yt_dlp/extractor/medialaan.py

   1 import re
   2
   3 from .common import InfoExtractor
   4 from ..utils import (
   5     extract_attributes,
   6     int_or_none,
   7     mimetype2ext,
   8     parse_iso8601,
   9 )
  10
  11
  12 class MedialaanIE(InfoExtractor):
  13     _VALID_URL = r'''(?x)
  14                     https?://
  15                         (?:
  16                             (?:embed\.)?mychannels.video/embed/|
  17                             embed\.mychannels\.video/(?:s(?:dk|cript)/)?production/|
  18                             (?:www\.)?(?:
  19                                 (?:
  20                                     7sur7|
  21                                     demorgen|
  22                                     hln|
  23                                     joe|
  24                                     qmusic
  25                                 )\.be|
  26                                 (?:
  27                                     [abe]d|
  28                                     bndestem|
  29                                     destentor|
  30                                     gelderlander|
  31                                     pzc|
  32                                     tubantia|
  33                                     volkskrant
  34                                 )\.nl
  35                             )/video/(?:[^/]+/)*[^/?&#]+~p
  36                         )
  37                         (?P<id>\d+)
  38                     '''
  39     _TESTS = [{
  40         'url': 'https://www.bndestem.nl/video/de-terugkeer-van-ally-de-aap-en-wie-vertrekt-er-nog-bij-nac~p193993',
  41         'info_dict': {
  42             'id': '193993',
  43             'ext': 'mp4',
  44             'title': 'De terugkeer van Ally de Aap en wie vertrekt er nog bij NAC?',
  45             'timestamp': 1611663540,
  46             'upload_date': '20210126',
  47             'duration': 238,
  48         },
  49         'params': {
  50             'skip_download': True,
  51         },
  52     }, {
  53         'url': 'https://www.gelderlander.nl/video/kanalen/degelderlander~c320/series/snel-nieuws~s984/noodbevel-in-doetinchem-politie-stuurt-mensen-centrum-uit~p194093',
  54         'only_matching': True,
  55     }, {
  56         'url': 'https://embed.mychannels.video/sdk/production/193993?options=TFTFF_default',
  57         'only_matching': True,
  58     }, {
  59         'url': 'https://embed.mychannels.video/script/production/193993',
  60         'only_matching': True,
  61     }, {
  62         'url': 'https://embed.mychannels.video/production/193993',
  63         'only_matching': True,
  64     }, {
  65         'url': 'https://mychannels.video/embed/193993',
  66         'only_matching': True,
  67     }, {
  68         'url': 'https://embed.mychannels.video/embed/193993',
  69         'only_matching': True,
  70     }]
  71
  72     @classmethod
  73     def _extract_embed_urls(cls, url, webpage):
  74         entries = []
  75         for element in re.findall(r'(<div[^>]+data-mychannels-type="video"[^>]*>)', webpage):
  76             mychannels_id = extract_attributes(element).get('data-mychannels-id')
  77             if mychannels_id:
  78                 entries.append('https://mychannels.video/embed/' + mychannels_id)
  79         return entries
  80
  81     def _real_extract(self, url):
  82         production_id = self._match_id(url)
  83         production = self._download_json(
  84             'https://embed.mychannels.video/sdk/production/' + production_id,
  85             production_id, query={'options': 'UUUU_default'})['productions'][0]
  86         title = production['title']
  87
  88         formats = []
  89         for source in (production.get('sources') or []):
  90             src = source.get('src')
  91             if not src:
  92                 continue
  93             ext = mimetype2ext(source.get('type'))
  94             if ext == 'm3u8':
  95                 formats.extend(self._extract_m3u8_formats(
  96                     src, production_id, 'mp4', 'm3u8_native',
  97                     m3u8_id='hls', fatal=False))
  98             else:
  99                 formats.append({
 100                     'ext': ext,
 101                     'url': src,
 102                 })
 103
 104         return {
 105             'id': production_id,
 106             'title': title,
 107             'formats': formats,
 108             'thumbnail': production.get('posterUrl'),
 109             'timestamp': parse_iso8601(production.get('publicationDate'), ' '),
 110             'duration': int_or_none(production.get('duration')) or None,
 111         }