]>
Commit | Line | Data |
---|---|---|
1ef15636 | 1 | # coding: utf-8 |
2 | from __future__ import unicode_literals | |
3 | ||
4 | import re | |
5 | ||
6 | from .common import InfoExtractor | |
7 | from ..utils import ( | |
8 | ExtractorError, | |
ec5e77c5 | 9 | float_or_none, |
10 | int_or_none, | |
1ef15636 | 11 | parse_iso8601, |
12 | qualities, | |
ec5e77c5 | 13 | try_get, |
1ef15636 | 14 | ) |
15 | ||
16 | ||
17 | class SRGSSRIE(InfoExtractor): | |
ec5e77c5 | 18 | _VALID_URL = r'''(?x) |
19 | (?: | |
20 | https?://tp\.srgssr\.ch/p(?:/[^/]+)+\?urn=urn| | |
21 | srgssr | |
22 | ): | |
23 | (?P<bu> | |
24 | srf|rts|rsi|rtr|swi | |
25 | ):(?:[^:]+:)? | |
26 | (?P<type> | |
27 | video|audio | |
28 | ): | |
29 | (?P<id> | |
30 | [0-9a-f\-]{36}|\d+ | |
31 | ) | |
32 | ''' | |
4248dad9 S |
33 | _GEO_BYPASS = False |
34 | _GEO_COUNTRIES = ['CH'] | |
1ef15636 | 35 | |
36 | _ERRORS = { | |
37 | 'AGERATING12': 'To protect children under the age of 12, this video is only available between 8 p.m. and 6 a.m.', | |
38 | 'AGERATING18': 'To protect children under the age of 18, this video is only available between 11 p.m. and 5 a.m.', | |
7db2897d | 39 | # 'ENDDATE': 'For legal reasons, this video was only available for a specified period of time.', |
1ef15636 | 40 | 'GEOBLOCK': 'For legal reasons, this video is only available in Switzerland.', |
41 | 'LEGAL': 'The video cannot be transmitted for legal reasons.', | |
42 | 'STARTDATE': 'This video is not yet available. Please try again later.', | |
43 | } | |
ec5e77c5 | 44 | _DEFAULT_LANGUAGE_CODES = { |
45 | 'srf': 'de', | |
46 | 'rts': 'fr', | |
47 | 'rsi': 'it', | |
48 | 'rtr': 'rm', | |
49 | 'swi': 'en', | |
50 | } | |
1ef15636 | 51 | |
68601ef3 | 52 | def _get_tokenized_src(self, url, video_id, format_id): |
68601ef3 | 53 | token = self._download_json( |
ec5e77c5 | 54 | 'http://tp.srgssr.ch/akahd/token?acl=*', |
68601ef3 | 55 | video_id, 'Downloading %s token' % format_id, fatal=False) or {} |
ec5e77c5 | 56 | auth_params = try_get(token, lambda x: x['token']['authparams']) |
68601ef3 | 57 | if auth_params: |
ec5e77c5 | 58 | url += ('?' if '?' not in url else '&') + auth_params |
68601ef3 RA |
59 | return url |
60 | ||
ec5e77c5 | 61 | def _get_media_data(self, bu, media_type, media_id): |
62 | query = {'onlyChapters': True} if media_type == 'video' else {} | |
63 | full_media_data = self._download_json( | |
64 | 'https://il.srgssr.ch/integrationlayer/2.0/%s/mediaComposition/%s/%s.json' | |
65 | % (bu, media_type, media_id), | |
66 | media_id, query=query)['chapterList'] | |
67 | try: | |
68 | media_data = next( | |
69 | x for x in full_media_data if x.get('id') == media_id) | |
70 | except StopIteration: | |
71 | raise ExtractorError('No media information found') | |
72 | ||
73 | block_reason = media_data.get('blockReason') | |
74 | if block_reason and block_reason in self._ERRORS: | |
75 | message = self._ERRORS[block_reason] | |
76 | if block_reason == 'GEOBLOCK': | |
4248dad9 S |
77 | self.raise_geo_restricted( |
78 | msg=message, countries=self._GEO_COUNTRIES) | |
71631862 S |
79 | raise ExtractorError( |
80 | '%s said: %s' % (self.IE_NAME, message), expected=True) | |
7db2897d | 81 | |
82 | return media_data | |
83 | ||
84 | def _real_extract(self, url): | |
85 | bu, media_type, media_id = re.match(self._VALID_URL, url).groups() | |
ec5e77c5 | 86 | media_data = self._get_media_data(bu, media_type, media_id) |
87 | title = media_data['title'] | |
7db2897d | 88 | |
1ef15636 | 89 | formats = [] |
efe9dba5 | 90 | subtitles = {} |
ec5e77c5 | 91 | q = qualities(['SD', 'HD']) |
92 | for source in (media_data.get('resourceList') or []): | |
93 | format_url = source.get('url') | |
94 | if not format_url: | |
95 | continue | |
96 | protocol = source.get('protocol') | |
97 | quality = source.get('quality') | |
98 | format_id = [] | |
99 | for e in (protocol, source.get('encoding'), quality): | |
100 | if e: | |
101 | format_id.append(e) | |
102 | format_id = '-'.join(format_id) | |
103 | ||
104 | if protocol in ('HDS', 'HLS'): | |
105 | if source.get('tokenType') == 'AKAMAI': | |
106 | format_url = self._get_tokenized_src( | |
107 | format_url, media_id, format_id) | |
efe9dba5 F |
108 | fmts, subs = self._extract_akamai_formats_and_subtitles( |
109 | format_url, media_id) | |
110 | formats.extend(fmts) | |
111 | subtitles = self._merge_subtitles(subtitles, subs) | |
ec5e77c5 | 112 | elif protocol == 'HLS': |
efe9dba5 | 113 | m3u8_fmts, m3u8_subs = self._extract_m3u8_formats_and_subtitles( |
ec5e77c5 | 114 | format_url, media_id, 'mp4', 'm3u8_native', |
efe9dba5 F |
115 | m3u8_id=format_id, fatal=False) |
116 | formats.extend(m3u8_fmts) | |
117 | subtitles = self._merge_subtitles(subtitles, m3u8_subs) | |
ec5e77c5 | 118 | elif protocol in ('HTTP', 'HTTPS'): |
119 | formats.append({ | |
120 | 'format_id': format_id, | |
121 | 'url': format_url, | |
122 | 'quality': q(quality), | |
123 | }) | |
124 | ||
125 | # This is needed because for audio medias the podcast url is usually | |
126 | # always included, even if is only an audio segment and not the | |
127 | # whole episode. | |
128 | if int_or_none(media_data.get('position')) == 0: | |
129 | for p in ('S', 'H'): | |
130 | podcast_url = media_data.get('podcast%sdUrl' % p) | |
131 | if not podcast_url: | |
132 | continue | |
133 | quality = p + 'D' | |
134 | formats.append({ | |
135 | 'format_id': 'PODCAST-' + quality, | |
136 | 'url': podcast_url, | |
137 | 'quality': q(quality), | |
138 | }) | |
1ef15636 | 139 | self._sort_formats(formats) |
140 | ||
ec5e77c5 | 141 | if media_type == 'video': |
142 | for sub in (media_data.get('subtitleList') or []): | |
143 | sub_url = sub.get('url') | |
144 | if not sub_url: | |
145 | continue | |
146 | lang = sub.get('locale') or self._DEFAULT_LANGUAGE_CODES[bu] | |
147 | subtitles.setdefault(lang, []).append({ | |
148 | 'url': sub_url, | |
149 | }) | |
150 | ||
1ef15636 | 151 | return { |
152 | 'id': media_id, | |
153 | 'title': title, | |
ec5e77c5 | 154 | 'description': media_data.get('description'), |
155 | 'timestamp': parse_iso8601(media_data.get('date')), | |
156 | 'thumbnail': media_data.get('imageUrl'), | |
157 | 'duration': float_or_none(media_data.get('duration'), 1000), | |
158 | 'subtitles': subtitles, | |
1ef15636 | 159 | 'formats': formats, |
160 | } | |
161 | ||
162 | ||
163 | class SRGSSRPlayIE(InfoExtractor): | |
ce7d243c | 164 | IE_DESC = 'srf.ch, rts.ch, rsi.ch, rtr.ch and swissinfo.ch play sites' |
25b83c2a S |
165 | _VALID_URL = r'''(?x) |
166 | https?:// | |
167 | (?:(?:www|play)\.)? | |
168 | (?P<bu>srf|rts|rsi|rtr|swissinfo)\.ch/play/(?:tv|radio)/ | |
169 | (?: | |
170 | [^/]+/(?P<type>video|audio)/[^?]+| | |
171 | popup(?P<type_2>video|audio)player | |
172 | ) | |
06cd4cdb | 173 | \?.*?\b(?:id=|urn=urn:[^:]+:video:)(?P<id>[0-9a-f\-]{36}|\d+) |
25b83c2a | 174 | ''' |
1ef15636 | 175 | |
176 | _TESTS = [{ | |
177 | 'url': 'http://www.srf.ch/play/tv/10vor10/video/snowden-beantragt-asyl-in-russland?id=28e1a57d-5b76-4399-8ab3-9097f071e6c5', | |
ec5e77c5 | 178 | 'md5': '6db2226ba97f62ad42ce09783680046c', |
1ef15636 | 179 | 'info_dict': { |
180 | 'id': '28e1a57d-5b76-4399-8ab3-9097f071e6c5', | |
68601ef3 | 181 | 'ext': 'mp4', |
1ef15636 | 182 | 'upload_date': '20130701', |
183 | 'title': 'Snowden beantragt Asyl in Russland', | |
ec5e77c5 | 184 | 'timestamp': 1372708215, |
185 | 'duration': 113.827, | |
186 | 'thumbnail': r're:^https?://.*1383719781\.png$', | |
1ef15636 | 187 | }, |
ec5e77c5 | 188 | 'expected_warnings': ['Unable to download f4m manifest'], |
7db2897d | 189 | }, { |
e09f58b3 | 190 | 'url': 'http://www.rtr.ch/play/radio/actualitad/audio/saira-tujetsch-tuttina-cuntinuar-cun-sedrun-muster-turissem?id=63cb0778-27f8-49af-9284-8c7a8c6d15fc', |
191 | 'info_dict': { | |
192 | 'id': '63cb0778-27f8-49af-9284-8c7a8c6d15fc', | |
193 | 'ext': 'mp3', | |
194 | 'upload_date': '20151013', | |
195 | 'title': 'Saira: Tujetsch - tuttina cuntinuar cun Sedrun Mustér Turissem', | |
ec5e77c5 | 196 | 'timestamp': 1444709160, |
197 | 'duration': 336.816, | |
e09f58b3 | 198 | }, |
199 | 'params': { | |
200 | # rtmp download | |
201 | 'skip_download': True, | |
202 | }, | |
7db2897d | 203 | }, { |
204 | 'url': 'http://www.rts.ch/play/tv/-/video/le-19h30?id=6348260', | |
205 | 'md5': '67a2a9ae4e8e62a68d0e9820cc9782df', | |
206 | 'info_dict': { | |
207 | 'id': '6348260', | |
208 | 'display_id': '6348260', | |
422f7c11 | 209 | 'ext': 'mp4', |
ec5e77c5 | 210 | 'duration': 1796.76, |
7db2897d | 211 | 'title': 'Le 19h30', |
7db2897d | 212 | 'upload_date': '20141201', |
213 | 'timestamp': 1417458600, | |
ec85ded8 | 214 | 'thumbnail': r're:^https?://.*\.image', |
7db2897d | 215 | }, |
422f7c11 | 216 | 'params': { |
217 | # m3u8 download | |
218 | 'skip_download': True, | |
219 | } | |
ec5e77c5 | 220 | }, { |
221 | 'url': 'http://play.swissinfo.ch/play/tv/business/video/why-people-were-against-tax-reforms?id=42960270', | |
222 | 'info_dict': { | |
223 | 'id': '42960270', | |
224 | 'ext': 'mp4', | |
225 | 'title': 'Why people were against tax reforms', | |
226 | 'description': 'md5:7ac442c558e9630e947427469c4b824d', | |
227 | 'duration': 94.0, | |
228 | 'upload_date': '20170215', | |
229 | 'timestamp': 1487173560, | |
230 | 'thumbnail': r're:https?://www\.swissinfo\.ch/srgscalableimage/42961964', | |
231 | 'subtitles': 'count:9', | |
232 | }, | |
233 | 'params': { | |
234 | 'skip_download': True, | |
235 | } | |
3fe77472 SP |
236 | }, { |
237 | 'url': 'https://www.srf.ch/play/tv/popupvideoplayer?id=c4dba0ca-e75b-43b2-a34f-f708a4932e01', | |
25b83c2a | 238 | 'only_matching': True, |
06cd4cdb S |
239 | }, { |
240 | 'url': 'https://www.srf.ch/play/tv/10vor10/video/snowden-beantragt-asyl-in-russland?urn=urn:srf:video:28e1a57d-5b76-4399-8ab3-9097f071e6c5', | |
241 | 'only_matching': True, | |
242 | }, { | |
243 | 'url': 'https://www.rts.ch/play/tv/19h30/video/le-19h30?urn=urn:rts:video:6348260', | |
244 | 'only_matching': True, | |
ec5e77c5 | 245 | }, { |
246 | # audio segment, has podcastSdUrl of the full episode | |
247 | 'url': 'https://www.srf.ch/play/radio/popupaudioplayer?id=50b20dc8-f05b-4972-bf03-e438ff2833eb', | |
248 | 'only_matching': True, | |
1ef15636 | 249 | }] |
250 | ||
251 | def _real_extract(self, url): | |
25b83c2a S |
252 | mobj = re.match(self._VALID_URL, url) |
253 | bu = mobj.group('bu') | |
254 | media_type = mobj.group('type') or mobj.group('type_2') | |
255 | media_id = mobj.group('id') | |
e09f58b3 | 256 | return self.url_result('srgssr:%s:%s:%s' % (bu[:3], media_type, media_id), 'SRGSSR') |