Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/fusion.py

   1 from __future__ import unicode_literals
   2
   3 from .common import InfoExtractor
   4 from ..utils import (
   5     determine_ext,
   6     int_or_none,
   7     mimetype2ext,
   8     parse_iso8601,
   9 )
  10
  11
  12 class FusionIE(InfoExtractor):
  13     _VALID_URL = r'https?://(?:www\.)?fusion\.(?:net|tv)/(?:video/|show/.+?\bvideo=)(?P<id>\d+)'
  14     _TESTS = [{
  15         'url': 'http://fusion.tv/video/201781/u-s-and-panamanian-forces-work-together-to-stop-a-vessel-smuggling-drugs/',
  16         'info_dict': {
  17             'id': '3145868',
  18             'ext': 'mp4',
  19             'title': 'U.S. and Panamanian forces work together to stop a vessel smuggling drugs',
  20             'description': 'md5:0cc84a9943c064c0f46b128b41b1b0d7',
  21             'duration': 140.0,
  22             'timestamp': 1442589635,
  23             'uploader': 'UNIVISON',
  24             'upload_date': '20150918',
  25         },
  26         'params': {
  27             'skip_download': True,
  28         },
  29         'add_ie': ['Anvato'],
  30     }, {
  31         'url': 'http://fusion.tv/video/201781',
  32         'only_matching': True,
  33     }, {
  34         'url': 'https://fusion.tv/show/food-exposed-with-nelufar-hedayat/?ancla=full-episodes&video=588644',
  35         'only_matching': True,
  36     }]
  37
  38     def _real_extract(self, url):
  39         video_id = self._match_id(url)
  40         video = self._download_json(
  41             'https://platform.fusion.net/wp-json/fusiondotnet/v1/video/' + video_id, video_id)
  42
  43         info = {
  44             'id': video_id,
  45             'title': video['title'],
  46             'description': video.get('excerpt'),
  47             'timestamp': parse_iso8601(video.get('published')),
  48             'series': video.get('show'),
  49         }
  50
  51         formats = []
  52         src = video.get('src') or {}
  53         for f_id, f in src.items():
  54             for q_id, q in f.items():
  55                 q_url = q.get('url')
  56                 if not q_url:
  57                     continue
  58                 ext = determine_ext(q_url, mimetype2ext(q.get('type')))
  59                 if ext == 'smil':
  60                     formats.extend(self._extract_smil_formats(q_url, video_id, fatal=False))
  61                 elif f_id == 'm3u8-variant' or (ext == 'm3u8' and q_id == 'Variant'):
  62                     formats.extend(self._extract_m3u8_formats(
  63                         q_url, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False))
  64                 else:
  65                     formats.append({
  66                         'format_id': '-'.join([f_id, q_id]),
  67                         'url': q_url,
  68                         'width': int_or_none(q.get('width')),
  69                         'height': int_or_none(q.get('height')),
  70                         'tbr': int_or_none(self._search_regex(r'_(\d+)\.m(?:p4|3u8)', q_url, 'bitrate')),
  71                         'ext': 'mp4' if ext == 'm3u8' else ext,
  72                         'protocol': 'm3u8_native' if ext == 'm3u8' else 'https',
  73                     })
  74         if formats:
  75             self._sort_formats(formats)
  76             info['formats'] = formats
  77         else:
  78             info.update({
  79                 '_type': 'url',
  80                 'url': 'anvato:uni:' + video['video_ids']['anvato'],
  81                 'ie_key': 'Anvato',
  82             })
  83
  84         return info