.jpg$',
114 'duration': 2553.08,
115 'subtitles': {
116 'nl': [{
117 'ext': 'vtt',
118 }],
119 },
120 },
121 'params': {
122 'skip_download': True,
123 },
124 'skip': 'Pagina niet gevonden',
125 }, {
126 'url': 'https://www.een.be/sorry-voor-alles/herbekijk-sorry-voor-alles',
127 'info_dict': {
128 'id': 'mz-ast-11a587f8-b921-4266-82e2-0bce3e80d07f',
129 'display_id': 'herbekijk-sorry-voor-alles',
130 'ext': 'mp4',
131 'title': 'Herbekijk Sorry voor alles',
132 'description': 'md5:8bb2805df8164e5eb95d6a7a29dc0dd3',
133 'thumbnail': r're:^https?://.*\.jpg$',
134 'duration': 3788.06,
135 },
136 'params': {
137 'skip_download': True,
138 },
139 'skip': 'Episode no longer available',
140 }, {
141 'url': 'https://www.canvas.be/check-point/najaar-2016/de-politie-uw-vriend',
142 'only_matching': True,
143 }]
144
145 def _real_extract(self, url):
146 mobj = re.match(self._VALID_URL, url)
147 site_id, display_id = mobj.group('site_id'), mobj.group('id')
148
149 webpage = self._download_webpage(url, display_id)
150
151 title = strip_or_none(self._search_regex(
152 r'<h1[^>]+class="video__body__header__title"[^>]*>(.+?)</h1>',
153 webpage, 'title', default=None) or self._og_search_title(
154 webpage, default=None))
155
156 video_id = self._html_search_regex(
157 r'data-video=(["\'])(?P<id>(?:(?!\1).)+)\1', webpage, 'video id',
158 group='id')
159
160 return {
161 '_type': 'url_transparent',
162 'url': 'https://mediazone.vrt.be/api/v1/%s/assets/%s' % (site_id, video_id),
163 'ie_key': CanvasIE.ie_key(),
164 'id': video_id,
165 'display_id': display_id,
166 'title': title,
167 'description': self._og_search_description(webpage),
168 }