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

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..utils import (
   7     ExtractorError,
   8     int_or_none,
   9     float_or_none,
  10     unescapeHTML,
  11 )
  12
  13
  14 class WistiaIE(InfoExtractor):
  15     _VALID_URL = r'(?:wistia:|https?://(?:fast\.)?wistia\.net/embed/iframe/)(?P<id>[a-z0-9]+)'
  16     _API_URL = 'http://fast.wistia.com/embed/medias/%s.json'
  17     _IFRAME_URL = 'http://fast.wistia.net/embed/iframe/%s'
  18
  19     _TESTS = [{
  20         'url': 'http://fast.wistia.net/embed/iframe/sh7fpupwlt',
  21         'md5': 'cafeb56ec0c53c18c97405eecb3133df',
  22         'info_dict': {
  23             'id': 'sh7fpupwlt',
  24             'ext': 'mov',
  25             'title': 'Being Resourceful',
  26             'description': 'a Clients From Hell Video Series video from worldwidewebhosting',
  27             'upload_date': '20131204',
  28             'timestamp': 1386185018,
  29             'duration': 117,
  30         },
  31     }, {
  32         'url': 'wistia:sh7fpupwlt',
  33         'only_matching': True,
  34     }, {
  35         # with hls video
  36         'url': 'wistia:807fafadvk',
  37         'only_matching': True,
  38     }]
  39
  40     @staticmethod
  41     def _extract_url(webpage):
  42         match = re.search(
  43             r'<(?:meta[^>]+?content|iframe[^>]+?src)=(["\'])(?P<url>(?:https?:)?//(?:fast\.)?wistia\.net/embed/iframe/.+?)\1', webpage)
  44         if match:
  45             return unescapeHTML(match.group('url'))
  46
  47         match = re.search(r'(?:id=["\']wistia_|data-wistia-?id=["\']|Wistia\.embed\(["\'])(?P<id>[^"\']+)', webpage)
  48         if match:
  49             return 'wistia:%s' % match.group('id')
  50
  51         match = re.search(
  52             r'''(?sx)
  53                 <script[^>]+src=(["'])(?:https?:)?//fast\.wistia\.com/assets/external/E-v1\.js\1[^>]*>.*?
  54                 <div[^>]+class=(["']).*?\bwistia_async_(?P<id>[a-z0-9]+)\b.*?\2
  55             ''', webpage)
  56         if match:
  57             return 'wistia:%s' % match.group('id')
  58
  59     def _real_extract(self, url):
  60         video_id = self._match_id(url)
  61
  62         data_json = self._download_json(
  63             self._API_URL % video_id, video_id,
  64             # Some videos require this.
  65             headers={
  66                 'Referer': url if url.startswith('http') else self._IFRAME_URL % video_id,
  67             })
  68
  69         if data_json.get('error'):
  70             raise ExtractorError(
  71                 'Error while getting the playlist', expected=True)
  72
  73         data = data_json['media']
  74         title = data['name']
  75
  76         formats = []
  77         thumbnails = []
  78         for a in data['assets']:
  79             aurl = a.get('url')
  80             if not aurl:
  81                 continue
  82             astatus = a.get('status')
  83             atype = a.get('type')
  84             if (astatus is not None and astatus != 2) or atype in ('preview', 'storyboard'):
  85                 continue
  86             elif atype in ('still', 'still_image'):
  87                 thumbnails.append({
  88                     'url': aurl,
  89                     'width': int_or_none(a.get('width')),
  90                     'height': int_or_none(a.get('height')),
  91                 })
  92             else:
  93                 aext = a.get('ext')
  94                 is_m3u8 = a.get('container') == 'm3u8' or aext == 'm3u8'
  95                 formats.append({
  96                     'format_id': atype,
  97                     'url': aurl,
  98                     'tbr': int_or_none(a.get('bitrate')),
  99                     'vbr': int_or_none(a.get('opt_vbitrate')),
 100                     'width': int_or_none(a.get('width')),
 101                     'height': int_or_none(a.get('height')),
 102                     'filesize': int_or_none(a.get('size')),
 103                     'vcodec': a.get('codec'),
 104                     'container': a.get('container'),
 105                     'ext': 'mp4' if is_m3u8 else aext,
 106                     'protocol': 'm3u8' if is_m3u8 else None,
 107                     'preference': 1 if atype == 'original' else None,
 108                 })
 109
 110         self._sort_formats(formats)
 111
 112         return {
 113             'id': video_id,
 114             'title': title,
 115             'description': data.get('seoDescription'),
 116             'formats': formats,
 117             'thumbnails': thumbnails,
 118             'duration': float_or_none(data.get('duration')),
 119             'timestamp': int_or_none(data.get('createdAt')),
 120         }