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

   1 from __future__ import unicode_literals
   2
   3 from .common import InfoExtractor
   4 from ..utils import (
   5     ExtractorError,
   6     int_or_none,
   7     float_or_none,
   8 )
   9
  10
  11 class WistiaIE(InfoExtractor):
  12     _VALID_URL = r'(?:wistia:|https?://(?:fast\.)?wistia\.net/embed/iframe/)(?P<id>[a-z0-9]+)'
  13     _API_URL = 'http://fast.wistia.com/embed/medias/%s.json'
  14     _IFRAME_URL = 'http://fast.wistia.net/embed/iframe/%s'
  15
  16     _TESTS = [{
  17         'url': 'http://fast.wistia.net/embed/iframe/sh7fpupwlt',
  18         'md5': 'cafeb56ec0c53c18c97405eecb3133df',
  19         'info_dict': {
  20             'id': 'sh7fpupwlt',
  21             'ext': 'mov',
  22             'title': 'Being Resourceful',
  23             'description': 'a Clients From Hell Video Series video from worldwidewebhosting',
  24             'upload_date': '20131204',
  25             'timestamp': 1386185018,
  26             'duration': 117,
  27         },
  28     }, {
  29         'url': 'wistia:sh7fpupwlt',
  30         'only_matching': True,
  31     }, {
  32         # with hls video
  33         'url': 'wistia:807fafadvk',
  34         'only_matching': True,
  35     }]
  36
  37     def _real_extract(self, url):
  38         video_id = self._match_id(url)
  39
  40         data_json = self._download_json(
  41             self._API_URL % video_id, video_id,
  42             # Some videos require this.
  43             headers={
  44                 'Referer': url if url.startswith('http') else self._IFRAME_URL % video_id,
  45             })
  46
  47         if data_json.get('error'):
  48             raise ExtractorError(
  49                 'Error while getting the playlist', expected=True)
  50
  51         data = data_json['media']
  52         title = data['name']
  53
  54         formats = []
  55         thumbnails = []
  56         for a in data['assets']:
  57             aurl = a.get('url')
  58             if not aurl:
  59                 continue
  60             astatus = a.get('status')
  61             atype = a.get('type')
  62             if (astatus is not None and astatus != 2) or atype in ('preview', 'storyboard'):
  63                 continue
  64             elif atype in ('still', 'still_image'):
  65                 thumbnails.append({
  66                     'url': aurl,
  67                     'width': int_or_none(a.get('width')),
  68                     'height': int_or_none(a.get('height')),
  69                 })
  70             else:
  71                 aext = a.get('ext')
  72                 is_m3u8 = a.get('container') == 'm3u8' or aext == 'm3u8'
  73                 formats.append({
  74                     'format_id': atype,
  75                     'url': aurl,
  76                     'tbr': int_or_none(a.get('bitrate')),
  77                     'vbr': int_or_none(a.get('opt_vbitrate')),
  78                     'width': int_or_none(a.get('width')),
  79                     'height': int_or_none(a.get('height')),
  80                     'filesize': int_or_none(a.get('size')),
  81                     'vcodec': a.get('codec'),
  82                     'container': a.get('container'),
  83                     'ext': 'mp4' if is_m3u8 else aext,
  84                     'protocol': 'm3u8' if is_m3u8 else None,
  85                     'preference': 1 if atype == 'original' else None,
  86                 })
  87
  88         self._sort_formats(formats)
  89
  90         return {
  91             'id': video_id,
  92             'title': title,
  93             'description': data.get('seoDescription'),
  94             'formats': formats,
  95             'thumbnails': thumbnails,
  96             'duration': float_or_none(data.get('duration')),
  97             'timestamp': int_or_none(data.get('createdAt')),
  98         }