Raphaël G. Git Repositories - youtubedl/blob - 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':
  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':
  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         print(description)
 278
 279         entries = []
 280         for mobj in re.finditer(
 281                 r'<a\b[^>]+\bhref=(["\'])(?P<path>/raiplay/video/.+?)\1',
 282                 webpage):
 283             video_url = urljoin(url, mobj.group('path'))
 284             entries.append(self.url_result(
 285                 video_url, ie=RaiPlayIE.ie_key(),
 286                 video_id=RaiPlayIE._match_id(video_url)))
 287
 288         return self.playlist_result(entries, playlist_id, title, description)
 289
 290
 291 class RaiIE(RaiBaseIE):
 292     _VALID_URL = r'https?://[^/]+\.(?:rai\.(?:it|tv)|rainews\.it)/dl/.+?-(?P<id>%s)(?:-.+?)?\.html' % RaiBaseIE._UUID_RE
 293     _TESTS = [{
 294         # var uniquename = "ContentItem-..."
 295         # data-id="ContentItem-..."
 296         'url': 'http://www.raisport.rai.it/dl/raiSport/media/rassegna-stampa-04a9f4bd-b563-40cf-82a6-aad3529cb4a9.html',
 297         'info_dict': {
 298             'id': '04a9f4bd-b563-40cf-82a6-aad3529cb4a9',
 299             'ext': 'mp4',
 300             'title': 'TG PRIMO TEMPO',
 301             'thumbnail': r're:^https?://.*\.jpg$',
 302             'duration': 1758,
 303             'upload_date': '20140612',
 304         }
 305     }, {
 306         # with ContentItem in many metas
 307         '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',
 308         'info_dict': {
 309             'id': '1632c009-c843-4836-bb65-80c33084a64b',
 310             'ext': 'mp4',
 311             'title': 'Weekend al cinema, da Hollywood arriva il thriller di Tate Taylor "La ragazza del treno"',
 312             'description': 'I film in uscita questa settimana.',
 313             'thumbnail': r're:^https?://.*\.png$',
 314             'duration': 833,
 315             'upload_date': '20161103',
 316         }
 317     }, {
 318         # with ContentItem in og:url
 319         'url': 'http://www.rai.it/dl/RaiTV/programmi/media/ContentItem-efb17665-691c-45d5-a60c-5301333cbb0c.html',
 320         'md5': '11959b4e44fa74de47011b5799490adf',
 321         'info_dict': {
 322             'id': 'efb17665-691c-45d5-a60c-5301333cbb0c',
 323             'ext': 'mp4',
 324             'title': 'TG1 ore 20:00 del 03/11/2016',
 325             'description': 'TG1 edizione integrale ore 20:00 del giorno 03/11/2016',
 326             'thumbnail': r're:^https?://.*\.jpg$',
 327             'duration': 2214,
 328             'upload_date': '20161103',
 329         }
 330     }, {
 331         # drawMediaRaiTV(...)
 332         'url': 'http://www.report.rai.it/dl/Report/puntata/ContentItem-0c7a664b-d0f4-4b2c-8835-3f82e46f433e.html',
 333         'md5': '2dd727e61114e1ee9c47f0da6914e178',
 334         'info_dict': {
 335             'id': '59d69d28-6bb6-409d-a4b5-ed44096560af',
 336             'ext': 'mp4',
 337             'title': 'Il pacco',
 338             'description': 'md5:4b1afae1364115ce5d78ed83cd2e5b3a',
 339             'thumbnail': r're:^https?://.*\.jpg$',
 340             'upload_date': '20141221',
 341         },
 342     }, {
 343         # initEdizione('ContentItem-...'
 344         'url': 'http://www.tg1.rai.it/dl/tg1/2010/edizioni/ContentSet-9b6e0cba-4bef-4aef-8cf0-9f7f665b7dfb-tg1.html?item=undefined',
 345         'info_dict': {
 346             'id': 'c2187016-8484-4e3a-8ac8-35e475b07303',
 347             'ext': 'mp4',
 348             'title': r're:TG1 ore \d{2}:\d{2} del \d{2}/\d{2}/\d{4}',
 349             'duration': 2274,
 350             'upload_date': '20170401',
 351         },
 352         'skip': 'Changes daily',
 353     }, {
 354         # HDS live stream with only relinker URL
 355         'url': 'http://www.rai.tv/dl/RaiTV/dirette/PublishingBlock-1912dbbf-3f96-44c3-b4cf-523681fbacbc.html?channel=EuroNews',
 356         'info_dict': {
 357             'id': '1912dbbf-3f96-44c3-b4cf-523681fbacbc',
 358             'ext': 'flv',
 359             'title': 'EuroNews',
 360         },
 361         'params': {
 362             'skip_download': True,
 363         },
 364     }, {
 365         # HLS live stream with ContentItem in og:url
 366         'url': 'http://www.rainews.it/dl/rainews/live/ContentItem-3156f2f2-dc70-4953-8e2f-70d7489d4ce9.html',
 367         'info_dict': {
 368             'id': '3156f2f2-dc70-4953-8e2f-70d7489d4ce9',
 369             'ext': 'mp4',
 370             'title': 'La diretta di Rainews24',
 371         },
 372         'params': {
 373             'skip_download': True,
 374         },
 375     }, {
 376         # Direct MMS URL
 377         'url': 'http://www.rai.it/dl/RaiTV/programmi/media/ContentItem-b63a4089-ac28-48cf-bca5-9f5b5bc46df5.html',
 378         'only_matching': True,
 379     }]
 380
 381     def _extract_from_content_id(self, content_id, url):
 382         media = self._download_json(
 383             'http://www.rai.tv/dl/RaiTV/programmi/media/ContentItem-%s.html?json' % content_id,
 384             content_id, 'Downloading video JSON')
 385
 386         title = media['name'].strip()
 387
 388         media_type = media['type']
 389         if 'Audio' in media_type:
 390             relinker_info = {
 391                 'formats': [{
 392                     'format_id': media.get('formatoAudio'),
 393                     'url': media['audioUrl'],
 394                     'ext': media.get('formatoAudio'),
 395                 }]
 396             }
 397         elif 'Video' in media_type:
 398             relinker_info = self._extract_relinker_info(media['mediaUri'], content_id)
 399         else:
 400             raise ExtractorError('not a media file')
 401
 402         self._sort_formats(relinker_info['formats'])
 403
 404         thumbnails = []
 405         for image_type in ('image', 'image_medium', 'image_300'):
 406             thumbnail_url = media.get(image_type)
 407             if thumbnail_url:
 408                 thumbnails.append({
 409                     'url': compat_urlparse.urljoin(url, thumbnail_url),
 410                 })
 411
 412         subtitles = self._extract_subtitles(url, media.get('subtitlesUrl'))
 413
 414         info = {
 415             'id': content_id,
 416             'title': title,
 417             'description': strip_or_none(media.get('desc')),
 418             'thumbnails': thumbnails,
 419             'uploader': media.get('author'),
 420             'upload_date': unified_strdate(media.get('date')),
 421             'duration': parse_duration(media.get('length')),
 422             'subtitles': subtitles,
 423         }
 424
 425         info.update(relinker_info)
 426
 427         return info
 428
 429     def _real_extract(self, url):
 430         video_id = self._match_id(url)
 431
 432         webpage = self._download_webpage(url, video_id)
 433
 434         content_item_id = None
 435
 436         content_item_url = self._html_search_meta(
 437             ('og:url', 'og:video', 'og:video:secure_url', 'twitter:url',
 438              'twitter:player', 'jsonlink'), webpage, default=None)
 439         if content_item_url:
 440             content_item_id = self._search_regex(
 441                 r'ContentItem-(%s)' % self._UUID_RE, content_item_url,
 442                 'content item id', default=None)
 443
 444         if not content_item_id:
 445             content_item_id = self._search_regex(
 446                 r'''(?x)
 447                     (?:
 448                         (?:initEdizione|drawMediaRaiTV)\(|
 449                         <(?:[^>]+\bdata-id|var\s+uniquename)=
 450                     )
 451                     (["\'])
 452                     (?:(?!\1).)*\bContentItem-(?P<id>%s)
 453                 ''' % self._UUID_RE,
 454                 webpage, 'content item id', default=None, group='id')
 455
 456         content_item_ids = set()
 457         if content_item_id:
 458             content_item_ids.add(content_item_id)
 459         if video_id not in content_item_ids:
 460             content_item_ids.add(video_id)
 461
 462         for content_item_id in content_item_ids:
 463             try:
 464                 return self._extract_from_content_id(content_item_id, url)
 465             except GeoRestrictedError:
 466                 raise
 467             except ExtractorError:
 468                 pass
 469
 470         relinker_url = self._search_regex(
 471             r'''(?x)
 472                 (?:
 473                     var\s+videoURL|
 474                     mediaInfo\.mediaUri
 475                 )\s*=\s*
 476                 ([\'"])
 477                 (?P<url>
 478                     (?:https?:)?
 479                     //mediapolis(?:vod)?\.rai\.it/relinker/relinkerServlet\.htm\?
 480                     (?:(?!\1).)*\bcont=(?:(?!\1).)+)\1
 481             ''',
 482             webpage, 'relinker URL', group='url')
 483
 484         relinker_info = self._extract_relinker_info(
 485             urljoin(url, relinker_url), video_id)
 486         self._sort_formats(relinker_info['formats'])
 487
 488         title = self._search_regex(
 489             r'var\s+videoTitolo\s*=\s*([\'"])(?P<title>[^\'"]+)\1',
 490             webpage, 'title', group='title',
 491             default=None) or self._og_search_title(webpage)
 492
 493         info = {
 494             'id': video_id,
 495             'title': title,
 496         }
 497
 498         info.update(relinker_info)
 499
 500         return info