yt_dlp/extractor/lrt.py

   1 from .common import InfoExtractor
   2 from ..utils import (
   3     clean_html,
   4     merge_dicts,
   5     traverse_obj,
   6     url_or_none,
   7 )
   8
   9
  10 class LRTBaseIE(InfoExtractor):
  11     def _extract_js_var(self, webpage, var_name, default=None):
  12         return self._search_regex(
  13             fr'{var_name}\s*=\s*(["\'])((?:(?!\1).)+)\1',
  14             webpage, var_name.replace('_', ' '), default, group=2)
  15
  16
  17 class LRTStreamIE(LRTBaseIE):
  18     _VALID_URL = r'https?://(?:www\.)?lrt\.lt/mediateka/tiesiogiai/(?P<id>[\w-]+)'
  19     _TESTS = [{
  20         'url': 'https://www.lrt.lt/mediateka/tiesiogiai/lrt-opus',
  21         'info_dict': {
  22             'id': 'lrt-opus',
  23             'live_status': 'is_live',
  24             'title': 're:^LRT Opus.+$',
  25             'ext': 'mp4'
  26         }
  27     }]
  28
  29     def _real_extract(self, url):
  30         video_id = self._match_id(url)
  31         webpage = self._download_webpage(url, video_id)
  32         streams_data = self._download_json(self._extract_js_var(webpage, 'tokenURL'), video_id)
  33
  34         formats, subtitles = [], {}
  35         for stream_url in traverse_obj(streams_data, (
  36                 'response', 'data', lambda k, _: k.startswith('content')), expected_type=url_or_none):
  37             fmts, subs = self._extract_m3u8_formats_and_subtitles(stream_url, video_id, 'mp4', m3u8_id='hls', live=True)
  38             formats.extend(fmts)
  39             subtitles = self._merge_subtitles(subtitles, subs)
  40
  41         stream_title = self._extract_js_var(webpage, 'video_title', 'LRT')
  42         return {
  43             'id': video_id,
  44             'formats': formats,
  45             'subtitles': subtitles,
  46             'is_live': True,
  47             'title': f'{self._og_search_title(webpage)} - {stream_title}'
  48         }
  49
  50
  51 class LRTVODIE(LRTBaseIE):
  52     _VALID_URL = r'https?://(?:www\.)?lrt\.lt(?P<path>/mediateka/irasas/(?P<id>[0-9]+))'
  53     _TESTS = [{
  54         # m3u8 download
  55         'url': 'https://www.lrt.lt/mediateka/irasas/2000127261/greita-ir-gardu-sicilijos-ikvepta-klasikiniu-makaronu-su-baklazanais-vakariene',
  56         'info_dict': {
  57             'id': '2000127261',
  58             'ext': 'mp4',
  59             'title': 'Greita ir gardu: Sicilijos įkvėpta klasikinių makaronų su baklažanais vakarienė',
  60             'description': 'md5:ad7d985f51b0dc1489ba2d76d7ed47fa',
  61             'duration': 3035,
  62             'timestamp': 1604079000,
  63             'upload_date': '20201030',
  64             'tags': ['LRT TELEVIZIJA', 'Beatos virtuvė', 'Beata Nicholson', 'Makaronai', 'Baklažanai', 'Vakarienė', 'Receptas'],
  65             'thumbnail': 'https://www.lrt.lt/img/2020/10/30/764041-126478-1287x836.jpg'
  66         },
  67     }, {
  68         # direct mp3 download
  69         'url': 'http://www.lrt.lt/mediateka/irasas/1013074524/',
  70         'md5': '389da8ca3cad0f51d12bed0c844f6a0a',
  71         'info_dict': {
  72             'id': '1013074524',
  73             'ext': 'mp3',
  74             'title': 'Kita tema 2016-09-05 15:05',
  75             'description': 'md5:1b295a8fc7219ed0d543fc228c931fb5',
  76             'duration': 3008,
  77             'view_count': int,
  78             'like_count': int,
  79         },
  80     }]
  81
  82     def _real_extract(self, url):
  83         path, video_id = self._match_valid_url(url).groups()
  84         webpage = self._download_webpage(url, video_id)
  85
  86         media_url = self._extract_js_var(webpage, 'main_url', path)
  87         media = self._download_json(self._extract_js_var(
  88             webpage, 'media_info_url',
  89             'https://www.lrt.lt/servisai/stream_url/vod/media_info/'),
  90             video_id, query={'url': media_url})
  91         jw_data = self._parse_jwplayer_data(
  92             media['playlist_item'], video_id, base_url=url)
  93
  94         json_ld_data = self._search_json_ld(webpage, video_id)
  95
  96         tags = []
  97         for tag in (media.get('tags') or []):
  98             tag_name = tag.get('name')
  99             if not tag_name:
 100                 continue
 101             tags.append(tag_name)
 102
 103         clean_info = {
 104             'description': clean_html(media.get('content')),
 105             'tags': tags,
 106         }
 107
 108         return merge_dicts(clean_info, jw_data, json_ld_data)