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     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': self._live_title(title) if relinker_info.get(
 195                 'is_live') else title,
 196             'alt_title': media.get('subtitle'),
 197             'description': media.get('description'),
 198             'uploader': strip_or_none(media.get('channel')),
 199             'creator': strip_or_none(media.get('editor')),
 200             'duration': parse_duration(video.get('duration')),
 201             'timestamp': timestamp,
 202             'thumbnails': thumbnails,
 203             'series': try_get(
 204                 media, lambda x: x['isPartOf']['name'], compat_str),
 205             'season_number': int_or_none(try_get(
 206                 media, lambda x: x['isPartOf']['numeroStagioni'])),
 207             'season': media.get('stagione') or None,
 208             'subtitles': subtitles,
 209         }
 210
 211         info.update(relinker_info)
 212         return info
 213
 214
 215 class RaiPlayLiveIE(RaiBaseIE):
 216     _VALID_URL = r'https?://(?:www\.)?raiplay\.it/dirette/(?P<id>[^/?#&]+)'
 217     _TEST = {
 218         'url': 'http://www.raiplay.it/dirette/rainews24',
 219         'info_dict': {
 220             'id': 'd784ad40-e0ae-4a69-aa76-37519d238a9c',
 221             'display_id': 'rainews24',
 222             'ext': 'mp4',
 223             'title': 're:^Diretta di Rai News 24 [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
 224             'description': 'md5:6eca31500550f9376819f174e5644754',
 225             'uploader': 'Rai News 24',
 226             'creator': 'Rai News 24',
 227             'is_live': True,
 228         },
 229         'params': {
 230             'skip_download': True,
 231         },
 232     }
 233
 234     def _real_extract(self, url):
 235         display_id = self._match_id(url)
 236
 237         webpage = self._download_webpage(url, display_id)
 238
 239         video_id = self._search_regex(
 240             r'data-uniquename=["\']ContentItem-(%s)' % RaiBaseIE._UUID_RE,
 241             webpage, 'content id')
 242
 243         return {
 244             '_type': 'url_transparent',
 245             'ie_key': RaiPlayIE.ie_key(),
 246             'url': 'http://www.raiplay.it/dirette/ContentItem-%s.html' % video_id,
 247             'id': video_id,
 248             'display_id': display_id,
 249         }
 250
 251
 252 class RaiIE(RaiBaseIE):
 253     _VALID_URL = r'https?://[^/]+\.(?:rai\.(?:it|tv)|rainews\.it)/dl/.+?-(?P<id>%s)(?:-.+?)?\.html' % RaiBaseIE._UUID_RE
 254     _TESTS = [{
 255         # var uniquename = "ContentItem-..."
 256         # data-id="ContentItem-..."
 257         'url': 'http://www.raisport.rai.it/dl/raiSport/media/rassegna-stampa-04a9f4bd-b563-40cf-82a6-aad3529cb4a9.html',
 258         'info_dict': {
 259             'id': '04a9f4bd-b563-40cf-82a6-aad3529cb4a9',
 260             'ext': 'mp4',
 261             'title': 'TG PRIMO TEMPO',
 262             'thumbnail': r're:^https?://.*\.jpg$',
 263             'duration': 1758,
 264             'upload_date': '20140612',
 265         }
 266     }, {
 267         # with ContentItem in many metas
 268         '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',
 269         'info_dict': {
 270             'id': '1632c009-c843-4836-bb65-80c33084a64b',
 271             'ext': 'mp4',
 272             'title': 'Weekend al cinema, da Hollywood arriva il thriller di Tate Taylor "La ragazza del treno"',
 273             'description': 'I film in uscita questa settimana.',
 274             'thumbnail': r're:^https?://.*\.png$',
 275             'duration': 833,
 276             'upload_date': '20161103',
 277         }
 278     }, {
 279         # with ContentItem in og:url
 280         'url': 'http://www.rai.it/dl/RaiTV/programmi/media/ContentItem-efb17665-691c-45d5-a60c-5301333cbb0c.html',
 281         'md5': '11959b4e44fa74de47011b5799490adf',
 282         'info_dict': {
 283             'id': 'efb17665-691c-45d5-a60c-5301333cbb0c',
 284             'ext': 'mp4',
 285             'title': 'TG1 ore 20:00 del 03/11/2016',
 286             'description': 'TG1 edizione integrale ore 20:00 del giorno 03/11/2016',
 287             'thumbnail': r're:^https?://.*\.jpg$',
 288             'duration': 2214,
 289             'upload_date': '20161103',
 290         }
 291     }, {
 292         # drawMediaRaiTV(...)
 293         'url': 'http://www.report.rai.it/dl/Report/puntata/ContentItem-0c7a664b-d0f4-4b2c-8835-3f82e46f433e.html',
 294         'md5': '2dd727e61114e1ee9c47f0da6914e178',
 295         'info_dict': {
 296             'id': '59d69d28-6bb6-409d-a4b5-ed44096560af',
 297             'ext': 'mp4',
 298             'title': 'Il pacco',
 299             'description': 'md5:4b1afae1364115ce5d78ed83cd2e5b3a',
 300             'thumbnail': r're:^https?://.*\.jpg$',
 301             'upload_date': '20141221',
 302         },
 303     }, {
 304         # initEdizione('ContentItem-...'
 305         'url': 'http://www.tg1.rai.it/dl/tg1/2010/edizioni/ContentSet-9b6e0cba-4bef-4aef-8cf0-9f7f665b7dfb-tg1.html?item=undefined',
 306         'info_dict': {
 307             'id': 'c2187016-8484-4e3a-8ac8-35e475b07303',
 308             'ext': 'mp4',
 309             'title': r're:TG1 ore \d{2}:\d{2} del \d{2}/\d{2}/\d{4}',
 310             'duration': 2274,
 311             'upload_date': '20170401',
 312         },
 313         'skip': 'Changes daily',
 314     }, {
 315         # HDS live stream with only relinker URL
 316         'url': 'http://www.rai.tv/dl/RaiTV/dirette/PublishingBlock-1912dbbf-3f96-44c3-b4cf-523681fbacbc.html?channel=EuroNews',
 317         'info_dict': {
 318             'id': '1912dbbf-3f96-44c3-b4cf-523681fbacbc',
 319             'ext': 'flv',
 320             'title': 'EuroNews',
 321         },
 322         'params': {
 323             'skip_download': True,
 324         },
 325     }, {
 326         # HLS live stream with ContentItem in og:url
 327         'url': 'http://www.rainews.it/dl/rainews/live/ContentItem-3156f2f2-dc70-4953-8e2f-70d7489d4ce9.html',
 328         'info_dict': {
 329             'id': '3156f2f2-dc70-4953-8e2f-70d7489d4ce9',
 330             'ext': 'mp4',
 331             'title': 'La diretta di Rainews24',
 332         },
 333         'params': {
 334             'skip_download': True,
 335         },
 336     }]
 337
 338     def _extract_from_content_id(self, content_id, url):
 339         media = self._download_json(
 340             'http://www.rai.tv/dl/RaiTV/programmi/media/ContentItem-%s.html?json' % content_id,
 341             content_id, 'Downloading video JSON')
 342
 343         title = media['name'].strip()
 344
 345         media_type = media['type']
 346         if 'Audio' in media_type:
 347             relinker_info = {
 348                 'formats': [{
 349                     'format_id': media.get('formatoAudio'),
 350                     'url': media['audioUrl'],
 351                     'ext': media.get('formatoAudio'),
 352                 }]
 353             }
 354         elif 'Video' in media_type:
 355             relinker_info = self._extract_relinker_info(media['mediaUri'], content_id)
 356         else:
 357             raise ExtractorError('not a media file')
 358
 359         self._sort_formats(relinker_info['formats'])
 360
 361         thumbnails = []
 362         for image_type in ('image', 'image_medium', 'image_300'):
 363             thumbnail_url = media.get(image_type)
 364             if thumbnail_url:
 365                 thumbnails.append({
 366                     'url': compat_urlparse.urljoin(url, thumbnail_url),
 367                 })
 368
 369         subtitles = self._extract_subtitles(url, media.get('subtitlesUrl'))
 370
 371         info = {
 372             'id': content_id,
 373             'title': title,
 374             'description': strip_or_none(media.get('desc')),
 375             'thumbnails': thumbnails,
 376             'uploader': media.get('author'),
 377             'upload_date': unified_strdate(media.get('date')),
 378             'duration': parse_duration(media.get('length')),
 379             'subtitles': subtitles,
 380         }
 381
 382         info.update(relinker_info)
 383
 384         return info
 385
 386     def _real_extract(self, url):
 387         video_id = self._match_id(url)
 388
 389         webpage = self._download_webpage(url, video_id)
 390
 391         content_item_id = None
 392
 393         content_item_url = self._html_search_meta(
 394             ('og:url', 'og:video', 'og:video:secure_url', 'twitter:url',
 395              'twitter:player', 'jsonlink'), webpage, default=None)
 396         if content_item_url:
 397             content_item_id = self._search_regex(
 398                 r'ContentItem-(%s)' % self._UUID_RE, content_item_url,
 399                 'content item id', default=None)
 400
 401         if not content_item_id:
 402             content_item_id = self._search_regex(
 403                 r'''(?x)
 404                     (?:
 405                         (?:initEdizione|drawMediaRaiTV)\(|
 406                         <(?:[^>]+\bdata-id|var\s+uniquename)=
 407                     )
 408                     (["\'])
 409                     (?:(?!\1).)*\bContentItem-(?P<id>%s)
 410                 ''' % self._UUID_RE,
 411                 webpage, 'content item id', default=None, group='id')
 412
 413         content_item_ids = set()
 414         if content_item_id:
 415             content_item_ids.add(content_item_id)
 416         if video_id not in content_item_ids:
 417             content_item_ids.add(video_id)
 418
 419         for content_item_id in content_item_ids:
 420             try:
 421                 return self._extract_from_content_id(content_item_id, url)
 422             except GeoRestrictedError:
 423                 raise
 424             except ExtractorError:
 425                 pass
 426
 427         relinker_url = self._search_regex(
 428             r'''(?x)
 429                 (?:
 430                     var\s+videoURL|
 431                     mediaInfo\.mediaUri
 432                 )\s*=\s*
 433                 ([\'"])
 434                 (?P<url>
 435                     (?:https?:)?
 436                     //mediapolis(?:vod)?\.rai\.it/relinker/relinkerServlet\.htm\?
 437                     (?:(?!\1).)*\bcont=(?:(?!\1).)+)\1
 438             ''',
 439             webpage, 'relinker URL', group='url')
 440
 441         relinker_info = self._extract_relinker_info(
 442             urljoin(url, relinker_url), video_id)
 443         self._sort_formats(relinker_info['formats'])
 444
 445         title = self._search_regex(
 446             r'var\s+videoTitolo\s*=\s*([\'"])(?P<title>[^\'"]+)\1',
 447             webpage, 'title', group='title',
 448             default=None) or self._og_search_title(webpage)
 449
 450         info = {
 451             'id': video_id,
 452             'title': title,
 453         }
 454
 455         info.update(relinker_info)
 456
 457         return info