youtube_dl/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
 158         video = media['video']
 159
 160         relinker_info = self._extract_relinker_info(video['content_url'], video_id)
 161         self._sort_formats(relinker_info['formats'])
 162
 163         thumbnails = []
 164         for _, value in media.get('images', {}).items():
 165             if value:
 166                 thumbnails.append({
 167                     'url': urljoin(url, value),
 168                 })
 169
 170         date_published = media.get('date_published')
 171         time_published = media.get('time_published')
 172         if date_published and time_published:
 173             date_published += ' ' + time_published
 174
 175         subtitles = self._extract_subtitles(url, video.get('subtitles'))
 176
 177         program_info = media.get('program_info') or {}
 178         season = media.get('season')
 179
 180         info = {
 181             'id': remove_start(media.get('id'), 'ContentItem-') or video_id,
 182             'display_id': video_id,
 183             'title': self._live_title(title) if relinker_info.get(
 184                 'is_live') else title,
 185             'alt_title': strip_or_none(media.get('subtitle')),
 186             'description': media.get('description'),
 187             'uploader': strip_or_none(media.get('channel')),
 188             'creator': strip_or_none(media.get('editor') or None),
 189             'duration': parse_duration(video.get('duration')),
 190             'timestamp': unified_timestamp(date_published),
 191             'thumbnails': thumbnails,
 192             'series': program_info.get('name'),
 193             'season_number': int_or_none(season),
 194             'season': season if (season and not season.isdigit()) else None,
 195             'episode': media.get('episode_title'),
 196             'episode_number': int_or_none(media.get('episode')),
 197             'subtitles': subtitles,
 198         }
 199
 200         info.update(relinker_info)
 201         return info
 202
 203
 204 class RaiPlayLiveIE(RaiPlayIE):
 205     _VALID_URL = r'(?P<base>https?://(?:www\.)?raiplay\.it/dirette/(?P<id>[^/?#&]+))'
 206     _TESTS = [{
 207         'url': 'http://www.raiplay.it/dirette/rainews24',
 208         'info_dict': {
 209             'id': 'd784ad40-e0ae-4a69-aa76-37519d238a9c',
 210             'display_id': 'rainews24',
 211             'ext': 'mp4',
 212             'title': 're:^Diretta di Rai News 24 [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
 213             'description': 'md5:4d00bcf6dc98b27c6ec480de329d1497',
 214             'uploader': 'Rai News 24',
 215             'creator': 'Rai News 24',
 216             'is_live': True,
 217         },
 218         'params': {
 219             'skip_download': True,
 220         },
 221     }]
 222
 223
 224 class RaiPlayPlaylistIE(InfoExtractor):
 225     _VALID_URL = r'(?P<base>https?://(?:www\.)?raiplay\.it/programmi/(?P<id>[^/?#&]+))'
 226     _TESTS = [{
 227         'url': 'http://www.raiplay.it/programmi/nondirloalmiocapo/',
 228         'info_dict': {
 229             'id': 'nondirloalmiocapo',
 230             'title': 'Non dirlo al mio capo',
 231             'description': 'md5:98ab6b98f7f44c2843fd7d6f045f153b',
 232         },
 233         'playlist_mincount': 12,
 234     }]
 235
 236     def _real_extract(self, url):
 237         base, playlist_id = re.match(self._VALID_URL, url).groups()
 238
 239         program = self._download_json(
 240             base + '.json', playlist_id, 'Downloading program JSON')
 241
 242         entries = []
 243         for b in (program.get('blocks') or []):
 244             for s in (b.get('sets') or []):
 245                 s_id = s.get('id')
 246                 if not s_id:
 247                     continue
 248                 medias = self._download_json(
 249                     '%s/%s.json' % (base, s_id), s_id,
 250                     'Downloading content set JSON', fatal=False)
 251                 if not medias:
 252                     continue
 253                 for m in (medias.get('items') or []):
 254                     path_id = m.get('path_id')
 255                     if not path_id:
 256                         continue
 257                     video_url = urljoin(url, path_id)
 258                     entries.append(self.url_result(
 259                         video_url, ie=RaiPlayIE.ie_key(),
 260                         video_id=RaiPlayIE._match_id(video_url)))
 261
 262         return self.playlist_result(
 263             entries, playlist_id, program.get('name'),
 264             try_get(program, lambda x: x['program_info']['description']))
 265
 266
 267 class RaiIE(RaiBaseIE):
 268     _VALID_URL = r'https?://[^/]+\.(?:rai\.(?:it|tv)|rainews\.it)/.+?-(?P<id>%s)(?:-.+?)?\.html' % RaiBaseIE._UUID_RE
 269     _TESTS = [{
 270         # var uniquename = "ContentItem-..."
 271         # data-id="ContentItem-..."
 272         'url': 'http://www.raisport.rai.it/dl/raiSport/media/rassegna-stampa-04a9f4bd-b563-40cf-82a6-aad3529cb4a9.html',
 273         'info_dict': {
 274             'id': '04a9f4bd-b563-40cf-82a6-aad3529cb4a9',
 275             'ext': 'mp4',
 276             'title': 'TG PRIMO TEMPO',
 277             'thumbnail': r're:^https?://.*\.jpg$',
 278             'duration': 1758,
 279             'upload_date': '20140612',
 280         },
 281         'skip': 'This content is available only in Italy',
 282     }, {
 283         # with ContentItem in many metas
 284         'url': 'http://www.rainews.it/dl/rainews/media/Weekend-al-cinema-da-Hollywood-arriva-il-thriller-di-Tate-Taylor-La-ragazza-del-treno-1632c009-c843-4836-bb65-80c33084a64b.html',
 285         'info_dict': {
 286             'id': '1632c009-c843-4836-bb65-80c33084a64b',
 287             'ext': 'mp4',
 288             'title': 'Weekend al cinema, da Hollywood arriva il thriller di Tate Taylor "La ragazza del treno"',
 289             'description': 'I film in uscita questa settimana.',
 290             'thumbnail': r're:^https?://.*\.png$',
 291             'duration': 833,
 292             'upload_date': '20161103',
 293         }
 294     }, {
 295         # with ContentItem in og:url
 296         'url': 'http://www.rai.it/dl/RaiTV/programmi/media/ContentItem-efb17665-691c-45d5-a60c-5301333cbb0c.html',
 297         'md5': '6865dd00cf0bbf5772fdd89d59bd768a',
 298         'info_dict': {
 299             'id': 'efb17665-691c-45d5-a60c-5301333cbb0c',
 300             'ext': 'mp4',
 301             'title': 'TG1 ore 20:00 del 03/11/2016',
 302             'description': 'TG1 edizione integrale ore 20:00 del giorno 03/11/2016',
 303             'thumbnail': r're:^https?://.*\.jpg$',
 304             'duration': 2214,
 305             'upload_date': '20161103',
 306         }
 307     }, {
 308         # initEdizione('ContentItem-...'
 309         'url': 'http://www.tg1.rai.it/dl/tg1/2010/edizioni/ContentSet-9b6e0cba-4bef-4aef-8cf0-9f7f665b7dfb-tg1.html?item=undefined',
 310         'info_dict': {
 311             'id': 'c2187016-8484-4e3a-8ac8-35e475b07303',
 312             'ext': 'mp4',
 313             'title': r're:TG1 ore \d{2}:\d{2} del \d{2}/\d{2}/\d{4}',
 314             'duration': 2274,
 315             'upload_date': '20170401',
 316         },
 317         'skip': 'Changes daily',
 318     }, {
 319         # HLS live stream with ContentItem in og:url
 320         'url': 'http://www.rainews.it/dl/rainews/live/ContentItem-3156f2f2-dc70-4953-8e2f-70d7489d4ce9.html',
 321         'info_dict': {
 322             'id': '3156f2f2-dc70-4953-8e2f-70d7489d4ce9',
 323             'ext': 'mp4',
 324             'title': 'La diretta di Rainews24',
 325         },
 326         'params': {
 327             'skip_download': True,
 328         },
 329     }, {
 330         # Direct MMS URL
 331         'url': 'http://www.rai.it/dl/RaiTV/programmi/media/ContentItem-b63a4089-ac28-48cf-bca5-9f5b5bc46df5.html',
 332         'only_matching': True,
 333     }, {
 334         'url': 'https://www.rainews.it/tgr/marche/notiziari/video/2019/02/ContentItem-6ba945a2-889c-4a80-bdeb-8489c70a8db9.html',
 335         'only_matching': True,
 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._proto_relative_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