]> Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/extractor/rai.py
81eb9db85b9127e292894e22c9dd6272b134832a
[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 unified_strdate,
21 unified_timestamp,
22 update_url_query,
23 urljoin,
24 xpath_text,
25 )
26
27
28 class RaiBaseIE(InfoExtractor):
29 _UUID_RE = r'[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12}'
30 _GEO_COUNTRIES = ['IT']
31 _GEO_BYPASS = False
32
33 def _extract_relinker_info(self, relinker_url, video_id):
34 formats = []
35 geoprotection = None
36 is_live = None
37 duration = None
38
39 for platform in ('mon', 'flash', 'native'):
40 relinker = self._download_xml(
41 relinker_url, video_id,
42 note='Downloading XML metadata for platform %s' % platform,
43 transform_source=fix_xml_ampersands,
44 query={'output': 45, 'pl': platform},
45 headers=self.geo_verification_headers())
46
47 if not geoprotection:
48 geoprotection = xpath_text(
49 relinker, './geoprotection', default=None) == 'Y'
50
51 if not is_live:
52 is_live = xpath_text(
53 relinker, './is_live', default=None) == 'Y'
54 if not duration:
55 duration = parse_duration(xpath_text(
56 relinker, './duration', default=None))
57
58 url_elem = find_xpath_attr(relinker, './url', 'type', 'content')
59 if url_elem is None:
60 continue
61
62 media_url = url_elem.text
63
64 # This does not imply geo restriction (e.g.
65 # http://www.raisport.rai.it/dl/raiSport/media/rassegna-stampa-04a9f4bd-b563-40cf-82a6-aad3529cb4a9.html)
66 if media_url == 'http://download.rai.it/video_no_available.mp4':
67 continue
68
69 ext = determine_ext(media_url)
70 if (ext == 'm3u8' and platform != 'mon') or (ext == 'f4m' and platform != 'flash'):
71 continue
72
73 if ext == 'm3u8':
74 formats.extend(self._extract_m3u8_formats(
75 media_url, video_id, 'mp4', 'm3u8_native',
76 m3u8_id='hls', fatal=False))
77 elif ext == 'f4m':
78 manifest_url = update_url_query(
79 media_url.replace('manifest#live_hds.f4m', 'manifest.f4m'),
80 {'hdcore': '3.7.0', 'plugin': 'aasp-3.7.0.39.44'})
81 formats.extend(self._extract_f4m_formats(
82 manifest_url, video_id, f4m_id='hds', fatal=False))
83 else:
84 bitrate = int_or_none(xpath_text(relinker, 'bitrate'))
85 formats.append({
86 'url': media_url,
87 'tbr': bitrate if bitrate > 0 else None,
88 'format_id': 'http-%d' % bitrate if bitrate > 0 else 'http',
89 })
90
91 if not formats and geoprotection is True:
92 self.raise_geo_restricted(countries=self._GEO_COUNTRIES)
93
94 return dict((k, v) for k, v in {
95 'is_live': is_live,
96 'duration': duration,
97 'formats': formats,
98 }.items() if v is not None)
99
100 @staticmethod
101 def _extract_subtitles(url, subtitle_url):
102 subtitles = {}
103 if subtitle_url and isinstance(subtitle_url, compat_str):
104 subtitle_url = urljoin(url, subtitle_url)
105 STL_EXT = '.stl'
106 SRT_EXT = '.srt'
107 subtitles['it'] = [{
108 'ext': 'stl',
109 'url': subtitle_url,
110 }]
111 if subtitle_url.endswith(STL_EXT):
112 srt_url = subtitle_url[:-len(STL_EXT)] + SRT_EXT
113 subtitles['it'].append({
114 'ext': 'srt',
115 'url': srt_url,
116 })
117 return subtitles
118
119
120 class RaiPlayIE(RaiBaseIE):
121 _VALID_URL = r'(?P<url>https?://(?:www\.)?raiplay\.it/.+?-(?P<id>%s)\.html)' % RaiBaseIE._UUID_RE
122 _TESTS = [{
123 'url': 'http://www.raiplay.it/video/2016/10/La-Casa-Bianca-e06118bb-59a9-4636-b914-498e4cfd2c66.html?source=twitter',
124 'md5': '340aa3b7afb54bfd14a8c11786450d76',
125 'info_dict': {
126 'id': 'e06118bb-59a9-4636-b914-498e4cfd2c66',
127 'ext': 'mp4',
128 'title': 'La Casa Bianca',
129 'alt_title': 'S2016 - Puntata del 23/10/2016',
130 'description': 'md5:a09d45890850458077d1f68bb036e0a5',
131 'thumbnail': r're:^https?://.*\.jpg$',
132 'uploader': 'Rai 3',
133 'creator': 'Rai 3',
134 'duration': 3278,
135 'timestamp': 1477764300,
136 'upload_date': '20161029',
137 'series': 'La Casa Bianca',
138 'season': '2016',
139 },
140 }, {
141 'url': 'http://www.raiplay.it/video/2014/04/Report-del-07042014-cb27157f-9dd0-4aee-b788-b1f67643a391.html',
142 'md5': '8970abf8caf8aef4696e7b1f2adfc696',
143 'info_dict': {
144 'id': 'cb27157f-9dd0-4aee-b788-b1f67643a391',
145 'ext': 'mp4',
146 'title': 'Report del 07/04/2014',
147 'alt_title': 'S2013/14 - Puntata del 07/04/2014',
148 'description': 'md5:f27c544694cacb46a078db84ec35d2d9',
149 'thumbnail': r're:^https?://.*\.jpg$',
150 'uploader': 'Rai 5',
151 'creator': 'Rai 5',
152 'duration': 6160,
153 'series': 'Report',
154 'season_number': 5,
155 'season': '2013/14',
156 },
157 'params': {
158 'skip_download': True,
159 },
160 }, {
161 'url': 'http://www.raiplay.it/video/2016/11/gazebotraindesi-efebe701-969c-4593-92f3-285f0d1ce750.html?',
162 'only_matching': True,
163 }]
164
165 def _real_extract(self, url):
166 mobj = re.match(self._VALID_URL, url)
167 url, video_id = mobj.group('url', 'id')
168
169 media = self._download_json(
170 '%s?json' % url, video_id, 'Downloading video JSON')
171
172 title = media['name']
173
174 video = media['video']
175
176 relinker_info = self._extract_relinker_info(video['contentUrl'], video_id)
177 self._sort_formats(relinker_info['formats'])
178
179 thumbnails = []
180 if 'images' in media:
181 for _, value in media.get('images').items():
182 if value:
183 thumbnails.append({
184 'url': value.replace('[RESOLUTION]', '600x400')
185 })
186
187 timestamp = unified_timestamp(try_get(
188 media, lambda x: x['availabilities'][0]['start'], compat_str))
189
190 subtitles = self._extract_subtitles(url, video.get('subtitles'))
191
192 info = {
193 'id': video_id,
194 'title': title,
195 'alt_title': media.get('subtitle'),
196 'description': media.get('description'),
197 'uploader': media.get('channel'),
198 'creator': media.get('editor'),
199 'duration': parse_duration(video.get('duration')),
200 'timestamp': timestamp,
201 'thumbnails': thumbnails,
202 'series': try_get(
203 media, lambda x: x['isPartOf']['name'], compat_str),
204 'season_number': int_or_none(try_get(
205 media, lambda x: x['isPartOf']['numeroStagioni'])),
206 'season': media.get('stagione') or None,
207 'subtitles': subtitles,
208 }
209
210 info.update(relinker_info)
211
212 return info
213
214
215 class RaiIE(RaiBaseIE):
216 _VALID_URL = r'https?://[^/]+\.(?:rai\.(?:it|tv)|rainews\.it)/dl/.+?-(?P<id>%s)(?:-.+?)?\.html' % RaiBaseIE._UUID_RE
217 _TESTS = [{
218 # var uniquename = "ContentItem-..."
219 # data-id="ContentItem-..."
220 'url': 'http://www.raisport.rai.it/dl/raiSport/media/rassegna-stampa-04a9f4bd-b563-40cf-82a6-aad3529cb4a9.html',
221 'info_dict': {
222 'id': '04a9f4bd-b563-40cf-82a6-aad3529cb4a9',
223 'ext': 'mp4',
224 'title': 'TG PRIMO TEMPO',
225 'thumbnail': r're:^https?://.*\.jpg$',
226 'duration': 1758,
227 'upload_date': '20140612',
228 }
229 }, {
230 # with ContentItem in many metas
231 'url': 'http://www.rainews.it/dl/rainews/media/Weekend-al-cinema-da-Hollywood-arriva-il-thriller-di-Tate-Taylor-La-ragazza-del-treno-1632c009-c843-4836-bb65-80c33084a64b.html',
232 'info_dict': {
233 'id': '1632c009-c843-4836-bb65-80c33084a64b',
234 'ext': 'mp4',
235 'title': 'Weekend al cinema, da Hollywood arriva il thriller di Tate Taylor "La ragazza del treno"',
236 'description': 'I film in uscita questa settimana.',
237 'thumbnail': r're:^https?://.*\.png$',
238 'duration': 833,
239 'upload_date': '20161103',
240 }
241 }, {
242 # with ContentItem in og:url
243 'url': 'http://www.rai.it/dl/RaiTV/programmi/media/ContentItem-efb17665-691c-45d5-a60c-5301333cbb0c.html',
244 'md5': '11959b4e44fa74de47011b5799490adf',
245 'info_dict': {
246 'id': 'efb17665-691c-45d5-a60c-5301333cbb0c',
247 'ext': 'mp4',
248 'title': 'TG1 ore 20:00 del 03/11/2016',
249 'description': 'TG1 edizione integrale ore 20:00 del giorno 03/11/2016',
250 'thumbnail': r're:^https?://.*\.jpg$',
251 'duration': 2214,
252 'upload_date': '20161103',
253 }
254 }, {
255 # drawMediaRaiTV(...)
256 'url': 'http://www.report.rai.it/dl/Report/puntata/ContentItem-0c7a664b-d0f4-4b2c-8835-3f82e46f433e.html',
257 'md5': '2dd727e61114e1ee9c47f0da6914e178',
258 'info_dict': {
259 'id': '59d69d28-6bb6-409d-a4b5-ed44096560af',
260 'ext': 'mp4',
261 'title': 'Il pacco',
262 'description': 'md5:4b1afae1364115ce5d78ed83cd2e5b3a',
263 'thumbnail': r're:^https?://.*\.jpg$',
264 'upload_date': '20141221',
265 },
266 }, {
267 # initEdizione('ContentItem-...'
268 'url': 'http://www.tg1.rai.it/dl/tg1/2010/edizioni/ContentSet-9b6e0cba-4bef-4aef-8cf0-9f7f665b7dfb-tg1.html?item=undefined',
269 'info_dict': {
270 'id': 'c2187016-8484-4e3a-8ac8-35e475b07303',
271 'ext': 'mp4',
272 'title': r're:TG1 ore \d{2}:\d{2} del \d{2}/\d{2}/\d{4}',
273 'duration': 2274,
274 'upload_date': '20170401',
275 },
276 'skip': 'Changes daily',
277 }, {
278 # HDS live stream with only relinker URL
279 'url': 'http://www.rai.tv/dl/RaiTV/dirette/PublishingBlock-1912dbbf-3f96-44c3-b4cf-523681fbacbc.html?channel=EuroNews',
280 'info_dict': {
281 'id': '1912dbbf-3f96-44c3-b4cf-523681fbacbc',
282 'ext': 'flv',
283 'title': 'EuroNews',
284 },
285 'params': {
286 'skip_download': True,
287 },
288 }, {
289 # HLS live stream with ContentItem in og:url
290 'url': 'http://www.rainews.it/dl/rainews/live/ContentItem-3156f2f2-dc70-4953-8e2f-70d7489d4ce9.html',
291 'info_dict': {
292 'id': '3156f2f2-dc70-4953-8e2f-70d7489d4ce9',
293 'ext': 'mp4',
294 'title': 'La diretta di Rainews24',
295 },
296 'params': {
297 'skip_download': True,
298 },
299 }]
300
301 def _extract_from_content_id(self, content_id, url):
302 media = self._download_json(
303 'http://www.rai.tv/dl/RaiTV/programmi/media/ContentItem-%s.html?json' % content_id,
304 content_id, 'Downloading video JSON')
305
306 title = media['name'].strip()
307
308 media_type = media['type']
309 if 'Audio' in media_type:
310 relinker_info = {
311 'formats': {
312 'format_id': media.get('formatoAudio'),
313 'url': media['audioUrl'],
314 'ext': media.get('formatoAudio'),
315 }
316 }
317 elif 'Video' in media_type:
318 relinker_info = self._extract_relinker_info(media['mediaUri'], content_id)
319 else:
320 raise ExtractorError('not a media file')
321
322 self._sort_formats(relinker_info['formats'])
323
324 thumbnails = []
325 for image_type in ('image', 'image_medium', 'image_300'):
326 thumbnail_url = media.get(image_type)
327 if thumbnail_url:
328 thumbnails.append({
329 'url': compat_urlparse.urljoin(url, thumbnail_url),
330 })
331
332 subtitles = self._extract_subtitles(url, media.get('subtitlesUrl'))
333
334 info = {
335 'id': content_id,
336 'title': title,
337 'description': strip_or_none(media.get('desc')),
338 'thumbnails': thumbnails,
339 'uploader': media.get('author'),
340 'upload_date': unified_strdate(media.get('date')),
341 'duration': parse_duration(media.get('length')),
342 'subtitles': subtitles,
343 }
344
345 info.update(relinker_info)
346
347 return info
348
349 def _real_extract(self, url):
350 video_id = self._match_id(url)
351
352 webpage = self._download_webpage(url, video_id)
353
354 content_item_id = None
355
356 content_item_url = self._html_search_meta(
357 ('og:url', 'og:video', 'og:video:secure_url', 'twitter:url',
358 'twitter:player', 'jsonlink'), webpage, default=None)
359 if content_item_url:
360 content_item_id = self._search_regex(
361 r'ContentItem-(%s)' % self._UUID_RE, content_item_url,
362 'content item id', default=None)
363
364 if not content_item_id:
365 content_item_id = self._search_regex(
366 r'''(?x)
367 (?:
368 (?:initEdizione|drawMediaRaiTV)\(|
369 <(?:[^>]+\bdata-id|var\s+uniquename)=
370 )
371 (["\'])
372 (?:(?!\1).)*\bContentItem-(?P<id>%s)
373 ''' % self._UUID_RE,
374 webpage, 'content item id', default=None, group='id')
375
376 content_item_ids = set()
377 if content_item_id:
378 content_item_ids.add(content_item_id)
379 if video_id not in content_item_ids:
380 content_item_ids.add(video_id)
381
382 for content_item_id in content_item_ids:
383 try:
384 return self._extract_from_content_id(content_item_id, url)
385 except GeoRestrictedError:
386 raise
387 except ExtractorError:
388 pass
389
390 relinker_url = self._search_regex(
391 r'''(?x)
392 (?:
393 var\s+videoURL|
394 mediaInfo\.mediaUri
395 )\s*=\s*
396 ([\'"])
397 (?P<url>
398 (?:https?:)?
399 //mediapolis(?:vod)?\.rai\.it/relinker/relinkerServlet\.htm\?
400 (?:(?!\1).)*\bcont=(?:(?!\1).)+)\1
401 ''',
402 webpage, 'relinker URL', group='url')
403
404 relinker_info = self._extract_relinker_info(
405 urljoin(url, relinker_url), video_id)
406 self._sort_formats(relinker_info['formats'])
407
408 title = self._search_regex(
409 r'var\s+videoTitolo\s*=\s*([\'"])(?P<title>[^\'"]+)\1',
410 webpage, 'title', group='title',
411 default=None) or self._og_search_title(webpage)
412
413 info = {
414 'id': video_id,
415 'title': title,
416 }
417
418 info.update(relinker_info)
419
420 return info