]> Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/extractor/npo.py
Prepare for release.
[youtubedl] / youtube_dl / extractor / npo.py
1 from __future__ import unicode_literals
2
3 from .subtitles import SubtitlesInfoExtractor
4 from ..utils import (
5 fix_xml_ampersands,
6 parse_duration,
7 qualities,
8 strip_jsonp,
9 unified_strdate,
10 url_basename,
11 )
12
13
14 class NPOBaseIE(SubtitlesInfoExtractor):
15 def _get_token(self, video_id):
16 token_page = self._download_webpage(
17 'http://ida.omroep.nl/npoplayer/i.js',
18 video_id, note='Downloading token')
19 return self._search_regex(
20 r'npoplayer\.token = "(.+?)"', token_page, 'token')
21
22
23 class NPOIE(NPOBaseIE):
24 IE_NAME = 'npo.nl'
25 _VALID_URL = r'https?://www\.npo\.nl/[^/]+/[^/]+/(?P<id>[^/?]+)'
26
27 _TESTS = [
28 {
29 'url': 'http://www.npo.nl/nieuwsuur/22-06-2014/VPWON_1220719',
30 'md5': '4b3f9c429157ec4775f2c9cb7b911016',
31 'info_dict': {
32 'id': 'VPWON_1220719',
33 'ext': 'm4v',
34 'title': 'Nieuwsuur',
35 'description': 'Dagelijks tussen tien en elf: nieuws, sport en achtergronden.',
36 'upload_date': '20140622',
37 },
38 },
39 {
40 'url': 'http://www.npo.nl/de-mega-mike-mega-thomas-show/27-02-2009/VARA_101191800',
41 'md5': 'da50a5787dbfc1603c4ad80f31c5120b',
42 'info_dict': {
43 'id': 'VARA_101191800',
44 'ext': 'm4v',
45 'title': 'De Mega Mike & Mega Thomas show',
46 'description': 'md5:3b74c97fc9d6901d5a665aac0e5400f4',
47 'upload_date': '20090227',
48 'duration': 2400,
49 },
50 },
51 {
52 'url': 'http://www.npo.nl/tegenlicht/25-02-2013/VPWON_1169289',
53 'md5': 'f8065e4e5a7824068ed3c7e783178f2c',
54 'info_dict': {
55 'id': 'VPWON_1169289',
56 'ext': 'm4v',
57 'title': 'Tegenlicht',
58 'description': 'md5:d6476bceb17a8c103c76c3b708f05dd1',
59 'upload_date': '20130225',
60 'duration': 3000,
61 },
62 },
63 {
64 'url': 'http://www.npo.nl/de-nieuwe-mens-deel-1/21-07-2010/WO_VPRO_043706',
65 'info_dict': {
66 'id': 'WO_VPRO_043706',
67 'ext': 'wmv',
68 'title': 'De nieuwe mens - Deel 1',
69 'description': 'md5:518ae51ba1293ffb80d8d8ce90b74e4b',
70 'duration': 4680,
71 },
72 'params': {
73 # mplayer mms download
74 'skip_download': True,
75 }
76 },
77 # non asf in streams
78 {
79 'url': 'http://www.npo.nl/hoe-gaat-europa-verder-na-parijs/10-01-2015/WO_NOS_762771',
80 'md5': 'b3da13de374cbe2d5332a7e910bef97f',
81 'info_dict': {
82 'id': 'WO_NOS_762771',
83 'ext': 'mp4',
84 'title': 'Hoe gaat Europa verder na Parijs?',
85 },
86 },
87 ]
88
89 def _real_extract(self, url):
90 video_id = self._match_id(url)
91 return self._get_info(video_id)
92
93 def _get_info(self, video_id):
94 metadata = self._download_json(
95 'http://e.omroep.nl/metadata/aflevering/%s' % video_id,
96 video_id,
97 # We have to remove the javascript callback
98 transform_source=strip_jsonp,
99 )
100
101 token = self._get_token(video_id)
102
103 formats = []
104
105 pubopties = metadata.get('pubopties')
106 if pubopties:
107 quality = qualities(['adaptive', 'wmv_sb', 'h264_sb', 'wmv_bb', 'h264_bb', 'wvc1_std', 'h264_std'])
108 for format_id in pubopties:
109 format_info = self._download_json(
110 'http://ida.omroep.nl/odi/?prid=%s&puboptions=%s&adaptive=yes&token=%s'
111 % (video_id, format_id, token),
112 video_id, 'Downloading %s JSON' % format_id)
113 if format_info.get('error_code', 0) or format_info.get('errorcode', 0):
114 continue
115 streams = format_info.get('streams')
116 if streams:
117 video_info = self._download_json(
118 streams[0] + '&type=json',
119 video_id, 'Downloading %s stream JSON' % format_id)
120 else:
121 video_info = format_info
122 video_url = video_info.get('url')
123 if not video_url:
124 continue
125 if format_id == 'adaptive':
126 formats.extend(self._extract_m3u8_formats(video_url, video_id))
127 else:
128 formats.append({
129 'url': video_url,
130 'format_id': format_id,
131 'quality': quality(format_id),
132 })
133
134 streams = metadata.get('streams')
135 if streams:
136 for i, stream in enumerate(streams):
137 stream_url = stream.get('url')
138 if not stream_url:
139 continue
140 if '.asf' not in stream_url:
141 formats.append({
142 'url': stream_url,
143 'quality': stream.get('kwaliteit'),
144 })
145 continue
146 asx = self._download_xml(
147 stream_url, video_id,
148 'Downloading stream %d ASX playlist' % i,
149 transform_source=fix_xml_ampersands)
150 ref = asx.find('./ENTRY/Ref')
151 if ref is None:
152 continue
153 video_url = ref.get('href')
154 if not video_url:
155 continue
156 formats.append({
157 'url': video_url,
158 'ext': stream.get('formaat', 'asf'),
159 'quality': stream.get('kwaliteit'),
160 })
161
162 self._sort_formats(formats)
163
164 subtitles = {}
165 if metadata.get('tt888') == 'ja':
166 subtitles['nl'] = 'http://e.omroep.nl/tt888/%s' % video_id
167
168 if self._downloader.params.get('listsubtitles', False):
169 self._list_available_subtitles(video_id, subtitles)
170 return
171
172 subtitles = self.extract_subtitles(video_id, subtitles)
173
174 return {
175 'id': video_id,
176 'title': metadata['titel'],
177 'description': metadata['info'],
178 'thumbnail': metadata.get('images', [{'url': None}])[-1]['url'],
179 'upload_date': unified_strdate(metadata.get('gidsdatum')),
180 'duration': parse_duration(metadata.get('tijdsduur')),
181 'formats': formats,
182 'subtitles': subtitles,
183 }
184
185
186 class NPOLiveIE(NPOBaseIE):
187 IE_NAME = 'npo.nl:live'
188 _VALID_URL = r'https?://www\.npo\.nl/live/(?P<id>.+)'
189
190 _TEST = {
191 'url': 'http://www.npo.nl/live/npo-1',
192 'info_dict': {
193 'id': 'LI_NEDERLAND1_136692',
194 'display_id': 'npo-1',
195 'ext': 'mp4',
196 'title': 're:^Nederland 1 [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
197 'description': 'Livestream',
198 'is_live': True,
199 },
200 'params': {
201 'skip_download': True,
202 }
203 }
204
205 def _real_extract(self, url):
206 display_id = self._match_id(url)
207
208 webpage = self._download_webpage(url, display_id)
209
210 live_id = self._search_regex(
211 r'data-prid="([^"]+)"', webpage, 'live id')
212
213 metadata = self._download_json(
214 'http://e.omroep.nl/metadata/%s' % live_id,
215 display_id, transform_source=strip_jsonp)
216
217 token = self._get_token(display_id)
218
219 formats = []
220
221 streams = metadata.get('streams')
222 if streams:
223 for stream in streams:
224 stream_type = stream.get('type').lower()
225 if stream_type == 'ss':
226 continue
227 stream_info = self._download_json(
228 'http://ida.omroep.nl/aapi/?stream=%s&token=%s&type=jsonp'
229 % (stream.get('url'), token),
230 display_id, 'Downloading %s JSON' % stream_type)
231 if stream_info.get('error_code', 0) or stream_info.get('errorcode', 0):
232 continue
233 stream_url = self._download_json(
234 stream_info['stream'], display_id,
235 'Downloading %s URL' % stream_type,
236 transform_source=strip_jsonp)
237 if stream_type == 'hds':
238 f4m_formats = self._extract_f4m_formats(stream_url, display_id)
239 # f4m downloader downloads only piece of live stream
240 for f4m_format in f4m_formats:
241 f4m_format['preference'] = -1
242 formats.extend(f4m_formats)
243 elif stream_type == 'hls':
244 formats.extend(self._extract_m3u8_formats(stream_url, display_id, 'mp4'))
245 else:
246 formats.append({
247 'url': stream_url,
248 })
249
250 self._sort_formats(formats)
251
252 return {
253 'id': live_id,
254 'display_id': display_id,
255 'title': self._live_title(metadata['titel']),
256 'description': metadata['info'],
257 'thumbnail': metadata.get('images', [{'url': None}])[-1]['url'],
258 'formats': formats,
259 'is_live': True,
260 }
261
262
263 class TegenlichtVproIE(NPOIE):
264 IE_NAME = 'tegenlicht.vpro.nl'
265 _VALID_URL = r'https?://tegenlicht\.vpro\.nl/afleveringen/.*?'
266
267 _TESTS = [
268 {
269 'url': 'http://tegenlicht.vpro.nl/afleveringen/2012-2013/de-toekomst-komt-uit-afrika.html',
270 'md5': 'f8065e4e5a7824068ed3c7e783178f2c',
271 'info_dict': {
272 'id': 'VPWON_1169289',
273 'ext': 'm4v',
274 'title': 'Tegenlicht',
275 'description': 'md5:d6476bceb17a8c103c76c3b708f05dd1',
276 'upload_date': '20130225',
277 },
278 },
279 ]
280
281 def _real_extract(self, url):
282 name = url_basename(url)
283 webpage = self._download_webpage(url, name)
284 urn = self._html_search_meta('mediaurn', webpage)
285 info_page = self._download_json(
286 'http://rs.vpro.nl/v2/api/media/%s.json' % urn, name)
287 return self._get_info(info_page['mid'])