]>
Commit | Line | Data |
---|---|---|
1a7dcca3 JJ |
1 | import functools |
2 | import json | |
3 | import time | |
4 | import urllib.error | |
5 | import urllib.parse | |
6 | ||
7 | from .gigya import GigyaBaseIE | |
ed6fb8b8 | 8 | from ..utils import ( |
1a7dcca3 JJ |
9 | ExtractorError, |
10 | clean_html, | |
82e91d20 | 11 | extract_attributes, |
ed6fb8b8 | 12 | float_or_none, |
82e91d20 | 13 | get_element_by_class, |
1a7dcca3 JJ |
14 | get_element_html_by_class, |
15 | int_or_none, | |
16 | join_nonempty, | |
17 | jwt_encode_hs256, | |
18 | make_archive_id, | |
19 | parse_age_limit, | |
20 | parse_iso8601, | |
21 | str_or_none, | |
82e91d20 | 22 | strip_or_none, |
1a7dcca3 JJ |
23 | traverse_obj, |
24 | url_or_none, | |
25 | urlencode_postdata, | |
ed6fb8b8 | 26 | ) |
911344e5 S |
27 | |
28 | ||
1a7dcca3 JJ |
29 | class VRTBaseIE(GigyaBaseIE): |
30 | _GEO_BYPASS = False | |
31 | _PLAYER_INFO = { | |
32 | 'platform': 'desktop', | |
33 | 'app': { | |
34 | 'type': 'browser', | |
35 | 'name': 'Chrome', | |
36 | }, | |
37 | 'device': 'undefined (undefined)', | |
38 | 'os': { | |
39 | 'name': 'Windows', | |
40 | 'version': 'x86_64' | |
41 | }, | |
42 | 'player': { | |
43 | 'name': 'VRT web player', | |
44 | 'version': '2.7.4-prod-2023-04-19T06:05:45' | |
45 | } | |
46 | } | |
47 | # From https://player.vrt.be/vrtnws/js/main.js & https://player.vrt.be/ketnet/js/main.fd1de01a40a1e3d842ea.js | |
48 | _JWT_KEY_ID = '0-0Fp51UZykfaiCJrfTE3+oMI8zvDteYfPtR+2n1R+z8w=' | |
49 | _JWT_SIGNING_KEY = '2a9251d782700769fb856da5725daf38661874ca6f80ae7dc2b05ec1a81a24ae' | |
50 | ||
51 | def _extract_formats_and_subtitles(self, data, video_id): | |
52 | if traverse_obj(data, 'drm'): | |
53 | self.report_drm(video_id) | |
54 | ||
55 | formats, subtitles = [], {} | |
56 | for target in traverse_obj(data, ('targetUrls', lambda _, v: url_or_none(v['url']) and v['type'])): | |
57 | format_type = target['type'].upper() | |
58 | format_url = target['url'] | |
59 | if format_type in ('HLS', 'HLS_AES'): | |
60 | fmts, subs = self._extract_m3u8_formats_and_subtitles( | |
61 | format_url, video_id, 'mp4', m3u8_id=format_type, fatal=False) | |
62 | formats.extend(fmts) | |
63 | self._merge_subtitles(subs, target=subtitles) | |
64 | elif format_type == 'HDS': | |
65 | formats.extend(self._extract_f4m_formats( | |
66 | format_url, video_id, f4m_id=format_type, fatal=False)) | |
67 | elif format_type == 'MPEG_DASH': | |
68 | fmts, subs = self._extract_mpd_formats_and_subtitles( | |
69 | format_url, video_id, mpd_id=format_type, fatal=False) | |
70 | formats.extend(fmts) | |
71 | self._merge_subtitles(subs, target=subtitles) | |
72 | elif format_type == 'HSS': | |
73 | fmts, subs = self._extract_ism_formats_and_subtitles( | |
74 | format_url, video_id, ism_id='mss', fatal=False) | |
75 | formats.extend(fmts) | |
76 | self._merge_subtitles(subs, target=subtitles) | |
77 | else: | |
78 | formats.append({ | |
79 | 'format_id': format_type, | |
80 | 'url': format_url, | |
81 | }) | |
82 | ||
83 | for sub in traverse_obj(data, ('subtitleUrls', lambda _, v: v['url'] and v['type'] == 'CLOSED')): | |
84 | subtitles.setdefault('nl', []).append({'url': sub['url']}) | |
85 | ||
86 | return formats, subtitles | |
87 | ||
88 | def _call_api(self, video_id, client='null', id_token=None, version='v2'): | |
89 | player_info = {'exp': (round(time.time(), 3) + 900), **self._PLAYER_INFO} | |
90 | player_token = self._download_json( | |
91 | 'https://media-services-public.vrt.be/vualto-video-aggregator-web/rest/external/v2/tokens', | |
92 | video_id, 'Downloading player token', headers={ | |
93 | **self.geo_verification_headers(), | |
94 | 'Content-Type': 'application/json', | |
95 | }, data=json.dumps({ | |
96 | 'identityToken': id_token or {}, | |
97 | 'playerInfo': jwt_encode_hs256(player_info, self._JWT_SIGNING_KEY, headers={ | |
98 | 'kid': self._JWT_KEY_ID | |
99 | }).decode() | |
100 | }, separators=(',', ':')).encode())['vrtPlayerToken'] | |
101 | ||
102 | return self._download_json( | |
103 | f'https://media-services-public.vrt.be/media-aggregator/{version}/media-items/{video_id}', | |
104 | video_id, 'Downloading API JSON', query={ | |
105 | 'vrtPlayerToken': player_token, | |
106 | 'client': client, | |
107 | }, expected_status=400) | |
108 | ||
109 | ||
110 | class VRTIE(VRTBaseIE): | |
82e91d20 RA |
111 | IE_DESC = 'VRT NWS, Flanders News, Flandern Info and Sporza' |
112 | _VALID_URL = r'https?://(?:www\.)?(?P<site>vrt\.be/vrtnws|sporza\.be)/[a-z]{2}/\d{4}/\d{2}/\d{2}/(?P<id>[^/?&#]+)' | |
113 | _TESTS = [{ | |
114 | 'url': 'https://www.vrt.be/vrtnws/nl/2019/05/15/beelden-van-binnenkant-notre-dame-een-maand-na-de-brand/', | |
82e91d20 RA |
115 | 'info_dict': { |
116 | 'id': 'pbs-pub-7855fc7b-1448-49bc-b073-316cb60caa71$vid-2ca50305-c38a-4762-9890-65cbd098b7bd', | |
117 | 'ext': 'mp4', | |
118 | 'title': 'Beelden van binnenkant Notre-Dame, één maand na de brand', | |
1a7dcca3 | 119 | 'description': 'md5:6fd85f999b2d1841aa5568f4bf02c3ff', |
82e91d20 | 120 | 'duration': 31.2, |
1a7dcca3 | 121 | 'thumbnail': 'https://images.vrt.be/orig/2019/05/15/2d914d61-7710-11e9-abcc-02b7b76bf47f.jpg', |
911344e5 | 122 | }, |
1a7dcca3 | 123 | 'params': {'skip_download': 'm3u8'}, |
82e91d20 RA |
124 | }, { |
125 | 'url': 'https://sporza.be/nl/2019/05/15/de-belgian-cats-zijn-klaar-voor-het-ek/', | |
82e91d20 RA |
126 | 'info_dict': { |
127 | 'id': 'pbs-pub-f2c86a46-8138-413a-a4b9-a0015a16ce2c$vid-1f112b31-e58e-4379-908d-aca6d80f8818', | |
128 | 'ext': 'mp4', | |
1a7dcca3 JJ |
129 | 'title': 'De Belgian Cats zijn klaar voor het EK', |
130 | 'description': 'Video: De Belgian Cats zijn klaar voor het EK mét Ann Wauters | basketbal, sport in het journaal', | |
82e91d20 | 131 | 'duration': 115.17, |
1a7dcca3 | 132 | 'thumbnail': 'https://images.vrt.be/orig/2019/05/15/11c0dba3-770e-11e9-abcc-02b7b76bf47f.jpg', |
911344e5 | 133 | }, |
1a7dcca3 | 134 | 'params': {'skip_download': 'm3u8'}, |
82e91d20 RA |
135 | }] |
136 | _CLIENT_MAP = { | |
137 | 'vrt.be/vrtnws': 'vrtnieuws', | |
138 | 'sporza.be': 'sporza', | |
139 | } | |
911344e5 S |
140 | |
141 | def _real_extract(self, url): | |
5ad28e7f | 142 | site, display_id = self._match_valid_url(url).groups() |
82e91d20 | 143 | webpage = self._download_webpage(url, display_id) |
1a7dcca3 | 144 | attrs = extract_attributes(get_element_html_by_class('vrtvideo', webpage) or '') |
82e91d20 | 145 | |
1a7dcca3 JJ |
146 | asset_id = attrs.get('data-video-id') or attrs['data-videoid'] |
147 | publication_id = traverse_obj(attrs, 'data-publication-id', 'data-publicationid') | |
82e91d20 | 148 | if publication_id: |
1a7dcca3 JJ |
149 | asset_id = f'{publication_id}${asset_id}' |
150 | client = traverse_obj(attrs, 'data-client-code', 'data-client') or self._CLIENT_MAP[site] | |
151 | ||
152 | data = self._call_api(asset_id, client) | |
153 | formats, subtitles = self._extract_formats_and_subtitles(data, asset_id) | |
82e91d20 | 154 | |
82e91d20 RA |
155 | description = self._html_search_meta( |
156 | ['og:description', 'twitter:description', 'description'], webpage) | |
157 | if description == '…': | |
158 | description = None | |
911344e5 S |
159 | |
160 | return { | |
82e91d20 | 161 | 'id': asset_id, |
1a7dcca3 JJ |
162 | 'formats': formats, |
163 | 'subtitles': subtitles, | |
911344e5 | 164 | 'description': description, |
1a7dcca3 | 165 | 'thumbnail': url_or_none(attrs.get('data-posterimage')), |
82e91d20 | 166 | 'duration': float_or_none(attrs.get('data-duration'), 1000), |
1a7dcca3 JJ |
167 | '_old_archive_ids': [make_archive_id('Canvas', asset_id)], |
168 | **traverse_obj(data, { | |
169 | 'title': ('title', {str}), | |
170 | 'description': ('shortDescription', {str}), | |
171 | 'duration': ('duration', {functools.partial(float_or_none, scale=1000)}), | |
172 | 'thumbnail': ('posterImageUrl', {url_or_none}), | |
173 | }), | |
174 | } | |
175 | ||
176 | ||
177 | class VrtNUIE(VRTBaseIE): | |
178 | IE_DESC = 'VRT MAX' | |
179 | _VALID_URL = r'https?://(?:www\.)?vrt\.be/vrtnu/a-z/(?:[^/]+/){2}(?P<id>[^/?#&]+)' | |
180 | _TESTS = [{ | |
181 | # CONTENT_IS_AGE_RESTRICTED | |
182 | 'url': 'https://www.vrt.be/vrtnu/a-z/de-ideale-wereld/2023-vj/de-ideale-wereld-d20230116/', | |
183 | 'info_dict': { | |
184 | 'id': 'pbs-pub-855b00a8-6ce2-4032-ac4f-1fcf3ae78524$vid-d2243aa1-ec46-4e34-a55b-92568459906f', | |
185 | 'ext': 'mp4', | |
186 | 'title': 'Tom Waes', | |
187 | 'description': 'Satirisch actualiteitenmagazine met Ella Leyers. Tom Waes is te gast.', | |
188 | 'timestamp': 1673905125, | |
189 | 'release_timestamp': 1673905125, | |
190 | 'series': 'De ideale wereld', | |
191 | 'season_id': '1672830988794', | |
192 | 'episode': 'Aflevering 1', | |
193 | 'episode_number': 1, | |
194 | 'episode_id': '1672830988861', | |
195 | 'display_id': 'de-ideale-wereld-d20230116', | |
196 | 'channel': 'VRT', | |
197 | 'duration': 1939.0, | |
198 | 'thumbnail': 'https://images.vrt.be/orig/2023/01/10/1bb39cb3-9115-11ed-b07d-02b7b76bf47f.jpg', | |
199 | 'release_date': '20230116', | |
200 | 'upload_date': '20230116', | |
201 | 'age_limit': 12, | |
202 | }, | |
203 | }, { | |
204 | 'url': 'https://www.vrt.be/vrtnu/a-z/buurman--wat-doet-u-nu-/6/buurman--wat-doet-u-nu--s6-trailer/', | |
205 | 'info_dict': { | |
206 | 'id': 'pbs-pub-ad4050eb-d9e5-48c2-9ec8-b6c355032361$vid-0465537a-34a8-4617-8352-4d8d983b4eee', | |
207 | 'ext': 'mp4', | |
208 | 'title': 'Trailer seizoen 6 \'Buurman, wat doet u nu?\'', | |
209 | 'description': 'md5:197424726c61384b4e5c519f16c0cf02', | |
210 | 'timestamp': 1652940000, | |
211 | 'release_timestamp': 1652940000, | |
212 | 'series': 'Buurman, wat doet u nu?', | |
213 | 'season': 'Seizoen 6', | |
214 | 'season_number': 6, | |
215 | 'season_id': '1652344200907', | |
216 | 'episode': 'Aflevering 0', | |
217 | 'episode_number': 0, | |
218 | 'episode_id': '1652951873524', | |
219 | 'display_id': 'buurman--wat-doet-u-nu--s6-trailer', | |
220 | 'channel': 'VRT', | |
221 | 'duration': 33.13, | |
222 | 'thumbnail': 'https://images.vrt.be/orig/2022/05/23/3c234d21-da83-11ec-b07d-02b7b76bf47f.jpg', | |
223 | 'release_date': '20220519', | |
224 | 'upload_date': '20220519', | |
225 | }, | |
226 | 'params': {'skip_download': 'm3u8'}, | |
227 | }] | |
228 | _NETRC_MACHINE = 'vrtnu' | |
229 | _authenticated = False | |
230 | ||
231 | def _perform_login(self, username, password): | |
232 | auth_info = self._gigya_login({ | |
233 | 'APIKey': '3_0Z2HujMtiWq_pkAjgnS2Md2E11a1AwZjYiBETtwNE-EoEHDINgtnvcAOpNgmrVGy', | |
234 | 'targetEnv': 'jssdk', | |
235 | 'loginID': username, | |
236 | 'password': password, | |
237 | 'authMode': 'cookie', | |
238 | }) | |
239 | ||
240 | if auth_info.get('errorDetails'): | |
241 | raise ExtractorError(f'Unable to login. VrtNU said: {auth_info["errorDetails"]}', expected=True) | |
242 | ||
243 | # Sometimes authentication fails for no good reason, retry | |
244 | for retry in self.RetryManager(): | |
245 | if retry.attempt > 1: | |
246 | self._sleep(1, None) | |
247 | try: | |
248 | self._request_webpage( | |
249 | 'https://token.vrt.be/vrtnuinitlogin', None, note='Requesting XSRF Token', | |
250 | errnote='Could not get XSRF Token', query={ | |
251 | 'provider': 'site', | |
252 | 'destination': 'https://www.vrt.be/vrtnu/', | |
253 | }) | |
254 | self._request_webpage( | |
255 | 'https://login.vrt.be/perform_login', None, | |
256 | note='Performing login', errnote='Login failed', | |
257 | query={'client_id': 'vrtnu-site'}, data=urlencode_postdata({ | |
258 | 'UID': auth_info['UID'], | |
259 | 'UIDSignature': auth_info['UIDSignature'], | |
260 | 'signatureTimestamp': auth_info['signatureTimestamp'], | |
261 | '_csrf': self._get_cookies('https://login.vrt.be').get('OIDCXSRF').value, | |
262 | })) | |
263 | except ExtractorError as e: | |
264 | if isinstance(e.cause, urllib.error.HTTPError) and e.cause.code == 401: | |
265 | retry.error = e | |
266 | continue | |
267 | raise | |
268 | ||
269 | self._authenticated = True | |
270 | ||
271 | def _real_extract(self, url): | |
272 | display_id = self._match_id(url) | |
273 | parsed_url = urllib.parse.urlparse(url) | |
274 | details = self._download_json( | |
275 | f'{parsed_url.scheme}://{parsed_url.netloc}{parsed_url.path.rstrip("/")}.model.json', | |
276 | display_id, 'Downloading asset JSON', 'Unable to download asset JSON')['details'] | |
277 | ||
278 | watch_info = traverse_obj(details, ( | |
279 | 'actions', lambda _, v: v['type'] == 'watch-episode', {dict}), get_all=False) or {} | |
280 | video_id = join_nonempty( | |
281 | 'episodePublicationId', 'episodeVideoId', delim='$', from_dict=watch_info) | |
282 | if '$' not in video_id: | |
283 | raise ExtractorError('Unable to extract video ID') | |
284 | ||
285 | vrtnutoken = self._download_json( | |
286 | 'https://token.vrt.be/refreshtoken', video_id, note='Retrieving vrtnutoken', | |
287 | errnote='Token refresh failed')['vrtnutoken'] if self._authenticated else None | |
288 | ||
289 | video_info = self._call_api(video_id, 'vrtnu-web@PROD', vrtnutoken) | |
290 | ||
291 | if 'title' not in video_info: | |
292 | code = video_info.get('code') | |
293 | if code in ('AUTHENTICATION_REQUIRED', 'CONTENT_IS_AGE_RESTRICTED'): | |
294 | self.raise_login_required(code, method='password') | |
295 | elif code in ('INVALID_LOCATION', 'CONTENT_AVAILABLE_ONLY_IN_BE'): | |
296 | self.raise_geo_restricted(countries=['BE']) | |
297 | elif code == 'CONTENT_AVAILABLE_ONLY_FOR_BE_RESIDENTS_AND_EXPATS': | |
298 | if not self._authenticated: | |
299 | self.raise_login_required(code, method='password') | |
300 | self.raise_geo_restricted(countries=['BE']) | |
301 | raise ExtractorError(code, expected=True) | |
302 | ||
303 | formats, subtitles = self._extract_formats_and_subtitles(video_info, video_id) | |
304 | ||
305 | return { | |
306 | **traverse_obj(details, { | |
307 | 'title': 'title', | |
308 | 'description': ('description', {clean_html}), | |
309 | 'timestamp': ('data', 'episode', 'onTime', 'raw', {parse_iso8601}), | |
310 | 'release_timestamp': ('data', 'episode', 'onTime', 'raw', {parse_iso8601}), | |
311 | 'series': ('data', 'program', 'title'), | |
312 | 'season': ('data', 'season', 'title', 'value'), | |
313 | 'season_number': ('data', 'season', 'title', 'raw', {int_or_none}), | |
314 | 'season_id': ('data', 'season', 'id', {str_or_none}), | |
315 | 'episode': ('data', 'episode', 'number', 'value', {str_or_none}), | |
316 | 'episode_number': ('data', 'episode', 'number', 'raw', {int_or_none}), | |
317 | 'episode_id': ('data', 'episode', 'id', {str_or_none}), | |
318 | 'age_limit': ('data', 'episode', 'age', 'raw', {parse_age_limit}), | |
319 | }), | |
320 | 'id': video_id, | |
321 | 'display_id': display_id, | |
322 | 'channel': 'VRT', | |
323 | 'formats': formats, | |
324 | 'duration': float_or_none(video_info.get('duration'), 1000), | |
325 | 'thumbnail': url_or_none(video_info.get('posterImageUrl')), | |
326 | 'subtitles': subtitles, | |
327 | '_old_archive_ids': [make_archive_id('Canvas', video_id)], | |
328 | } | |
329 | ||
330 | ||
331 | class KetnetIE(VRTBaseIE): | |
332 | _VALID_URL = r'https?://(?:www\.)?ketnet\.be/(?P<id>(?:[^/]+/)*[^/?#&]+)' | |
333 | _TESTS = [{ | |
334 | 'url': 'https://www.ketnet.be/kijken/m/meisjes/6/meisjes-s6a5', | |
335 | 'info_dict': { | |
336 | 'id': 'pbs-pub-39f8351c-a0a0-43e6-8394-205d597d6162$vid-5e306921-a9aa-4fa9-9f39-5b82c8f1028e', | |
337 | 'ext': 'mp4', | |
338 | 'title': 'Meisjes', | |
339 | 'episode': 'Reeks 6: Week 5', | |
340 | 'season': 'Reeks 6', | |
341 | 'series': 'Meisjes', | |
342 | 'timestamp': 1685251800, | |
343 | 'upload_date': '20230528', | |
344 | }, | |
345 | 'params': {'skip_download': 'm3u8'}, | |
346 | }] | |
347 | ||
348 | def _real_extract(self, url): | |
349 | display_id = self._match_id(url) | |
350 | ||
351 | video = self._download_json( | |
352 | 'https://senior-bff.ketnet.be/graphql', display_id, query={ | |
353 | 'query': '''{ | |
354 | video(id: "content/ketnet/nl/%s.model.json") { | |
355 | description | |
356 | episodeNr | |
357 | imageUrl | |
358 | mediaReference | |
359 | programTitle | |
360 | publicationDate | |
361 | seasonTitle | |
362 | subtitleVideodetail | |
363 | titleVideodetail | |
364 | } | |
365 | }''' % display_id, | |
366 | })['data']['video'] | |
367 | ||
368 | video_id = urllib.parse.unquote(video['mediaReference']) | |
369 | data = self._call_api(video_id, 'ketnet@PROD', version='v1') | |
370 | formats, subtitles = self._extract_formats_and_subtitles(data, video_id) | |
371 | ||
372 | return { | |
373 | 'id': video_id, | |
374 | 'formats': formats, | |
375 | 'subtitles': subtitles, | |
376 | '_old_archive_ids': [make_archive_id('Canvas', video_id)], | |
377 | **traverse_obj(video, { | |
378 | 'title': ('titleVideodetail', {str}), | |
379 | 'description': ('description', {str}), | |
380 | 'thumbnail': ('thumbnail', {url_or_none}), | |
381 | 'timestamp': ('publicationDate', {parse_iso8601}), | |
382 | 'series': ('programTitle', {str}), | |
383 | 'season': ('seasonTitle', {str}), | |
384 | 'episode': ('subtitleVideodetail', {str}), | |
385 | 'episode_number': ('episodeNr', {int_or_none}), | |
386 | }), | |
387 | } | |
388 | ||
389 | ||
390 | class DagelijkseKostIE(VRTBaseIE): | |
391 | IE_DESC = 'dagelijksekost.een.be' | |
392 | _VALID_URL = r'https?://dagelijksekost\.een\.be/gerechten/(?P<id>[^/?#&]+)' | |
393 | _TESTS = [{ | |
394 | 'url': 'https://dagelijksekost.een.be/gerechten/hachis-parmentier-met-witloof', | |
395 | 'info_dict': { | |
396 | 'id': 'md-ast-27a4d1ff-7d7b-425e-b84f-a4d227f592fa', | |
397 | 'ext': 'mp4', | |
398 | 'title': 'Hachis parmentier met witloof', | |
399 | 'description': 'md5:9960478392d87f63567b5b117688cdc5', | |
400 | 'display_id': 'hachis-parmentier-met-witloof', | |
401 | }, | |
402 | 'params': {'skip_download': 'm3u8'}, | |
403 | }] | |
404 | ||
405 | def _real_extract(self, url): | |
406 | display_id = self._match_id(url) | |
407 | webpage = self._download_webpage(url, display_id) | |
408 | video_id = self._html_search_regex( | |
409 | r'data-url=(["\'])(?P<id>(?:(?!\1).)+)\1', webpage, 'video id', group='id') | |
410 | ||
411 | data = self._call_api(video_id, 'dako@prod', version='v1') | |
412 | formats, subtitles = self._extract_formats_and_subtitles(data, video_id) | |
413 | ||
414 | return { | |
415 | 'id': video_id, | |
416 | 'formats': formats, | |
417 | 'subtitles': subtitles, | |
418 | 'display_id': display_id, | |
419 | 'title': strip_or_none(get_element_by_class( | |
420 | 'dish-metadata__title', webpage) or self._html_search_meta('twitter:title', webpage)), | |
421 | 'description': clean_html(get_element_by_class( | |
422 | 'dish-description', webpage)) or self._html_search_meta( | |
423 | ['description', 'twitter:description', 'og:description'], webpage), | |
424 | '_old_archive_ids': [make_archive_id('Canvas', video_id)], | |
5f6a1245 | 425 | } |