yt_dlp/extractor/hrfensehen.py

   1 import json
   2 import re
   3
   4 from ..utils import int_or_none, unified_timestamp, unescapeHTML
   5 from .common import InfoExtractor
   6
   7
   8 class HRFernsehenIE(InfoExtractor):
   9     IE_NAME = 'hrfernsehen'
  10     _VALID_URL = r'^https?://www\.(?:hr-fernsehen|hessenschau)\.de/.*,video-(?P<id>[0-9]{6})\.html'
  11
  12     _TESTS = [{
  13         'url': 'https://www.hessenschau.de/tv-sendung/hessenschau-vom-26082020,video-130546.html',
  14         'md5': '5c4e0ba94677c516a2f65a84110fc536',
  15         'info_dict': {
  16             'id': '130546',
  17             'ext': 'mp4',
  18             'description': 'Sturmtief Kirsten fegt über Hessen / Die Corona-Pandemie – eine Chronologie / '
  19                            'Sterbehilfe: Die Lage in Hessen / Miss Hessen leitet zwei eigene Unternehmen / '
  20                            'Pop-Up Museum zeigt Schwarze Unterhaltung und Black Music',
  21             'subtitles': {'de': [{
  22                 'url': 'https://hr-a.akamaihd.net/video/as/hessenschau/2020_08/hrLogo_200826200407_L385592_512x288-25p-500kbit.vtt'
  23             }]},
  24             'timestamp': 1598470200,
  25             'upload_date': '20200826',
  26             'thumbnail': 'https://www.hessenschau.de/tv-sendung/hs_ganz-1554~_t-1598465545029_v-16to9__medium.jpg',
  27             'title': 'hessenschau vom 26.08.2020'
  28         }
  29     }, {
  30         'url': 'https://www.hr-fernsehen.de/sendungen-a-z/mex/sendungen/fair-und-gut---was-hinter-aldis-eigenem-guetesiegel-steckt,video-130544.html',
  31         'only_matching': True
  32     }]
  33
  34     _GEO_COUNTRIES = ['DE']
  35
  36     def extract_airdate(self, loader_data):
  37         airdate_str = loader_data.get('mediaMetadata', {}).get('agf', {}).get('airdate')
  38
  39         if airdate_str is None:
  40             return None
  41
  42         return unified_timestamp(airdate_str)
  43
  44     def extract_formats(self, loader_data):
  45         stream_formats = []
  46         for stream_obj in loader_data["videoResolutionLevels"]:
  47             stream_format = {
  48                 'format_id': str(stream_obj['verticalResolution']) + "p",
  49                 'height': stream_obj['verticalResolution'],
  50                 'url': stream_obj['url'],
  51             }
  52
  53             quality_information = re.search(r'([0-9]{3,4})x([0-9]{3,4})-([0-9]{2})p-([0-9]{3,4})kbit',
  54                                             stream_obj['url'])
  55             if quality_information:
  56                 stream_format['width'] = int_or_none(quality_information.group(1))
  57                 stream_format['height'] = int_or_none(quality_information.group(2))
  58                 stream_format['fps'] = int_or_none(quality_information.group(3))
  59                 stream_format['tbr'] = int_or_none(quality_information.group(4))
  60
  61             stream_formats.append(stream_format)
  62
  63         self._sort_formats(stream_formats)
  64         return stream_formats
  65
  66     def _real_extract(self, url):
  67         video_id = self._match_id(url)
  68         webpage = self._download_webpage(url, video_id)
  69
  70         title = self._html_search_meta(
  71             ['og:title', 'twitter:title', 'name'], webpage)
  72         description = self._html_search_meta(
  73             ['description'], webpage)
  74
  75         loader_str = unescapeHTML(self._search_regex(r"data-new-hr-mediaplayer-loader='([^']*)'", webpage, "ardloader"))
  76         loader_data = json.loads(loader_str)
  77
  78         info = {
  79             'id': video_id,
  80             'title': title,
  81             'description': description,
  82             'formats': self.extract_formats(loader_data),
  83             'timestamp': self.extract_airdate(loader_data)
  84         }
  85
  86         if "subtitle" in loader_data:
  87             info["subtitles"] = {"de": [{"url": loader_data["subtitle"]}]}
  88
  89         thumbnails = list(set([t for t in loader_data.get("previewImageUrl", {}).values()]))
  90         if len(thumbnails) > 0:
  91             info["thumbnails"] = [{"url": t} for t in thumbnails]
  92
  93         return info