]> Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/extractor/rai.py
Update upstream source from tag 'upstream/2019.09.28'
[youtubedl] / youtube_dl / extractor / rai.py
1 from __future__ import unicode_literals
2
3 import re
4
5 from .common import InfoExtractor
6 from ..compat import (
7 compat_urlparse,
8 compat_str,
9 )
10 from ..utils import (
11 ExtractorError,
12 determine_ext,
13 find_xpath_attr,
14 fix_xml_ampersands,
15 GeoRestrictedError,
16 int_or_none,
17 parse_duration,
18 strip_or_none,
19 try_get,
20 unescapeHTML,
21 unified_strdate,
22 unified_timestamp,
23 update_url_query,
24 urljoin,
25 xpath_text,
26 )
27
28
29 class RaiBaseIE(InfoExtractor):
30 _UUID_RE = r'[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12}'
31 _GEO_COUNTRIES = ['IT']
32 _GEO_BYPASS = False
33
34 def _extract_relinker_info(self, relinker_url, video_id):
35 if not re.match(r'https?://', relinker_url):
36 return {'formats': [{'url': relinker_url}]}
37
38 formats = []
39 geoprotection = None
40 is_live = None
41 duration = None
42
43 for platform in ('mon', 'flash', 'native'):
44 relinker = self._download_xml(
45 relinker_url, video_id,
46 note='Downloading XML metadata for platform %s' % platform,
47 transform_source=fix_xml_ampersands,
48 query={'output': 45, 'pl': platform},
49 headers=self.geo_verification_headers())
50
51 if not geoprotection:
52 geoprotection = xpath_text(
53 relinker, './geoprotection', default=None) == 'Y'
54
55 if not is_live:
56 is_live = xpath_text(
57 relinker, './is_live', default=None) == 'Y'
58 if not duration:
59 duration = parse_duration(xpath_text(
60 relinker, './duration', default=None))
61
62 url_elem = find_xpath_attr(relinker, './url', 'type', 'content')
63 if url_elem is None:
64 continue
65
66 media_url = url_elem.text
67
68 # This does not imply geo restriction (e.g.
69 # http://www.raisport.rai.it/dl/raiSport/media/rassegna-stampa-04a9f4bd-b563-40cf-82a6-aad3529cb4a9.html)
70 if media_url == 'http://download.rai.it/video_no_available.mp4':
71 continue
72
73 ext = determine_ext(media_url)
74 if (ext == 'm3u8' and platform != 'mon') or (ext == 'f4m' and platform != 'flash'):
75 continue
76
77 if ext == 'm3u8' or 'format=m3u8' in media_url or platform == 'mon':
78 formats.extend(self._extract_m3u8_formats(
79 media_url, video_id, 'mp4', 'm3u8_native',
80 m3u8_id='hls', fatal=False))
81 elif ext == 'f4m' or platform == 'flash':
82 manifest_url = update_url_query(
83 media_url.replace('manifest#live_hds.f4m', 'manifest.f4m'),
84 {'hdcore': '3.7.0', 'plugin': 'aasp-3.7.0.39.44'})
85 formats.extend(self._extract_f4m_formats(
86 manifest_url, video_id, f4m_id='hds', fatal=False))
87 else:
88 bitrate = int_or_none(xpath_text(relinker, 'bitrate'))
89 formats.append({
90 'url': media_url,
91 'tbr': bitrate if bitrate > 0 else None,
92 'format_id': 'http-%d' % bitrate if bitrate > 0 else 'http',
93 })
94
95 if not formats and geoprotection is True:
96 self.raise_geo_restricted(countries=self._GEO_COUNTRIES)
97
98 return dict((k, v) for k, v in {
99 'is_live': is_live,
100 'duration': duration,
101 'formats': formats,
102 }.items() if v is not None)
103
104 @staticmethod
105 def _extract_subtitles(url, subtitle_url):
106 subtitles = {}
107 if subtitle_url and isinstance(subtitle_url, compat_str):
108 subtitle_url = urljoin(url, subtitle_url)
109 STL_EXT = '.stl'
110 SRT_EXT = '.srt'
111 subtitles['it'] = [{
112 'ext': 'stl',
113 'url': subtitle_url,
114 }]
115 if subtitle_url.endswith(STL_EXT):
116 srt_url = subtitle_url[:-len(STL_EXT)] + SRT_EXT
117 subtitles['it'].append({
118 'ext': 'srt',
119 'url': srt_url,
120 })
121 return subtitles
122
123
124 class RaiPlayIE(RaiBaseIE):
125 _VALID_URL = r'(?P<url>https?://(?:www\.)?raiplay\.it/.+?-(?P<id>%s)\.html)' % RaiBaseIE._UUID_RE
126 _TESTS = [{
127 'url': 'http://www.raiplay.it/video/2016/10/La-Casa-Bianca-e06118bb-59a9-4636-b914-498e4cfd2c66.html?source=twitter',
128 'md5': '340aa3b7afb54bfd14a8c11786450d76',
129 'info_dict': {
130 'id': 'e06118bb-59a9-4636-b914-498e4cfd2c66',
131 'ext': 'mp4',
132 'title': 'La Casa Bianca',
133 'alt_title': 'S2016 - Puntata del 23/10/2016',
134 'description': 'md5:a09d45890850458077d1f68bb036e0a5',
135 'thumbnail': r're:^https?://.*\.jpg$',
136 'uploader': 'Rai 3',
137 'creator': 'Rai 3',
138 'duration': 3278,
139 'timestamp': 1477764300,
140 'upload_date': '20161029',
141 'series': 'La Casa Bianca',
142 'season': '2016',
143 },
144 }, {
145 'url': 'http://www.raiplay.it/video/2014/04/Report-del-07042014-cb27157f-9dd0-4aee-b788-b1f67643a391.html',
146 'md5': '8970abf8caf8aef4696e7b1f2adfc696',
147 'info_dict': {
148 'id': 'cb27157f-9dd0-4aee-b788-b1f67643a391',
149 'ext': 'mp4',
150 'title': 'Report del 07/04/2014',
151 'alt_title': 'S2013/14 - Puntata del 07/04/2014',
152 'description': 'md5:f27c544694cacb46a078db84ec35d2d9',
153 'thumbnail': r're:^https?://.*\.jpg$',
154 'uploader': 'Rai 5',
155 'creator': 'Rai 5',
156 'duration': 6160,
157 'series': 'Report',
158 'season_number': 5,
159 'season': '2013/14',
160 },
161 'params': {
162 'skip_download': True,
163 },
164 }, {
165 'url': 'http://www.raiplay.it/video/2016/11/gazebotraindesi-efebe701-969c-4593-92f3-285f0d1ce750.html?',
166 'only_matching': True,
167 }]
168
169 def _real_extract(self, url):
170 mobj = re.match(self._VALID_URL, url)
171 url, video_id = mobj.group('url', 'id')
172
173 media = self._download_json(
174 '%s?json' % url, video_id, 'Downloading video JSON')
175
176 title = media['name']
177
178 video = media['video']
179
180 relinker_info = self._extract_relinker_info(video['contentUrl'], video_id)
181 self._sort_formats(relinker_info['formats'])
182
183 thumbnails = []
184 if 'images' in media:
185 for _, value in media.get('images').items():
186 if value:
187 thumbnails.append({
188 'url': value.replace('[RESOLUTION]', '600x400')
189 })
190
191 timestamp = unified_timestamp(try_get(
192 media, lambda x: x['availabilities'][0]['start'], compat_str))
193
194 subtitles = self._extract_subtitles(url, video.get('subtitles'))
195
196 info = {
197 'id': video_id,
198 'title': self._live_title(title) if relinker_info.get(
199 'is_live') else title,
200 'alt_title': media.get('subtitle'),
201 'description': media.get('description'),
202 'uploader': strip_or_none(media.get('channel')),
203 'creator': strip_or_none(media.get('editor')),
204 'duration': parse_duration(video.get('duration')),
205 'timestamp': timestamp,
206 'thumbnails': thumbnails,
207 'series': try_get(
208 media, lambda x: x['isPartOf']['name'], compat_str),
209 'season_number': int_or_none(try_get(
210 media, lambda x: x['isPartOf']['numeroStagioni'])),
211 'season': media.get('stagione') or None,
212 'subtitles': subtitles,
213 }
214
215 info.update(relinker_info)
216 return info
217
218
219 class RaiPlayLiveIE(RaiBaseIE):
220 _VALID_URL = r'https?://(?:www\.)?raiplay\.it/dirette/(?P<id>[^/?#&]+)'
221 _TEST = {
222 'url': 'http://www.raiplay.it/dirette/rainews24',
223 'info_dict': {
224 'id': 'd784ad40-e0ae-4a69-aa76-37519d238a9c',
225 'display_id': 'rainews24',
226 'ext': 'mp4',
227 'title': 're:^Diretta di Rai News 24 [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
228 'description': 'md5:6eca31500550f9376819f174e5644754',
229 'uploader': 'Rai News 24',
230 'creator': 'Rai News 24',
231 'is_live': True,
232 },
233 'params': {
234 'skip_download': True,
235 },
236 }
237
238 def _real_extract(self, url):
239 display_id = self._match_id(url)
240
241 webpage = self._download_webpage(url, display_id)
242
243 video_id = self._search_regex(
244 r'data-uniquename=["\']ContentItem-(%s)' % RaiBaseIE._UUID_RE,
245 webpage, 'content id')
246
247 return {
248 '_type': 'url_transparent',
249 'ie_key': RaiPlayIE.ie_key(),
250 'url': 'http://www.raiplay.it/dirette/ContentItem-%s.html' % video_id,
251 'id': video_id,
252 'display_id': display_id,
253 }
254
255
256 class RaiPlayPlaylistIE(InfoExtractor):
257 _VALID_URL = r'https?://(?:www\.)?raiplay\.it/programmi/(?P<id>[^/?#&]+)'
258 _TESTS = [{
259 'url': 'http://www.raiplay.it/programmi/nondirloalmiocapo/',
260 'info_dict': {
261 'id': 'nondirloalmiocapo',
262 'title': 'Non dirlo al mio capo',
263 'description': 'md5:9f3d603b2947c1c7abb098f3b14fac86',
264 },
265 'playlist_mincount': 12,
266 }]
267
268 def _real_extract(self, url):
269 playlist_id = self._match_id(url)
270
271 webpage = self._download_webpage(url, playlist_id)
272
273 title = self._html_search_meta(
274 ('programma', 'nomeProgramma'), webpage, 'title')
275 description = unescapeHTML(self._html_search_meta(
276 ('description', 'og:description'), webpage, 'description'))
277
278 entries = []
279 for mobj in re.finditer(
280 r'<a\b[^>]+\bhref=(["\'])(?P<path>/raiplay/video/.+?)\1',
281 webpage):
282 video_url = urljoin(url, mobj.group('path'))
283 entries.append(self.url_result(
284 video_url, ie=RaiPlayIE.ie_key(),
285 video_id=RaiPlayIE._match_id(video_url)))
286
287 return self.playlist_result(entries, playlist_id, title, description)
288
289
290 class RaiIE(RaiBaseIE):
291 _VALID_URL = r'https?://[^/]+\.(?:rai\.(?:it|tv)|rainews\.it)/.+?-(?P<id>%s)(?:-.+?)?\.html' % RaiBaseIE._UUID_RE
292 _TESTS = [{
293 # var uniquename = "ContentItem-..."
294 # data-id="ContentItem-..."
295 'url': 'http://www.raisport.rai.it/dl/raiSport/media/rassegna-stampa-04a9f4bd-b563-40cf-82a6-aad3529cb4a9.html',
296 'info_dict': {
297 'id': '04a9f4bd-b563-40cf-82a6-aad3529cb4a9',
298 'ext': 'mp4',
299 'title': 'TG PRIMO TEMPO',
300 'thumbnail': r're:^https?://.*\.jpg$',
301 'duration': 1758,
302 'upload_date': '20140612',
303 }
304 }, {
305 # with ContentItem in many metas
306 'url': 'http://www.rainews.it/dl/rainews/media/Weekend-al-cinema-da-Hollywood-arriva-il-thriller-di-Tate-Taylor-La-ragazza-del-treno-1632c009-c843-4836-bb65-80c33084a64b.html',
307 'info_dict': {
308 'id': '1632c009-c843-4836-bb65-80c33084a64b',
309 'ext': 'mp4',
310 'title': 'Weekend al cinema, da Hollywood arriva il thriller di Tate Taylor "La ragazza del treno"',
311 'description': 'I film in uscita questa settimana.',
312 'thumbnail': r're:^https?://.*\.png$',
313 'duration': 833,
314 'upload_date': '20161103',
315 }
316 }, {
317 # with ContentItem in og:url
318 'url': 'http://www.rai.it/dl/RaiTV/programmi/media/ContentItem-efb17665-691c-45d5-a60c-5301333cbb0c.html',
319 'md5': '11959b4e44fa74de47011b5799490adf',
320 'info_dict': {
321 'id': 'efb17665-691c-45d5-a60c-5301333cbb0c',
322 'ext': 'mp4',
323 'title': 'TG1 ore 20:00 del 03/11/2016',
324 'description': 'TG1 edizione integrale ore 20:00 del giorno 03/11/2016',
325 'thumbnail': r're:^https?://.*\.jpg$',
326 'duration': 2214,
327 'upload_date': '20161103',
328 }
329 }, {
330 # drawMediaRaiTV(...)
331 'url': 'http://www.report.rai.it/dl/Report/puntata/ContentItem-0c7a664b-d0f4-4b2c-8835-3f82e46f433e.html',
332 'md5': '2dd727e61114e1ee9c47f0da6914e178',
333 'info_dict': {
334 'id': '59d69d28-6bb6-409d-a4b5-ed44096560af',
335 'ext': 'mp4',
336 'title': 'Il pacco',
337 'description': 'md5:4b1afae1364115ce5d78ed83cd2e5b3a',
338 'thumbnail': r're:^https?://.*\.jpg$',
339 'upload_date': '20141221',
340 },
341 }, {
342 # initEdizione('ContentItem-...'
343 'url': 'http://www.tg1.rai.it/dl/tg1/2010/edizioni/ContentSet-9b6e0cba-4bef-4aef-8cf0-9f7f665b7dfb-tg1.html?item=undefined',
344 'info_dict': {
345 'id': 'c2187016-8484-4e3a-8ac8-35e475b07303',
346 'ext': 'mp4',
347 'title': r're:TG1 ore \d{2}:\d{2} del \d{2}/\d{2}/\d{4}',
348 'duration': 2274,
349 'upload_date': '20170401',
350 },
351 'skip': 'Changes daily',
352 }, {
353 # HDS live stream with only relinker URL
354 'url': 'http://www.rai.tv/dl/RaiTV/dirette/PublishingBlock-1912dbbf-3f96-44c3-b4cf-523681fbacbc.html?channel=EuroNews',
355 'info_dict': {
356 'id': '1912dbbf-3f96-44c3-b4cf-523681fbacbc',
357 'ext': 'flv',
358 'title': 'EuroNews',
359 },
360 'params': {
361 'skip_download': True,
362 },
363 }, {
364 # HLS live stream with ContentItem in og:url
365 'url': 'http://www.rainews.it/dl/rainews/live/ContentItem-3156f2f2-dc70-4953-8e2f-70d7489d4ce9.html',
366 'info_dict': {
367 'id': '3156f2f2-dc70-4953-8e2f-70d7489d4ce9',
368 'ext': 'mp4',
369 'title': 'La diretta di Rainews24',
370 },
371 'params': {
372 'skip_download': True,
373 },
374 }, {
375 # Direct MMS URL
376 'url': 'http://www.rai.it/dl/RaiTV/programmi/media/ContentItem-b63a4089-ac28-48cf-bca5-9f5b5bc46df5.html',
377 'only_matching': True,
378 }, {
379 'url': 'https://www.rainews.it/tgr/marche/notiziari/video/2019/02/ContentItem-6ba945a2-889c-4a80-bdeb-8489c70a8db9.html',
380 'only_matching': True,
381 }]
382
383 def _extract_from_content_id(self, content_id, url):
384 media = self._download_json(
385 'http://www.rai.tv/dl/RaiTV/programmi/media/ContentItem-%s.html?json' % content_id,
386 content_id, 'Downloading video JSON')
387
388 title = media['name'].strip()
389
390 media_type = media['type']
391 if 'Audio' in media_type:
392 relinker_info = {
393 'formats': [{
394 'format_id': media.get('formatoAudio'),
395 'url': media['audioUrl'],
396 'ext': media.get('formatoAudio'),
397 }]
398 }
399 elif 'Video' in media_type:
400 relinker_info = self._extract_relinker_info(media['mediaUri'], content_id)
401 else:
402 raise ExtractorError('not a media file')
403
404 self._sort_formats(relinker_info['formats'])
405
406 thumbnails = []
407 for image_type in ('image', 'image_medium', 'image_300'):
408 thumbnail_url = media.get(image_type)
409 if thumbnail_url:
410 thumbnails.append({
411 'url': compat_urlparse.urljoin(url, thumbnail_url),
412 })
413
414 subtitles = self._extract_subtitles(url, media.get('subtitlesUrl'))
415
416 info = {
417 'id': content_id,
418 'title': title,
419 'description': strip_or_none(media.get('desc')),
420 'thumbnails': thumbnails,
421 'uploader': media.get('author'),
422 'upload_date': unified_strdate(media.get('date')),
423 'duration': parse_duration(media.get('length')),
424 'subtitles': subtitles,
425 }
426
427 info.update(relinker_info)
428
429 return info
430
431 def _real_extract(self, url):
432 video_id = self._match_id(url)
433
434 webpage = self._download_webpage(url, video_id)
435
436 content_item_id = None
437
438 content_item_url = self._html_search_meta(
439 ('og:url', 'og:video', 'og:video:secure_url', 'twitter:url',
440 'twitter:player', 'jsonlink'), webpage, default=None)
441 if content_item_url:
442 content_item_id = self._search_regex(
443 r'ContentItem-(%s)' % self._UUID_RE, content_item_url,
444 'content item id', default=None)
445
446 if not content_item_id:
447 content_item_id = self._search_regex(
448 r'''(?x)
449 (?:
450 (?:initEdizione|drawMediaRaiTV)\(|
451 <(?:[^>]+\bdata-id|var\s+uniquename)=
452 )
453 (["\'])
454 (?:(?!\1).)*\bContentItem-(?P<id>%s)
455 ''' % self._UUID_RE,
456 webpage, 'content item id', default=None, group='id')
457
458 content_item_ids = set()
459 if content_item_id:
460 content_item_ids.add(content_item_id)
461 if video_id not in content_item_ids:
462 content_item_ids.add(video_id)
463
464 for content_item_id in content_item_ids:
465 try:
466 return self._extract_from_content_id(content_item_id, url)
467 except GeoRestrictedError:
468 raise
469 except ExtractorError:
470 pass
471
472 relinker_url = self._search_regex(
473 r'''(?x)
474 (?:
475 var\s+videoURL|
476 mediaInfo\.mediaUri
477 )\s*=\s*
478 ([\'"])
479 (?P<url>
480 (?:https?:)?
481 //mediapolis(?:vod)?\.rai\.it/relinker/relinkerServlet\.htm\?
482 (?:(?!\1).)*\bcont=(?:(?!\1).)+)\1
483 ''',
484 webpage, 'relinker URL', group='url')
485
486 relinker_info = self._extract_relinker_info(
487 urljoin(url, relinker_url), video_id)
488 self._sort_formats(relinker_info['formats'])
489
490 title = self._search_regex(
491 r'var\s+videoTitolo\s*=\s*([\'"])(?P<title>[^\'"]+)\1',
492 webpage, 'title', group='title',
493 default=None) or self._og_search_title(webpage)
494
495 info = {
496 'id': video_id,
497 'title': title,
498 }
499
500 info.update(relinker_info)
501
502 return info