youtube_dlc/extractor/rai.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..compat import (
   8     compat_urlparse,
   9     compat_str,
  10 )
  11 from ..utils import (
  12     ExtractorError,
  13     determine_ext,
  14     find_xpath_attr,
  15     fix_xml_ampersands,
  16     GeoRestrictedError,
  17     int_or_none,
  18     parse_duration,
  19     strip_or_none,
  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|json))' % 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$',
 136             'uploader': 'Rai 3',
 137             'creator': 'Rai 3',
 138             'duration': 3278,
 139             'timestamp': 1477764300,
 140             'upload_date': '20161029',
 141             'series': 'La Casa Bianca',
 142             'season': '2016',
 143         },
 144         'skip': 'This content is not available',
 145     }, {
 146         'url': 'http://www.raiplay.it/video/2014/04/Report-del-07042014-cb27157f-9dd0-4aee-b788-b1f67643a391.html',
 147         'md5': '8970abf8caf8aef4696e7b1f2adfc696',
 148         'info_dict': {
 149             'id': 'cb27157f-9dd0-4aee-b788-b1f67643a391',
 150             'ext': 'mp4',
 151             'title': 'Report del 07/04/2014',
 152             'alt_title': 'St 2013/14 - Espresso nel caffè - 07/04/2014',
 153             'description': 'md5:d730c168a58f4bb35600fc2f881ec04e',
 154             'thumbnail': r're:^https?://.*\.jpg$',
 155             'uploader': 'Rai Gulp',
 156             'duration': 6160,
 157             'series': 'Report',
 158             'season': '2013/14',
 159         },
 160         'params': {
 161             'skip_download': True,
 162         },
 163     }, {
 164         'url': 'http://www.raiplay.it/video/2016/11/gazebotraindesi-efebe701-969c-4593-92f3-285f0d1ce750.html?',
 165         'only_matching': True,
 166     }]
 167
 168     def _real_extract(self, url):
 169         url, video_id = re.match(self._VALID_URL, url).groups()
 170
 171         media = self._download_json(
 172             url.replace('.html', '.json'), video_id, 'Downloading video JSON')
 173
 174         title = media['name']
 175         video = media['video']
 176
 177         relinker_info = self._extract_relinker_info(video['content_url'], video_id)
 178         self._sort_formats(relinker_info['formats'])
 179
 180         thumbnails = []
 181         for _, value in media.get('images', {}).items():
 182             if value:
 183                 thumbnails.append({
 184                     'url': urljoin(url, value),
 185                 })
 186
 187         date_published = media.get('date_published')
 188         time_published = media.get('time_published')
 189         if date_published and time_published:
 190             date_published += ' ' + time_published
 191
 192         subtitles = self._extract_subtitles(url, video.get('subtitles'))
 193
 194         program_info = media.get('program_info') or {}
 195         season = media.get('season')
 196
 197         info = {
 198             'id': video_id,
 199             'title': self._live_title(title) if relinker_info.get(
 200                 'is_live') else title,
 201             'alt_title': strip_or_none(media.get('subtitle')),
 202             'description': media.get('description'),
 203             'uploader': strip_or_none(media.get('channel')),
 204             'creator': strip_or_none(media.get('editor') or None),
 205             'duration': parse_duration(video.get('duration')),
 206             'timestamp': unified_timestamp(date_published),
 207             'thumbnails': thumbnails,
 208             'series': program_info.get('name'),
 209             'season_number': int_or_none(season),
 210             'season': season if (season and not season.isdigit()) else None,
 211             'episode': media.get('episode_title'),
 212             'episode_number': int_or_none(media.get('episode')),
 213             'subtitles': subtitles,
 214         }
 215
 216         info.update(relinker_info)
 217         return info
 218
 219
 220 class RaiPlayLiveIE(RaiBaseIE):
 221     _VALID_URL = r'https?://(?:www\.)?raiplay\.it/dirette/(?P<id>[^/?#&]+)'
 222     _TEST = {
 223         'url': 'http://www.raiplay.it/dirette/rainews24',
 224         'info_dict': {
 225             'id': 'd784ad40-e0ae-4a69-aa76-37519d238a9c',
 226             'display_id': 'rainews24',
 227             'ext': 'mp4',
 228             'title': 're:^Diretta di Rai News 24 [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
 229             'description': 'md5:6eca31500550f9376819f174e5644754',
 230             'uploader': 'Rai News 24',
 231             'creator': 'Rai News 24',
 232             'is_live': True,
 233         },
 234         'params': {
 235             'skip_download': True,
 236         },
 237     }
 238
 239     def _real_extract(self, url):
 240         display_id = self._match_id(url)
 241
 242         webpage = self._download_webpage(url, display_id)
 243
 244         video_id = self._search_regex(
 245             r'data-uniquename=["\']ContentItem-(%s)' % RaiBaseIE._UUID_RE,
 246             webpage, 'content id')
 247
 248         return {
 249             '_type': 'url_transparent',
 250             'ie_key': RaiPlayIE.ie_key(),
 251             'url': 'http://www.raiplay.it/dirette/ContentItem-%s.html' % video_id,
 252             'id': video_id,
 253             'display_id': display_id,
 254         }
 255
 256
 257 class RaiPlayPlaylistIE(InfoExtractor):
 258     _VALID_URL = r'https?://(?:www\.)?raiplay\.it/programmi/(?P<id>[^/?#&]+)'
 259     _TESTS = [{
 260         'url': 'http://www.raiplay.it/programmi/nondirloalmiocapo/',
 261         'info_dict': {
 262             'id': 'nondirloalmiocapo',
 263             'title': 'Non dirlo al mio capo',
 264             'description': 'md5:9f3d603b2947c1c7abb098f3b14fac86',
 265         },
 266         'playlist_mincount': 12,
 267     }]
 268
 269     def _real_extract(self, url):
 270         playlist_id = self._match_id(url)
 271
 272         webpage = self._download_webpage(url, playlist_id)
 273
 274         title = self._html_search_meta(
 275             ('programma', 'nomeProgramma'), webpage, 'title')
 276         description = unescapeHTML(self._html_search_meta(
 277             ('description', 'og:description'), webpage, '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)/.+?-(?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         'skip': 'This content is available only in Italy',
 306     }, {
 307         # with ContentItem in many metas
 308         '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',
 309         'info_dict': {
 310             'id': '1632c009-c843-4836-bb65-80c33084a64b',
 311             'ext': 'mp4',
 312             'title': 'Weekend al cinema, da Hollywood arriva il thriller di Tate Taylor "La ragazza del treno"',
 313             'description': 'I film in uscita questa settimana.',
 314             'thumbnail': r're:^https?://.*\.png$',
 315             'duration': 833,
 316             'upload_date': '20161103',
 317         }
 318     }, {
 319         # with ContentItem in og:url
 320         'url': 'http://www.rai.it/dl/RaiTV/programmi/media/ContentItem-efb17665-691c-45d5-a60c-5301333cbb0c.html',
 321         'md5': '6865dd00cf0bbf5772fdd89d59bd768a',
 322         'info_dict': {
 323             'id': 'efb17665-691c-45d5-a60c-5301333cbb0c',
 324             'ext': 'mp4',
 325             'title': 'TG1 ore 20:00 del 03/11/2016',
 326             'description': 'TG1 edizione integrale ore 20:00 del giorno 03/11/2016',
 327             'thumbnail': r're:^https?://.*\.jpg$',
 328             'duration': 2214,
 329             'upload_date': '20161103',
 330         }
 331     }, {
 332         # drawMediaRaiTV(...)
 333         'url': 'http://www.report.rai.it/dl/Report/puntata/ContentItem-0c7a664b-d0f4-4b2c-8835-3f82e46f433e.html',
 334         'md5': '2dd727e61114e1ee9c47f0da6914e178',
 335         'info_dict': {
 336             'id': '59d69d28-6bb6-409d-a4b5-ed44096560af',
 337             'ext': 'mp4',
 338             'title': 'Il pacco',
 339             'description': 'md5:4b1afae1364115ce5d78ed83cd2e5b3a',
 340             'thumbnail': r're:^https?://.*\.jpg$',
 341             'upload_date': '20141221',
 342         },
 343         'skip': 'This content is not available',
 344     }, {
 345         # initEdizione('ContentItem-...'
 346         'url': 'http://www.tg1.rai.it/dl/tg1/2010/edizioni/ContentSet-9b6e0cba-4bef-4aef-8cf0-9f7f665b7dfb-tg1.html?item=undefined',
 347         'info_dict': {
 348             'id': 'c2187016-8484-4e3a-8ac8-35e475b07303',
 349             'ext': 'mp4',
 350             'title': r're:TG1 ore \d{2}:\d{2} del \d{2}/\d{2}/\d{4}',
 351             'duration': 2274,
 352             'upload_date': '20170401',
 353         },
 354         'skip': 'Changes daily',
 355     }, {
 356         # HDS live stream with only relinker URL
 357         'url': 'http://www.rai.tv/dl/RaiTV/dirette/PublishingBlock-1912dbbf-3f96-44c3-b4cf-523681fbacbc.html?channel=EuroNews',
 358         'info_dict': {
 359             'id': '1912dbbf-3f96-44c3-b4cf-523681fbacbc',
 360             'ext': 'flv',
 361             'title': 'EuroNews',
 362         },
 363         'params': {
 364             'skip_download': True,
 365         },
 366         'skip': 'This content is available only in Italy',
 367     }, {
 368         # HLS live stream with ContentItem in og:url
 369         'url': 'http://www.rainews.it/dl/rainews/live/ContentItem-3156f2f2-dc70-4953-8e2f-70d7489d4ce9.html',
 370         'info_dict': {
 371             'id': '3156f2f2-dc70-4953-8e2f-70d7489d4ce9',
 372             'ext': 'mp4',
 373             'title': 'La diretta di Rainews24',
 374         },
 375         'params': {
 376             'skip_download': True,
 377         },
 378     }, {
 379         # Direct MMS URL
 380         'url': 'http://www.rai.it/dl/RaiTV/programmi/media/ContentItem-b63a4089-ac28-48cf-bca5-9f5b5bc46df5.html',
 381         'only_matching': True,
 382     }, {
 383         'url': 'https://www.rainews.it/tgr/marche/notiziari/video/2019/02/ContentItem-6ba945a2-889c-4a80-bdeb-8489c70a8db9.html',
 384         'only_matching': True,
 385     }]
 386
 387     def _extract_from_content_id(self, content_id, url):
 388         media = self._download_json(
 389             'http://www.rai.tv/dl/RaiTV/programmi/media/ContentItem-%s.html?json' % content_id,
 390             content_id, 'Downloading video JSON')
 391
 392         title = media['name'].strip()
 393
 394         media_type = media['type']
 395         if 'Audio' in media_type:
 396             relinker_info = {
 397                 'formats': [{
 398                     'format_id': media.get('formatoAudio'),
 399                     'url': media['audioUrl'],
 400                     'ext': media.get('formatoAudio'),
 401                 }]
 402             }
 403         elif 'Video' in media_type:
 404             relinker_info = self._extract_relinker_info(media['mediaUri'], content_id)
 405         else:
 406             raise ExtractorError('not a media file')
 407
 408         self._sort_formats(relinker_info['formats'])
 409
 410         thumbnails = []
 411         for image_type in ('image', 'image_medium', 'image_300'):
 412             thumbnail_url = media.get(image_type)
 413             if thumbnail_url:
 414                 thumbnails.append({
 415                     'url': compat_urlparse.urljoin(url, thumbnail_url),
 416                 })
 417
 418         subtitles = self._extract_subtitles(url, media.get('subtitlesUrl'))
 419
 420         info = {
 421             'id': content_id,
 422             'title': title,
 423             'description': strip_or_none(media.get('desc')),
 424             'thumbnails': thumbnails,
 425             'uploader': media.get('author'),
 426             'upload_date': unified_strdate(media.get('date')),
 427             'duration': parse_duration(media.get('length')),
 428             'subtitles': subtitles,
 429         }
 430
 431         info.update(relinker_info)
 432
 433         return info
 434
 435     def _real_extract(self, url):
 436         video_id = self._match_id(url)
 437
 438         webpage = self._download_webpage(url, video_id)
 439
 440         content_item_id = None
 441
 442         content_item_url = self._html_search_meta(
 443             ('og:url', 'og:video', 'og:video:secure_url', 'twitter:url',
 444              'twitter:player', 'jsonlink'), webpage, default=None)
 445         if content_item_url:
 446             content_item_id = self._search_regex(
 447                 r'ContentItem-(%s)' % self._UUID_RE, content_item_url,
 448                 'content item id', default=None)
 449
 450         if not content_item_id:
 451             content_item_id = self._search_regex(
 452                 r'''(?x)
 453                     (?:
 454                         (?:initEdizione|drawMediaRaiTV)\(|
 455                         <(?:[^>]+\bdata-id|var\s+uniquename)=
 456                     )
 457                     (["\'])
 458                     (?:(?!\1).)*\bContentItem-(?P<id>%s)
 459                 ''' % self._UUID_RE,
 460                 webpage, 'content item id', default=None, group='id')
 461
 462         content_item_ids = set()
 463         if content_item_id:
 464             content_item_ids.add(content_item_id)
 465         if video_id not in content_item_ids:
 466             content_item_ids.add(video_id)
 467
 468         for content_item_id in content_item_ids:
 469             try:
 470                 return self._extract_from_content_id(content_item_id, url)
 471             except GeoRestrictedError:
 472                 raise
 473             except ExtractorError:
 474                 pass
 475
 476         relinker_url = self._search_regex(
 477             r'''(?x)
 478                 (?:
 479                     var\s+videoURL|
 480                     mediaInfo\.mediaUri
 481                 )\s*=\s*
 482                 ([\'"])
 483                 (?P<url>
 484                     (?:https?:)?
 485                     //mediapolis(?:vod)?\.rai\.it/relinker/relinkerServlet\.htm\?
 486                     (?:(?!\1).)*\bcont=(?:(?!\1).)+)\1
 487             ''',
 488             webpage, 'relinker URL', group='url')
 489
 490         relinker_info = self._extract_relinker_info(
 491             urljoin(url, relinker_url), video_id)
 492         self._sort_formats(relinker_info['formats'])
 493
 494         title = self._search_regex(
 495             r'var\s+videoTitolo\s*=\s*([\'"])(?P<title>[^\'"]+)\1',
 496             webpage, 'title', group='title',
 497             default=None) or self._og_search_title(webpage)
 498
 499         info = {
 500             'id': video_id,
 501             'title': title,
 502         }
 503
 504         info.update(relinker_info)
 505
 506         return info