]> jfr.im git - yt-dlp.git/blob - yt_dlp/extractor/tvp.py
[extractor] Deprecate `_sort_formats`
[yt-dlp.git] / yt_dlp / extractor / tvp.py
1 import itertools
2 import random
3 import re
4
5 from .common import InfoExtractor
6 from ..utils import (
7 clean_html,
8 determine_ext,
9 dict_get,
10 ExtractorError,
11 int_or_none,
12 js_to_json,
13 str_or_none,
14 strip_or_none,
15 traverse_obj,
16 try_get,
17 url_or_none,
18 )
19
20
21 class TVPIE(InfoExtractor):
22 IE_NAME = 'tvp'
23 IE_DESC = 'Telewizja Polska'
24 _VALID_URL = r'https?://(?:[^/]+\.)?(?:tvp(?:parlament)?\.(?:pl|info)|tvpworld\.com|swipeto\.pl)/(?:(?!\d+/)[^/]+/)*(?P<id>\d+)'
25
26 _TESTS = [{
27 # TVPlayer 2 in js wrapper
28 'url': 'https://swipeto.pl/64095316/uliczny-foxtrot-wypozyczalnia-kaset-kto-pamieta-dvdvideo',
29 'info_dict': {
30 'id': '64095316',
31 'ext': 'mp4',
32 'title': 'Uliczny Foxtrot — Wypożyczalnia kaset. Kto pamięta DVD-Video?',
33 'age_limit': 0,
34 'duration': 374,
35 'thumbnail': r're:https://.+',
36 },
37 'expected_warnings': [
38 'Failed to download ISM manifest: HTTP Error 404: Not Found',
39 'Failed to download m3u8 information: HTTP Error 404: Not Found',
40 ],
41 }, {
42 # TVPlayer legacy
43 'url': 'https://www.tvp.pl/polska-press-video-uploader/wideo/62042351',
44 'info_dict': {
45 'id': '62042351',
46 'ext': 'mp4',
47 'title': 'Wideo',
48 'description': 'Wideo Kamera',
49 'duration': 24,
50 'age_limit': 0,
51 'thumbnail': r're:https://.+',
52 },
53 }, {
54 # TVPlayer 2 in iframe
55 'url': 'https://wiadomosci.tvp.pl/50725617/dzieci-na-sprzedaz-dla-homoseksualistow',
56 'info_dict': {
57 'id': '50725617',
58 'ext': 'mp4',
59 'title': 'Dzieci na sprzedaż dla homoseksualistów',
60 'description': 'md5:7d318eef04e55ddd9f87a8488ac7d590',
61 'age_limit': 12,
62 'duration': 259,
63 'thumbnail': r're:https://.+',
64 },
65 }, {
66 # TVPlayer 2 in client-side rendered website (regional; window.__newsData)
67 'url': 'https://warszawa.tvp.pl/25804446/studio-yayo',
68 'info_dict': {
69 'id': '25804446',
70 'ext': 'mp4',
71 'title': 'Studio Yayo',
72 'upload_date': '20160616',
73 'timestamp': 1466075700,
74 'age_limit': 0,
75 'duration': 20,
76 'thumbnail': r're:https://.+',
77 },
78 'skip': 'Geo-blocked outside PL',
79 }, {
80 # TVPlayer 2 in client-side rendered website (tvp.info; window.__videoData)
81 'url': 'https://www.tvp.info/52880236/09042021-0800',
82 'info_dict': {
83 'id': '52880236',
84 'ext': 'mp4',
85 'title': '09.04.2021, 08:00',
86 'age_limit': 0,
87 'thumbnail': r're:https://.+',
88 },
89 'skip': 'Geo-blocked outside PL',
90 }, {
91 # client-side rendered (regional) program (playlist) page
92 'url': 'https://opole.tvp.pl/9660819/rozmowa-dnia',
93 'info_dict': {
94 'id': '9660819',
95 'description': 'Od poniedziałku do piątku o 18:55',
96 'title': 'Rozmowa dnia',
97 },
98 'playlist_mincount': 1800,
99 'params': {
100 'skip_download': True,
101 }
102 }, {
103 # ABC-specific video embeding
104 # moved to https://bajkowakraina.tvp.pl/wideo/50981130,teleranek,51027049,zubr,51116450
105 'url': 'https://abc.tvp.pl/48636269/zubry-odc-124',
106 'info_dict': {
107 'id': '48320456',
108 'ext': 'mp4',
109 'title': 'Teleranek, Żubr',
110 },
111 'skip': 'unavailable',
112 }, {
113 # yet another vue page
114 'url': 'https://jp2.tvp.pl/46925618/filmy',
115 'info_dict': {
116 'id': '46925618',
117 'title': 'Filmy',
118 },
119 'playlist_mincount': 19,
120 }, {
121 'url': 'http://vod.tvp.pl/seriale/obyczajowe/na-sygnale/sezon-2-27-/odc-39/17834272',
122 'only_matching': True,
123 }, {
124 'url': 'http://wiadomosci.tvp.pl/25169746/24052016-1200',
125 'only_matching': True,
126 }, {
127 'url': 'http://krakow.tvp.pl/25511623/25lecie-mck-wyjatkowe-miejsce-na-mapie-krakowa',
128 'only_matching': True,
129 }, {
130 'url': 'http://teleexpress.tvp.pl/25522307/wierni-wzieli-udzial-w-procesjach',
131 'only_matching': True,
132 }, {
133 'url': 'http://sport.tvp.pl/25522165/krychowiak-uspokaja-w-sprawie-kontuzji-dwa-tygodnie-to-maksimum',
134 'only_matching': True,
135 }, {
136 'url': 'http://www.tvp.info/25511919/trwa-rewolucja-wladza-zdecydowala-sie-na-pogwalcenie-konstytucji',
137 'only_matching': True,
138 }, {
139 'url': 'https://tvp.info/49193823/teczowe-flagi-na-pomnikach-prokuratura-wszczela-postepowanie-wieszwiecej',
140 'only_matching': True,
141 }, {
142 'url': 'https://www.tvpparlament.pl/retransmisje-vod/inne/wizyta-premiera-mateusza-morawieckiego-w-firmie-berotu-sp-z-oo/48857277',
143 'only_matching': True,
144 }, {
145 'url': 'https://tvpworld.com/48583640/tescos-polish-business-bought-by-danish-chain-netto',
146 'only_matching': True,
147 }]
148
149 def _parse_vue_website_data(self, webpage, page_id):
150 website_data = self._search_regex([
151 # website - regiony, tvp.info
152 # directory - jp2.tvp.pl
153 r'window\.__(?:website|directory)Data\s*=\s*({(?:.|\s)+?});',
154 ], webpage, 'website data')
155 if not website_data:
156 return None
157 return self._parse_json(website_data, page_id, transform_source=js_to_json)
158
159 def _extract_vue_video(self, video_data, page_id=None):
160 if isinstance(video_data, str):
161 video_data = self._parse_json(video_data, page_id, transform_source=js_to_json)
162 thumbnails = []
163 image = video_data.get('image')
164 if image:
165 for thumb in (image if isinstance(image, list) else [image]):
166 thmb_url = str_or_none(thumb.get('url'))
167 if thmb_url:
168 thumbnails.append({
169 'url': thmb_url,
170 })
171 is_website = video_data.get('type') == 'website'
172 if is_website:
173 url = video_data['url']
174 else:
175 url = 'tvp:' + str_or_none(video_data.get('_id') or page_id)
176 return {
177 '_type': 'url_transparent',
178 'id': str_or_none(video_data.get('_id') or page_id),
179 'url': url,
180 'ie_key': (TVPIE if is_website else TVPEmbedIE).ie_key(),
181 'title': str_or_none(video_data.get('title')),
182 'description': str_or_none(video_data.get('lead')),
183 'timestamp': int_or_none(video_data.get('release_date_long')),
184 'duration': int_or_none(video_data.get('duration')),
185 'thumbnails': thumbnails,
186 }
187
188 def _handle_vuejs_page(self, url, webpage, page_id):
189 # vue client-side rendered sites (all regional pages + tvp.info)
190 video_data = self._search_regex([
191 r'window\.__(?:news|video)Data\s*=\s*({(?:.|\s)+?})\s*;',
192 ], webpage, 'video data', default=None)
193 if video_data:
194 return self._extract_vue_video(video_data, page_id=page_id)
195 # paged playlists
196 website_data = self._parse_vue_website_data(webpage, page_id)
197 if website_data:
198 entries = self._vuejs_entries(url, website_data, page_id)
199
200 return {
201 '_type': 'playlist',
202 'id': page_id,
203 'title': str_or_none(website_data.get('title')),
204 'description': str_or_none(website_data.get('lead')),
205 'entries': entries,
206 }
207 raise ExtractorError('Could not extract video/website data')
208
209 def _vuejs_entries(self, url, website_data, page_id):
210
211 def extract_videos(wd):
212 if wd.get('latestVideo'):
213 yield self._extract_vue_video(wd['latestVideo'])
214 for video in wd.get('videos') or []:
215 yield self._extract_vue_video(video)
216 for video in wd.get('items') or []:
217 yield self._extract_vue_video(video)
218
219 yield from extract_videos(website_data)
220
221 if website_data.get('items_total_count') > website_data.get('items_per_page'):
222 for page in itertools.count(2):
223 page_website_data = self._parse_vue_website_data(
224 self._download_webpage(url, page_id, note='Downloading page #%d' % page,
225 query={'page': page}),
226 page_id)
227 if not page_website_data.get('videos') and not page_website_data.get('items'):
228 break
229 yield from extract_videos(page_website_data)
230
231 def _real_extract(self, url):
232 page_id = self._match_id(url)
233 webpage, urlh = self._download_webpage_handle(url, page_id)
234
235 # The URL may redirect to a VOD
236 # example: https://vod.tvp.pl/48463890/wadowickie-spotkania-z-janem-pawlem-ii
237 for ie_cls in (TVPVODSeriesIE, TVPVODVideoIE):
238 if ie_cls.suitable(urlh.url):
239 return self.url_result(urlh.url, ie=ie_cls.ie_key(), video_id=page_id)
240
241 if re.search(
242 r'window\.__(?:video|news|website|directory)Data\s*=',
243 webpage):
244 return self._handle_vuejs_page(url, webpage, page_id)
245
246 # classic server-side rendered sites
247 video_id = self._search_regex([
248 r'<iframe[^>]+src="[^"]*?embed\.php\?(?:[^&]+&)*ID=(\d+)',
249 r'<iframe[^>]+src="[^"]*?object_id=(\d+)',
250 r"object_id\s*:\s*'(\d+)'",
251 r'data-video-id="(\d+)"',
252
253 # abc.tvp.pl - somehow there are more than one video IDs that seem to be the same video?
254 # the first one is referenced to as "copyid", and seems to be unused by the website
255 r'<script>\s*tvpabc\.video\.init\(\s*\d+,\s*(\d+)\s*\)\s*</script>',
256 ], webpage, 'video id', default=page_id)
257 return {
258 '_type': 'url_transparent',
259 'url': 'tvp:' + video_id,
260 'description': self._og_search_description(
261 webpage, default=None) or (self._html_search_meta(
262 'description', webpage, default=None)
263 if '//s.tvp.pl/files/portal/v' in webpage else None),
264 'thumbnail': self._og_search_thumbnail(webpage, default=None),
265 'ie_key': 'TVPEmbed',
266 }
267
268
269 class TVPStreamIE(InfoExtractor):
270 IE_NAME = 'tvp:stream'
271 _VALID_URL = r'(?:tvpstream:|https?://tvpstream\.vod\.tvp\.pl/(?:\?(?:[^&]+[&;])*channel_id=)?)(?P<id>\d*)'
272 _TESTS = [{
273 # untestable as "video" id changes many times across a day
274 'url': 'https://tvpstream.vod.tvp.pl/?channel_id=1455',
275 'only_matching': True,
276 }, {
277 'url': 'tvpstream:39821455',
278 'only_matching': True,
279 }, {
280 # the default stream when you provide no channel_id, most probably TVP Info
281 'url': 'tvpstream:',
282 'only_matching': True,
283 }, {
284 'url': 'https://tvpstream.vod.tvp.pl/',
285 'only_matching': True,
286 }]
287
288 _PLAYER_BOX_RE = r'<div\s[^>]*id\s*=\s*["\']?tvp_player_box["\']?[^>]+data-%s-id\s*=\s*["\']?(\d+)'
289 _BUTTON_RE = r'<div\s[^>]*data-channel-id=["\']?%s["\']?[^>]*\sdata-title=(?:"([^"]*)"|\'([^\']*)\')[^>]*\sdata-stationname=(?:"([^"]*)"|\'([^\']*)\')'
290
291 def _real_extract(self, url):
292 channel_id = self._match_id(url)
293 channel_url = self._proto_relative_url('//tvpstream.vod.tvp.pl/?channel_id=%s' % channel_id or 'default')
294 webpage = self._download_webpage(channel_url, channel_id, 'Downloading channel webpage')
295 if not channel_id:
296 channel_id = self._search_regex(self._PLAYER_BOX_RE % 'channel',
297 webpage, 'default channel id')
298 video_id = self._search_regex(self._PLAYER_BOX_RE % 'video',
299 webpage, 'video id')
300 audition_title, station_name = self._search_regex(
301 self._BUTTON_RE % (re.escape(channel_id)), webpage,
302 'audition title and station name',
303 group=(1, 2))
304 return {
305 '_type': 'url_transparent',
306 'id': channel_id,
307 'url': 'tvp:%s' % video_id,
308 'title': audition_title,
309 'alt_title': station_name,
310 'is_live': True,
311 'ie_key': 'TVPEmbed',
312 }
313
314
315 class TVPEmbedIE(InfoExtractor):
316 IE_NAME = 'tvp:embed'
317 IE_DESC = 'Telewizja Polska'
318 _GEO_BYPASS = False
319 _VALID_URL = r'''(?x)
320 (?:
321 tvp:
322 |https?://
323 (?:[^/]+\.)?
324 (?:tvp(?:parlament)?\.pl|tvp\.info|tvpworld\.com|swipeto\.pl)/
325 (?:sess/
326 (?:tvplayer\.php\?.*?object_id
327 |TVPlayer2/(?:embed|api)\.php\?.*[Ii][Dd])
328 |shared/details\.php\?.*?object_id)
329 =)
330 (?P<id>\d+)
331 '''
332 _EMBED_REGEX = [rf'(?x)<iframe[^>]+?src=(["\'])(?P<url>{_VALID_URL[4:]})']
333
334 _TESTS = [{
335 'url': 'tvp:194536',
336 'info_dict': {
337 'id': '194536',
338 'ext': 'mp4',
339 'title': 'Czas honoru, odc. 13 – Władek',
340 'description': 'md5:76649d2014f65c99477be17f23a4dead',
341 'age_limit': 12,
342 'duration': 2652,
343 'series': 'Czas honoru',
344 'episode': 'Episode 13',
345 'episode_number': 13,
346 'season': 'sezon 1',
347 'thumbnail': r're:https://.+',
348 },
349 }, {
350 'url': 'https://www.tvp.pl/sess/tvplayer.php?object_id=51247504&amp;autoplay=false',
351 'info_dict': {
352 'id': '51247504',
353 'ext': 'mp4',
354 'title': 'Razmova 091220',
355 'duration': 876,
356 'age_limit': 0,
357 'thumbnail': r're:https://.+',
358 },
359 }, {
360 # TVPlayer2 embed URL
361 'url': 'https://tvp.info/sess/TVPlayer2/embed.php?ID=50595757',
362 'only_matching': True,
363 }, {
364 'url': 'https://wiadomosci.tvp.pl/sess/TVPlayer2/api.php?id=51233452',
365 'only_matching': True,
366 }, {
367 # pulsembed on dziennik.pl
368 'url': 'https://www.tvp.pl/shared/details.php?copy_id=52205981&object_id=52204505&autoplay=false&is_muted=false&allowfullscreen=true&template=external-embed/video/iframe-video.html',
369 'only_matching': True,
370 }]
371
372 def _real_extract(self, url):
373 video_id = self._match_id(url)
374
375 # it could be anything that is a valid JS function name
376 callback = random.choice((
377 'jebac_pis',
378 'jebacpis',
379 'ziobro',
380 'sasin70',
381 'sasin_przejebal_70_milionow_PLN',
382 'tvp_is_a_state_propaganda_service',
383 ))
384
385 webpage = self._download_webpage(
386 ('https://www.tvp.pl/sess/TVPlayer2/api.php?id=%s'
387 + '&@method=getTvpConfig&@callback=%s') % (video_id, callback), video_id)
388
389 # stripping JSONP padding
390 datastr = webpage[15 + len(callback):-3]
391 if datastr.startswith('null,'):
392 error = self._parse_json(datastr[5:], video_id, fatal=False)
393 error_desc = traverse_obj(error, (0, 'desc'))
394
395 if error_desc == 'Obiekt wymaga płatności':
396 raise ExtractorError('Video requires payment and log-in, but log-in is not implemented')
397
398 raise ExtractorError(error_desc or 'unexpected JSON error')
399
400 content = self._parse_json(datastr, video_id)['content']
401 info = content['info']
402 is_live = try_get(info, lambda x: x['isLive'], bool)
403
404 if info.get('isGeoBlocked'):
405 # actual country list is not provided, we just assume it's always available in PL
406 self.raise_geo_restricted(countries=['PL'])
407
408 formats = []
409 for file in content['files']:
410 video_url = url_or_none(file.get('url'))
411 if not video_url:
412 continue
413 ext = determine_ext(video_url, None)
414 if ext == 'm3u8':
415 formats.extend(self._extract_m3u8_formats(video_url, video_id, m3u8_id='hls', fatal=False, live=is_live))
416 elif ext == 'mpd':
417 if is_live:
418 # doesn't work with either ffmpeg or native downloader
419 continue
420 formats.extend(self._extract_mpd_formats(video_url, video_id, mpd_id='dash', fatal=False))
421 elif ext == 'f4m':
422 formats.extend(self._extract_f4m_formats(video_url, video_id, f4m_id='hds', fatal=False))
423 elif video_url.endswith('.ism/manifest'):
424 formats.extend(self._extract_ism_formats(video_url, video_id, ism_id='mss', fatal=False))
425 else:
426 formats.append({
427 'format_id': 'direct',
428 'url': video_url,
429 'ext': ext or file.get('type'),
430 'fps': int_or_none(traverse_obj(file, ('quality', 'fps'))),
431 'tbr': int_or_none(traverse_obj(file, ('quality', 'bitrate')), scale=1000),
432 'width': int_or_none(traverse_obj(file, ('quality', 'width'))),
433 'height': int_or_none(traverse_obj(file, ('quality', 'height'))),
434 })
435
436 title = dict_get(info, ('subtitle', 'title', 'seoTitle'))
437 description = dict_get(info, ('description', 'seoDescription'))
438 thumbnails = []
439 for thumb in content.get('posters') or ():
440 thumb_url = thumb.get('src')
441 if not thumb_url or '{width}' in thumb_url or '{height}' in thumb_url:
442 continue
443 thumbnails.append({
444 'url': thumb.get('src'),
445 'width': thumb.get('width'),
446 'height': thumb.get('height'),
447 })
448 age_limit = try_get(info, lambda x: x['ageGroup']['minAge'], int)
449 if age_limit == 1:
450 age_limit = 0
451 duration = try_get(info, lambda x: x['duration'], int) if not is_live else None
452
453 subtitles = {}
454 for sub in content.get('subtitles') or []:
455 if not sub.get('url'):
456 continue
457 subtitles.setdefault(sub['lang'], []).append({
458 'url': sub['url'],
459 'ext': sub.get('type'),
460 })
461
462 info_dict = {
463 'id': video_id,
464 'title': title,
465 'description': description,
466 'thumbnails': thumbnails,
467 'age_limit': age_limit,
468 'is_live': is_live,
469 'duration': duration,
470 'formats': formats,
471 'subtitles': subtitles,
472 }
473
474 # vod.tvp.pl
475 if info.get('vortalName') == 'vod':
476 info_dict.update({
477 'title': '%s, %s' % (info.get('title'), info.get('subtitle')),
478 'series': info.get('title'),
479 'season': info.get('season'),
480 'episode_number': info.get('episode'),
481 })
482
483 return info_dict
484
485
486 class TVPVODBaseIE(InfoExtractor):
487 _API_BASE_URL = 'https://vod.tvp.pl/api/products'
488
489 def _call_api(self, resource, video_id, **kwargs):
490 return self._download_json(
491 f'{self._API_BASE_URL}/{resource}', video_id,
492 query={'lang': 'pl', 'platform': 'BROWSER'}, **kwargs)
493
494 def _parse_video(self, video):
495 return {
496 '_type': 'url',
497 'url': 'tvp:' + video['externalUid'],
498 'ie_key': TVPEmbedIE.ie_key(),
499 'title': video.get('title'),
500 'description': traverse_obj(video, ('lead', 'description')),
501 'age_limit': int_or_none(video.get('rating')),
502 'duration': int_or_none(video.get('duration')),
503 }
504
505
506 class TVPVODVideoIE(TVPVODBaseIE):
507 IE_NAME = 'tvp:vod'
508 _VALID_URL = r'https?://vod\.tvp\.pl/[a-z\d-]+,\d+/[a-z\d-]+(?<!-odcinki)(?:-odcinki,\d+/odcinek-\d+,S\d+E\d+)?,(?P<id>\d+)(?:\?[^#]+)?(?:#.+)?$'
509
510 _TESTS = [{
511 'url': 'https://vod.tvp.pl/dla-dzieci,24/laboratorium-alchemika-odcinki,309338/odcinek-24,S01E24,311357',
512 'info_dict': {
513 'id': '60468609',
514 'ext': 'mp4',
515 'title': 'Laboratorium alchemika, Tusze termiczne. Jak zobaczyć niewidoczne. Odcinek 24',
516 'description': 'md5:1d4098d3e537092ccbac1abf49b7cd4c',
517 'duration': 300,
518 'episode_number': 24,
519 'episode': 'Episode 24',
520 'age_limit': 0,
521 'series': 'Laboratorium alchemika',
522 'thumbnail': 're:https://.+',
523 },
524 }, {
525 'url': 'https://vod.tvp.pl/filmy-dokumentalne,163/ukrainski-sluga-narodu,339667',
526 'info_dict': {
527 'id': '51640077',
528 'ext': 'mp4',
529 'title': 'Ukraiński sługa narodu, Ukraiński sługa narodu',
530 'series': 'Ukraiński sługa narodu',
531 'description': 'md5:b7940c0a8e439b0c81653a986f544ef3',
532 'age_limit': 12,
533 'episode': 'Episode 0',
534 'episode_number': 0,
535 'duration': 3051,
536 'thumbnail': 're:https://.+',
537 },
538 }]
539
540 def _real_extract(self, url):
541 video_id = self._match_id(url)
542
543 return self._parse_video(self._call_api(f'vods/{video_id}', video_id))
544
545
546 class TVPVODSeriesIE(TVPVODBaseIE):
547 IE_NAME = 'tvp:vod:series'
548 _VALID_URL = r'https?://vod\.tvp\.pl/[a-z\d-]+,\d+/[a-z\d-]+-odcinki,(?P<id>\d+)(?:\?[^#]+)?(?:#.+)?$'
549
550 _TESTS = [{
551 'url': 'https://vod.tvp.pl/seriale,18/ranczo-odcinki,316445',
552 'info_dict': {
553 'id': '316445',
554 'title': 'Ranczo',
555 'age_limit': 12,
556 'categories': ['seriale'],
557 },
558 'playlist_count': 129,
559 }, {
560 'url': 'https://vod.tvp.pl/programy,88/rolnik-szuka-zony-odcinki,284514',
561 'only_matching': True,
562 }, {
563 'url': 'https://vod.tvp.pl/dla-dzieci,24/laboratorium-alchemika-odcinki,309338',
564 'only_matching': True,
565 }]
566
567 def _entries(self, seasons, playlist_id):
568 for season in seasons:
569 episodes = self._call_api(
570 f'vods/serials/{playlist_id}/seasons/{season["id"]}/episodes', playlist_id,
571 note=f'Downloading episode list for {season["title"]}')
572 yield from map(self._parse_video, episodes)
573
574 def _real_extract(self, url):
575 playlist_id = self._match_id(url)
576 metadata = self._call_api(
577 f'vods/serials/{playlist_id}', playlist_id,
578 note='Downloading serial metadata')
579 seasons = self._call_api(
580 f'vods/serials/{playlist_id}/seasons', playlist_id,
581 note='Downloading season list')
582 return self.playlist_result(
583 self._entries(seasons, playlist_id), playlist_id, strip_or_none(metadata.get('title')),
584 clean_html(traverse_obj(metadata, ('description', 'lead'), expected_type=strip_or_none)),
585 categories=[traverse_obj(metadata, ('mainCategory', 'name'))],
586 age_limit=int_or_none(metadata.get('rating')),
587 )