yt_dlp/extractor/hrfensehen.py

   1 import json
   2 import re
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     int_or_none,
   7     traverse_obj,
   8     try_call,
   9     unescapeHTML,
  10     unified_timestamp,
  11 )
  12
  13
  14 class HRFernsehenIE(InfoExtractor):
  15     IE_NAME = 'hrfernsehen'
  16     _VALID_URL = r'https?://www\.(?:hr-fernsehen|hessenschau)\.de/.*,video-(?P<id>[0-9]{6})\.html'
  17     _TESTS = [{
  18         'url': 'https://www.hessenschau.de/tv-sendung/hessenschau-vom-26082020,video-130546.html',
  19         'md5': '5c4e0ba94677c516a2f65a84110fc536',
  20         'info_dict': {
  21             'id': '130546',
  22             'ext': 'mp4',
  23             'description': 'Sturmtief Kirsten fegt über Hessen / Die Corona-Pandemie – eine Chronologie / '
  24                            'Sterbehilfe: Die Lage in Hessen / Miss Hessen leitet zwei eigene Unternehmen / '
  25                            'Pop-Up Museum zeigt Schwarze Unterhaltung und Black Music',
  26             'subtitles': {'de': [{
  27                 'url': 'https://hr-a.akamaihd.net/video/as/hessenschau/2020_08/hrLogo_200826200407_L385592_512x288-25p-500kbit.vtt',
  28             }]},
  29             'timestamp': 1598400000,
  30             'upload_date': '20200826',
  31             'thumbnail': 'https://www.hessenschau.de/tv-sendung/hs_ganz-1554~_t-1598465545029_v-16to9.jpg',
  32             'title': 'hessenschau vom 26.08.2020',
  33             'duration': 1654,
  34         },
  35     }, {
  36         'url': 'https://www.hr-fernsehen.de/sendungen-a-z/mex/sendungen/fair-und-gut---was-hinter-aldis-eigenem-guetesiegel-steckt,video-130544.html',
  37         'only_matching': True,
  38     }]
  39
  40     _GEO_COUNTRIES = ['DE']
  41
  42     def extract_formats(self, loader_data):
  43         stream_formats = []
  44         data = loader_data['mediaCollection']['streams'][0]['media']
  45         for inner in data[1:]:
  46             stream_format = {
  47                 'format_id': try_call(lambda: f'{inner["maxHResolutionPx"]}p'),
  48                 'height': inner.get('maxHResolutionPx'),
  49                 'url': inner['url'],
  50             }
  51
  52             quality_information = re.search(r'([0-9]{3,4})x([0-9]{3,4})-([0-9]{2})p-([0-9]{3,4})kbit',
  53                                             inner['url'])
  54             if quality_information:
  55                 stream_format['width'] = int_or_none(quality_information.group(1))
  56                 stream_format['height'] = int_or_none(quality_information.group(2))
  57                 stream_format['fps'] = int_or_none(quality_information.group(3))
  58                 stream_format['tbr'] = int_or_none(quality_information.group(4))
  59
  60             stream_formats.append(stream_format)
  61         return stream_formats
  62
  63     def _real_extract(self, url):
  64         video_id = self._match_id(url)
  65         webpage = self._download_webpage(url, video_id)
  66
  67         title = self._html_search_meta(
  68             ['og:title', 'twitter:title', 'name'], webpage)
  69         description = self._html_search_meta(
  70             ['description'], webpage)
  71
  72         loader_str = unescapeHTML(self._search_regex(r"data-(?:new-)?hr-mediaplayer-loader='([^']*)'", webpage, 'ardloader'))
  73         loader_data = json.loads(loader_str)
  74
  75         subtitle = traverse_obj(loader_data, ('mediaCollection', 'subTitles', 0, 'sources', 0, 'url'))
  76
  77         return {
  78             'id': video_id,
  79             'title': title,
  80             'description': description,
  81             'formats': self.extract_formats(loader_data),
  82             'subtitles': {'de': [{'url': subtitle}]},
  83             'timestamp': unified_timestamp(self._search_regex(
  84                 r'<time\sdatetime="(\d{4}\W\d{1,2}\W\d{1,2})', webpage, 'datetime', fatal=False)),
  85             'duration': int_or_none(traverse_obj(
  86                 loader_data, ('playerConfig', 'pluginData', 'trackingAti@all', 'richMedia', 'duration'))),
  87             'thumbnail': self._search_regex(r'thumbnailUrl\W*([^"]+)', webpage, 'thumbnail', default=None),
  88         }