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     remove_start,
  20     strip_or_none,
  21     try_get,
  22     unified_strdate,
  23     unified_timestamp,
  24     update_url_query,
  25     urljoin,
  26     xpath_text,
  27 )
  28
  29
  30 class RaiBaseIE(InfoExtractor):
  31     _UUID_RE = r'[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12}'
  32     _GEO_COUNTRIES = ['IT']
  33     _GEO_BYPASS = False
  34
  35     def _extract_relinker_info(self, relinker_url, video_id):
  36         if not re.match(r'https?://', relinker_url):
  37             return {'formats': [{'url': relinker_url}]}
  38
  39         formats = []
  40         geoprotection = None
  41         is_live = None
  42         duration = None
  43
  44         for platform in ('mon', 'flash', 'native'):
  45             relinker = self._download_xml(
  46                 relinker_url, video_id,
  47                 note='Downloading XML metadata for platform %s' % platform,
  48                 transform_source=fix_xml_ampersands,
  49                 query={'output': 45, 'pl': platform},
  50                 headers=self.geo_verification_headers())
  51
  52             if not geoprotection:
  53                 geoprotection = xpath_text(
  54                     relinker, './geoprotection', default=None) == 'Y'
  55
  56             if not is_live:
  57                 is_live = xpath_text(
  58                     relinker, './is_live', default=None) == 'Y'
  59             if not duration:
  60                 duration = parse_duration(xpath_text(
  61                     relinker, './duration', default=None))
  62
  63             url_elem = find_xpath_attr(relinker, './url', 'type', 'content')
  64             if url_elem is None:
  65                 continue
  66
  67             media_url = url_elem.text
  68
  69             # This does not imply geo restriction (e.g.
  70             # http://www.raisport.rai.it/dl/raiSport/media/rassegna-stampa-04a9f4bd-b563-40cf-82a6-aad3529cb4a9.html)
  71             if '/video_no_available.mp4' in media_url:
  72                 continue
  73
  74             ext = determine_ext(media_url)
  75             if (ext == 'm3u8' and platform != 'mon') or (ext == 'f4m' and platform != 'flash'):
  76                 continue
  77
  78             if ext == 'm3u8' or 'format=m3u8' in media_url or platform == 'mon':
  79                 formats.extend(self._extract_m3u8_formats(
  80                     media_url, video_id, 'mp4', 'm3u8_native',
  81                     m3u8_id='hls', fatal=False))
  82             elif ext == 'f4m' or platform == 'flash':
  83                 manifest_url = update_url_query(
  84                     media_url.replace('manifest#live_hds.f4m', 'manifest.f4m'),
  85                     {'hdcore': '3.7.0', 'plugin': 'aasp-3.7.0.39.44'})
  86                 formats.extend(self._extract_f4m_formats(
  87                     manifest_url, video_id, f4m_id='hds', fatal=False))
  88             else:
  89                 bitrate = int_or_none(xpath_text(relinker, 'bitrate'))
  90                 formats.append({
  91                     'url': media_url,
  92                     'tbr': bitrate if bitrate > 0 else None,
  93                     'format_id': 'http-%d' % bitrate if bitrate > 0 else 'http',
  94                 })
  95
  96         if not formats and geoprotection is True:
  97             self.raise_geo_restricted(countries=self._GEO_COUNTRIES)
  98
  99         return dict((k, v) for k, v in {
 100             'is_live': is_live,
 101             'duration': duration,
 102             'formats': formats,
 103         }.items() if v is not None)
 104
 105     @staticmethod
 106     def _extract_subtitles(url, subtitle_url):
 107         subtitles = {}
 108         if subtitle_url and isinstance(subtitle_url, compat_str):
 109             subtitle_url = urljoin(url, subtitle_url)
 110             STL_EXT = '.stl'
 111             SRT_EXT = '.srt'
 112             subtitles['it'] = [{
 113                 'ext': 'stl',
 114                 'url': subtitle_url,
 115             }]
 116             if subtitle_url.endswith(STL_EXT):
 117                 srt_url = subtitle_url[:-len(STL_EXT)] + SRT_EXT
 118                 subtitles['it'].append({
 119                     'ext': 'srt',
 120                     'url': srt_url,
 121                 })
 122         return subtitles
 123
 124
 125 class RaiPlayIE(RaiBaseIE):
 126     _VALID_URL = r'(?P<base>https?://(?:www\.)?raiplay\.it/.+?-(?P<id>%s))\.(?:html|json)' % RaiBaseIE._UUID_RE
 127     _TESTS = [{
 128         'url': 'http://www.raiplay.it/video/2014/04/Report-del-07042014-cb27157f-9dd0-4aee-b788-b1f67643a391.html',
 129         'md5': '8970abf8caf8aef4696e7b1f2adfc696',
 130         'info_dict': {
 131             'id': 'cb27157f-9dd0-4aee-b788-b1f67643a391',
 132             'ext': 'mp4',
 133             'title': 'Report del 07/04/2014',
 134             'alt_title': 'St 2013/14 - Espresso nel caffè - 07/04/2014',
 135             'description': 'md5:d730c168a58f4bb35600fc2f881ec04e',
 136             'thumbnail': r're:^https?://.*\.jpg$',
 137             'uploader': 'Rai Gulp',
 138             'duration': 6160,
 139             'series': 'Report',
 140             'season': '2013/14',
 141         },
 142         'params': {
 143             'skip_download': True,
 144         },
 145     }, {
 146         'url': 'http://www.raiplay.it/video/2016/11/gazebotraindesi-efebe701-969c-4593-92f3-285f0d1ce750.html?',
 147         'only_matching': True,
 148     }]
 149
 150     def _real_extract(self, url):
 151         base, video_id = re.match(self._VALID_URL, url).groups()
 152
 153         media = self._download_json(
 154             base + '.json', video_id, 'Downloading video JSON')
 155
 156         title = media['name']
 157         video = media['video']
 158
 159         relinker_info = self._extract_relinker_info(video['content_url'], video_id)
 160         self._sort_formats(relinker_info['formats'])
 161
 162         thumbnails = []
 163         for _, value in media.get('images', {}).items():
 164             if value:
 165                 thumbnails.append({
 166                     'url': urljoin(url, value),
 167                 })
 168
 169         date_published = media.get('date_published')
 170         time_published = media.get('time_published')
 171         if date_published and time_published:
 172             date_published += ' ' + time_published
 173
 174         subtitles = self._extract_subtitles(url, video.get('subtitles'))
 175
 176         program_info = media.get('program_info') or {}
 177         season = media.get('season')
 178
 179         info = {
 180             'id': remove_start(media.get('id'), 'ContentItem-') or video_id,
 181             'display_id': video_id,
 182             'title': self._live_title(title) if relinker_info.get(
 183                 'is_live') else title,
 184             'alt_title': strip_or_none(media.get('subtitle')),
 185             'description': media.get('description'),
 186             'uploader': strip_or_none(media.get('channel')),
 187             'creator': strip_or_none(media.get('editor') or None),
 188             'duration': parse_duration(video.get('duration')),
 189             'timestamp': unified_timestamp(date_published),
 190             'thumbnails': thumbnails,
 191             'series': program_info.get('name'),
 192             'season_number': int_or_none(season),
 193             'season': season if (season and not season.isdigit()) else None,
 194             'episode': media.get('episode_title'),
 195             'episode_number': int_or_none(media.get('episode')),
 196             'subtitles': subtitles,
 197         }
 198
 199         info.update(relinker_info)
 200         return info
 201
 202
 203 class RaiPlayLiveIE(RaiPlayIE):
 204     _VALID_URL = r'(?P<base>https?://(?:www\.)?raiplay\.it/dirette/(?P<id>[^/?#&]+))'
 205     _TESTS = [{
 206         'url': 'http://www.raiplay.it/dirette/rainews24',
 207         'info_dict': {
 208             'id': 'd784ad40-e0ae-4a69-aa76-37519d238a9c',
 209             'display_id': 'rainews24',
 210             'ext': 'mp4',
 211             'title': 're:^Diretta di Rai News 24 [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
 212             'description': 'md5:4d00bcf6dc98b27c6ec480de329d1497',
 213             'uploader': 'Rai News 24',
 214             'creator': 'Rai News 24',
 215             'is_live': True,
 216         },
 217         'params': {
 218             'skip_download': True,
 219         },
 220     }]
 221
 222
 223 class RaiPlayPlaylistIE(InfoExtractor):
 224     _VALID_URL = r'(?P<base>https?://(?:www\.)?raiplay\.it/programmi/(?P<id>[^/?#&]+))'
 225     _TESTS = [{
 226         'url': 'http://www.raiplay.it/programmi/nondirloalmiocapo/',
 227         'info_dict': {
 228             'id': 'nondirloalmiocapo',
 229             'title': 'Non dirlo al mio capo',
 230             'description': 'md5:98ab6b98f7f44c2843fd7d6f045f153b',
 231         },
 232         'playlist_mincount': 12,
 233     }]
 234
 235     def _real_extract(self, url):
 236         base, playlist_id = re.match(self._VALID_URL, url).groups()
 237
 238         program = self._download_json(
 239             base + '.json', playlist_id, 'Downloading program JSON')
 240
 241         entries = []
 242         for b in (program.get('blocks') or []):
 243             for s in (b.get('sets') or []):
 244                 s_id = s.get('id')
 245                 if not s_id:
 246                     continue
 247                 medias = self._download_json(
 248                     '%s/%s.json' % (base, s_id), s_id,
 249                     'Downloading content set JSON', fatal=False)
 250                 if not medias:
 251                     continue
 252                 for m in (medias.get('items') or []):
 253                     path_id = m.get('path_id')
 254                     if not path_id:
 255                         continue
 256                     video_url = urljoin(url, path_id)
 257                     entries.append(self.url_result(
 258                         video_url, ie=RaiPlayIE.ie_key(),
 259                         video_id=RaiPlayIE._match_id(video_url)))
 260
 261         return self.playlist_result(
 262             entries, playlist_id, program.get('name'),
 263             try_get(program, lambda x: x['program_info']['description']))
 264
 265
 266 class RaiIE(RaiBaseIE):
 267     _VALID_URL = r'https?://[^/]+\.(?:rai\.(?:it|tv)|rainews\.it)/.+?-(?P<id>%s)(?:-.+?)?\.html' % RaiBaseIE._UUID_RE
 268     _TESTS = [{
 269         # var uniquename = "ContentItem-..."
 270         # data-id="ContentItem-..."
 271         'url': 'http://www.raisport.rai.it/dl/raiSport/media/rassegna-stampa-04a9f4bd-b563-40cf-82a6-aad3529cb4a9.html',
 272         'info_dict': {
 273             'id': '04a9f4bd-b563-40cf-82a6-aad3529cb4a9',
 274             'ext': 'mp4',
 275             'title': 'TG PRIMO TEMPO',
 276             'thumbnail': r're:^https?://.*\.jpg$',
 277             'duration': 1758,
 278             'upload_date': '20140612',
 279         },
 280         'skip': 'This content is available only in Italy',
 281     }, {
 282         # with ContentItem in many metas
 283         'url': 'http://www.rainews.it/dl/rainews/media/Weekend-al-cinema-da-Hollywood-arriva-il-thriller-di-Tate-Taylor-La-ragazza-del-treno-1632c009-c843-4836-bb65-80c33084a64b.html',
 284         'info_dict': {
 285             'id': '1632c009-c843-4836-bb65-80c33084a64b',
 286             'ext': 'mp4',
 287             'title': 'Weekend al cinema, da Hollywood arriva il thriller di Tate Taylor "La ragazza del treno"',
 288             'description': 'I film in uscita questa settimana.',
 289             'thumbnail': r're:^https?://.*\.png$',
 290             'duration': 833,
 291             'upload_date': '20161103',
 292         }
 293     }, {
 294         # with ContentItem in og:url
 295         'url': 'http://www.rai.it/dl/RaiTV/programmi/media/ContentItem-efb17665-691c-45d5-a60c-5301333cbb0c.html',
 296         'md5': '6865dd00cf0bbf5772fdd89d59bd768a',
 297         'info_dict': {
 298             'id': 'efb17665-691c-45d5-a60c-5301333cbb0c',
 299             'ext': 'mp4',
 300             'title': 'TG1 ore 20:00 del 03/11/2016',
 301             'description': 'TG1 edizione integrale ore 20:00 del giorno 03/11/2016',
 302             'thumbnail': r're:^https?://.*\.jpg$',
 303             'duration': 2214,
 304             'upload_date': '20161103',
 305         }
 306     }, {
 307         # initEdizione('ContentItem-...'
 308         'url': 'http://www.tg1.rai.it/dl/tg1/2010/edizioni/ContentSet-9b6e0cba-4bef-4aef-8cf0-9f7f665b7dfb-tg1.html?item=undefined',
 309         'info_dict': {
 310             'id': 'c2187016-8484-4e3a-8ac8-35e475b07303',
 311             'ext': 'mp4',
 312             'title': r're:TG1 ore \d{2}:\d{2} del \d{2}/\d{2}/\d{4}',
 313             'duration': 2274,
 314             'upload_date': '20170401',
 315         },
 316         'skip': 'Changes daily',
 317     }, {
 318         # HLS live stream with ContentItem in og:url
 319         'url': 'http://www.rainews.it/dl/rainews/live/ContentItem-3156f2f2-dc70-4953-8e2f-70d7489d4ce9.html',
 320         'info_dict': {
 321             'id': '3156f2f2-dc70-4953-8e2f-70d7489d4ce9',
 322             'ext': 'mp4',
 323             'title': 'La diretta di Rainews24',
 324         },
 325         'params': {
 326             'skip_download': True,
 327         },
 328     }, {
 329         # Direct MMS URL
 330         'url': 'http://www.rai.it/dl/RaiTV/programmi/media/ContentItem-b63a4089-ac28-48cf-bca5-9f5b5bc46df5.html',
 331         'only_matching': True,
 332     }, {
 333         'url': 'https://www.rainews.it/tgr/marche/notiziari/video/2019/02/ContentItem-6ba945a2-889c-4a80-bdeb-8489c70a8db9.html',
 334         'only_matching': True,
 335     }]
 336
 337     def _extract_from_content_id(self, content_id, url):
 338         media = self._download_json(
 339             'http://www.rai.tv/dl/RaiTV/programmi/media/ContentItem-%s.html?json' % content_id,
 340             content_id, 'Downloading video JSON')
 341
 342         title = media['name'].strip()
 343
 344         media_type = media['type']
 345         if 'Audio' in media_type:
 346             relinker_info = {
 347                 'formats': [{
 348                     'format_id': media.get('formatoAudio'),
 349                     'url': media['audioUrl'],
 350                     'ext': media.get('formatoAudio'),
 351                 }]
 352             }
 353         elif 'Video' in media_type:
 354             relinker_info = self._extract_relinker_info(media['mediaUri'], content_id)
 355         else:
 356             raise ExtractorError('not a media file')
 357
 358         self._sort_formats(relinker_info['formats'])
 359
 360         thumbnails = []
 361         for image_type in ('image', 'image_medium', 'image_300'):
 362             thumbnail_url = media.get(image_type)
 363             if thumbnail_url:
 364                 thumbnails.append({
 365                     'url': compat_urlparse.urljoin(url, thumbnail_url),
 366                 })
 367
 368         subtitles = self._extract_subtitles(url, media.get('subtitlesUrl'))
 369
 370         info = {
 371             'id': content_id,
 372             'title': title,
 373             'description': strip_or_none(media.get('desc')),
 374             'thumbnails': thumbnails,
 375             'uploader': media.get('author'),
 376             'upload_date': unified_strdate(media.get('date')),
 377             'duration': parse_duration(media.get('length')),
 378             'subtitles': subtitles,
 379         }
 380
 381         info.update(relinker_info)
 382
 383         return info
 384
 385     def _real_extract(self, url):
 386         video_id = self._match_id(url)
 387
 388         webpage = self._download_webpage(url, video_id)
 389
 390         content_item_id = None
 391
 392         content_item_url = self._html_search_meta(
 393             ('og:url', 'og:video', 'og:video:secure_url', 'twitter:url',
 394              'twitter:player', 'jsonlink'), webpage, default=None)
 395         if content_item_url:
 396             content_item_id = self._search_regex(
 397                 r'ContentItem-(%s)' % self._UUID_RE, content_item_url,
 398                 'content item id', default=None)
 399
 400         if not content_item_id:
 401             content_item_id = self._search_regex(
 402                 r'''(?x)
 403                     (?:
 404                         (?:initEdizione|drawMediaRaiTV)\(|
 405                         <(?:[^>]+\bdata-id|var\s+uniquename)=
 406                     )
 407                     (["\'])
 408                     (?:(?!\1).)*\bContentItem-(?P<id>%s)
 409                 ''' % self._UUID_RE,
 410                 webpage, 'content item id', default=None, group='id')
 411
 412         content_item_ids = set()
 413         if content_item_id:
 414             content_item_ids.add(content_item_id)
 415         if video_id not in content_item_ids:
 416             content_item_ids.add(video_id)
 417
 418         for content_item_id in content_item_ids:
 419             try:
 420                 return self._extract_from_content_id(content_item_id, url)
 421             except GeoRestrictedError:
 422                 raise
 423             except ExtractorError:
 424                 pass
 425
 426         relinker_url = self._proto_relative_url(self._search_regex(
 427             r'''(?x)
 428                 (?:
 429                     var\s+videoURL|
 430                     mediaInfo\.mediaUri
 431                 )\s*=\s*
 432                 ([\'"])
 433                 (?P<url>
 434                     (?:https?:)?
 435                     //mediapolis(?:vod)?\.rai\.it/relinker/relinkerServlet\.htm\?
 436                     (?:(?!\1).)*\bcont=(?:(?!\1).)+)\1
 437             ''',
 438             webpage, 'relinker URL', group='url'))
 439
 440         relinker_info = self._extract_relinker_info(
 441             urljoin(url, relinker_url), video_id)
 442         self._sort_formats(relinker_info['formats'])
 443
 444         title = self._search_regex(
 445             r'var\s+videoTitolo\s*=\s*([\'"])(?P<title>[^\'"]+)\1',
 446             webpage, 'title', group='title',
 447             default=None) or self._og_search_title(webpage)
 448
 449         info = {
 450             'id': video_id,
 451             'title': title,
 452         }
 453
 454         info.update(relinker_info)
 455
 456         return info