yt_dlp/extractor/mediaworksnz.py

   1 import re
   2
   3 from .common import InfoExtractor
   4 from ..utils import (
   5     bug_reports_message,
   6     float_or_none,
   7     traverse_obj,
   8     unified_timestamp,
   9 )
  10
  11
  12 class MediaWorksNZVODIE(InfoExtractor):
  13     _VALID_URL_BASE_RE = r'https?://vodupload-api\.mediaworks\.nz/library/asset/published/'
  14     _VALID_URL_ID_RE = r'(?P<id>[A-Za-z0-9-]+)'
  15     _VALID_URL = rf'{_VALID_URL_BASE_RE}{_VALID_URL_ID_RE}'
  16     _TESTS = [{
  17         'url': 'https://vodupload-api.mediaworks.nz/library/asset/published/VID00359',
  18         'info_dict': {
  19             'id': 'VID00359',
  20             'ext': 'mp4',
  21             'title': 'GRG Jacinda Ardern safe drug testing 1920x1080',
  22             'description': 'md5:d4d7dc366742e86d8130b257dcb520ba',
  23             'duration': 142.76,
  24             'timestamp': 1604268608,
  25             'upload_date': '20201101',
  26             'thumbnail': r're:^https?://.*\.jpg$',
  27             'channel': 'George FM'
  28         }
  29     }, {
  30         # has audio-only format
  31         'url': 'https://vodupload-api.mediaworks.nz/library/asset/published/VID02627',
  32         'info_dict': {
  33             'id': 'VID02627',
  34             'ext': 'mp3',
  35             'title': 'Tova O\'Brien meets Ukraine President Volodymyr Zelensky',
  36             'channel': 'Today FM',
  37             'description': 'Watch in full the much anticipated interview of Volodymyr Zelensky',
  38             'duration': 2061.16,
  39             'thumbnail': r're:^https?://.*\.jpg$',
  40             'upload_date': '20220822',
  41             'timestamp': 1661152289,
  42         },
  43         'params': {'format': 'ba[ext=mp3]'}
  44     }]
  45
  46     _WEBPAGE_TESTS = [{
  47         'url': 'https://www.rova.nz/home/podcasts/socrates-walks-into-a-bar/the-trolley-problem---episode-1.html',
  48         'info_dict': {
  49             'id': 'VID02494',
  50             'ext': 'mp4',
  51             'title': 'The Trolley Problem',
  52             'duration': 2843.56,
  53             'channel': 'Other',
  54             'timestamp': 1658356489,
  55             'thumbnail': r're:^https?://.*\.jpg$',
  56             'description': 'Socrates Walks Into A Bar Podcast Episode 1',
  57             'upload_date': '20220720',
  58         }
  59     }]
  60
  61     @classmethod
  62     def _extract_embed_urls(cls, url, webpage):
  63         for mobj in re.finditer(
  64             rf'''(?x)<div\s+\bid=["']Player-Attributes-JWID[^>]+\b
  65             data-request-url=["']{cls._VALID_URL_BASE_RE}["'][^>]+\b
  66             data-asset-id=["']{cls._VALID_URL_ID_RE}["']''', webpage
  67         ):
  68             yield f'https://vodupload-api.mediaworks.nz/library/asset/published/{mobj.group("id")}'
  69
  70     def _real_extract(self, url):
  71         video_id = self._match_id(url)
  72         asset = self._download_json(url, video_id)['asset']
  73
  74         if asset.get('drm') not in ('NonDRM', None):
  75             self.report_drm(video_id)
  76
  77         content_type = asset.get('type')
  78         if content_type and content_type != 'video':
  79             self.report_warning(f'Unknown content type: {content_type}' + bug_reports_message(), video_id)
  80
  81         formats, subtitles = self._extract_m3u8_formats_and_subtitles(asset['streamingUrl'], video_id)
  82
  83         audio_streaming_url = traverse_obj(
  84             asset, 'palyoutPathAudio', 'playoutpathaudio', expected_type=str)
  85         if audio_streaming_url:
  86             audio_formats = self._extract_m3u8_formats(audio_streaming_url, video_id, fatal=False, ext='mp3')
  87             for audio_format in audio_formats:
  88                 # all the audio streams appear to be aac
  89                 audio_format.setdefault('vcodec', 'none')
  90                 audio_format.setdefault('acodec', 'aac')
  91                 formats.append(audio_format)
  92
  93         return {
  94             'id': video_id,
  95             'title': asset.get('title'),
  96             'description': asset.get('description'),
  97             'duration': float_or_none(asset.get('duration')),
  98             'timestamp': unified_timestamp(asset.get('dateadded')),
  99             'channel': asset.get('brand'),
 100             'thumbnails': [{'url': thumbnail_url} for thumbnail_url in asset.get('thumbnails') or []],
 101             'formats': formats,
 102             'subtitles': subtitles,
 103         }