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