yt_dlp/extractor/fusion.py

   1 from .common import InfoExtractor
   2 from ..utils import (
   3     determine_ext,
   4     int_or_none,
   5     mimetype2ext,
   6     parse_iso8601,
   7 )
   8
   9
  10 class FusionIE(InfoExtractor):
  11     _VALID_URL = r'https?://(?:www\.)?fusion\.(?:net|tv)/(?:video/|show/.+?\bvideo=)(?P<id>\d+)'
  12     _TESTS = [{
  13         'url': 'http://fusion.tv/video/201781/u-s-and-panamanian-forces-work-together-to-stop-a-vessel-smuggling-drugs/',
  14         'info_dict': {
  15             'id': '3145868',
  16             'ext': 'mp4',
  17             'title': 'U.S. and Panamanian forces work together to stop a vessel smuggling drugs',
  18             'description': 'md5:0cc84a9943c064c0f46b128b41b1b0d7',
  19             'duration': 140.0,
  20             'timestamp': 1442589635,
  21             'uploader': 'UNIVISON',
  22             'upload_date': '20150918',
  23         },
  24         'params': {
  25             'skip_download': True,
  26         },
  27         'add_ie': ['Anvato'],
  28     }, {
  29         'url': 'http://fusion.tv/video/201781',
  30         'only_matching': True,
  31     }, {
  32         'url': 'https://fusion.tv/show/food-exposed-with-nelufar-hedayat/?ancla=full-episodes&video=588644',
  33         'only_matching': True,
  34     }]
  35
  36     def _real_extract(self, url):
  37         video_id = self._match_id(url)
  38         video = self._download_json(
  39             'https://platform.fusion.net/wp-json/fusiondotnet/v1/video/' + video_id, video_id)
  40
  41         info = {
  42             'id': video_id,
  43             'title': video['title'],
  44             'description': video.get('excerpt'),
  45             'timestamp': parse_iso8601(video.get('published')),
  46             'series': video.get('show'),
  47         }
  48
  49         formats = []
  50         src = video.get('src') or {}
  51         for f_id, f in src.items():
  52             for q_id, q in f.items():
  53                 q_url = q.get('url')
  54                 if not q_url:
  55                     continue
  56                 ext = determine_ext(q_url, mimetype2ext(q.get('type')))
  57                 if ext == 'smil':
  58                     formats.extend(self._extract_smil_formats(q_url, video_id, fatal=False))
  59                 elif f_id == 'm3u8-variant' or (ext == 'm3u8' and q_id == 'Variant'):
  60                     formats.extend(self._extract_m3u8_formats(
  61                         q_url, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False))
  62                 else:
  63                     formats.append({
  64                         'format_id': '-'.join([f_id, q_id]),
  65                         'url': q_url,
  66                         'width': int_or_none(q.get('width')),
  67                         'height': int_or_none(q.get('height')),
  68                         'tbr': int_or_none(self._search_regex(r'_(\d+)\.m(?:p4|3u8)', q_url, 'bitrate')),
  69                         'ext': 'mp4' if ext == 'm3u8' else ext,
  70                         'protocol': 'm3u8_native' if ext == 'm3u8' else 'https',
  71                     })
  72         if formats:
  73             info['formats'] = formats
  74         else:
  75             info.update({
  76                 '_type': 'url',
  77                 'url': 'anvato:uni:' + video['video_ids']['anvato'],
  78                 'ie_key': 'Anvato',
  79             })
  80
  81         return info