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|com)/embed/(?:iframe|medias)/)(?P<id>[a-z0-9]{10})'
  16     _EMBED_BASE_URL = 'http://fast.wistia.com/embed/'
  17
  18     _TESTS = [{
  19         'url': 'http://fast.wistia.net/embed/iframe/sh7fpupwlt',
  20         'md5': 'cafeb56ec0c53c18c97405eecb3133df',
  21         'info_dict': {
  22             'id': 'sh7fpupwlt',
  23             'ext': 'mov',
  24             'title': 'Being Resourceful',
  25             'description': 'a Clients From Hell Video Series video from worldwidewebhosting',
  26             'upload_date': '20131204',
  27             'timestamp': 1386185018,
  28             'duration': 117,
  29         },
  30     }, {
  31         'url': 'wistia:sh7fpupwlt',
  32         'only_matching': True,
  33     }, {
  34         # with hls video
  35         'url': 'wistia:807fafadvk',
  36         'only_matching': True,
  37     }, {
  38         'url': 'http://fast.wistia.com/embed/iframe/sh7fpupwlt',
  39         'only_matching': True,
  40     }, {
  41         'url': 'http://fast.wistia.net/embed/medias/sh7fpupwlt.json',
  42         'only_matching': True,
  43     }]
  44
  45     # https://wistia.com/support/embed-and-share/video-on-your-website
  46     @staticmethod
  47     def _extract_url(webpage):
  48         match = re.search(
  49             r'<(?:meta[^>]+?content|(?:iframe|script)[^>]+?src)=["\'](?P<url>(?:https?:)?//(?:fast\.)?wistia\.(?:net|com)/embed/(?:iframe|medias)/[a-z0-9]{10})', webpage)
  50         if match:
  51             return unescapeHTML(match.group('url'))
  52
  53         match = re.search(
  54             r'''(?sx)
  55                 <script[^>]+src=(["'])(?:https?:)?//fast\.wistia\.com/assets/external/E-v1\.js\1[^>]*>.*?
  56                 <div[^>]+class=(["']).*?\bwistia_async_(?P<id>[a-z0-9]{10})\b.*?\2
  57             ''', webpage)
  58         if match:
  59             return 'wistia:%s' % match.group('id')
  60
  61         match = re.search(r'(?:data-wistia-?id=["\']|Wistia\.embed\(["\']|id=["\']wistia_)(?P<id>[a-z0-9]{10})', webpage)
  62         if match:
  63             return 'wistia:%s' % match.group('id')
  64
  65     def _real_extract(self, url):
  66         video_id = self._match_id(url)
  67
  68         data_json = self._download_json(
  69             self._EMBED_BASE_URL + 'medias/%s.json' % video_id, video_id,
  70             # Some videos require this.
  71             headers={
  72                 'Referer': url if url.startswith('http') else self._EMBED_BASE_URL + 'iframe/' + video_id,
  73             })
  74
  75         if data_json.get('error'):
  76             raise ExtractorError(
  77                 'Error while getting the playlist', expected=True)
  78
  79         data = data_json['media']
  80         title = data['name']
  81
  82         formats = []
  83         thumbnails = []
  84         for a in data['assets']:
  85             aurl = a.get('url')
  86             if not aurl:
  87                 continue
  88             astatus = a.get('status')
  89             atype = a.get('type')
  90             if (astatus is not None and astatus != 2) or atype in ('preview', 'storyboard'):
  91                 continue
  92             elif atype in ('still', 'still_image'):
  93                 thumbnails.append({
  94                     'url': aurl,
  95                     'width': int_or_none(a.get('width')),
  96                     'height': int_or_none(a.get('height')),
  97                     'filesize': int_or_none(a.get('size')),
  98                 })
  99             else:
 100                 aext = a.get('ext')
 101                 display_name = a.get('display_name')
 102                 format_id = atype
 103                 if atype and atype.endswith('_video') and display_name:
 104                     format_id = '%s-%s' % (atype[:-6], display_name)
 105                 f = {
 106                     'format_id': format_id,
 107                     'url': aurl,
 108                     'tbr': int_or_none(a.get('bitrate')) or None,
 109                     'preference': 1 if atype == 'original' else None,
 110                 }
 111                 if display_name == 'Audio':
 112                     f.update({
 113                         'vcodec': 'none',
 114                     })
 115                 else:
 116                     f.update({
 117                         'width': int_or_none(a.get('width')),
 118                         'height': int_or_none(a.get('height')),
 119                         'vcodec': a.get('codec'),
 120                     })
 121                 if a.get('container') == 'm3u8' or aext == 'm3u8':
 122                     ts_f = f.copy()
 123                     ts_f.update({
 124                         'ext': 'ts',
 125                         'format_id': f['format_id'].replace('hls-', 'ts-'),
 126                         'url': f['url'].replace('.bin', '.ts'),
 127                     })
 128                     formats.append(ts_f)
 129                     f.update({
 130                         'ext': 'mp4',
 131                         'protocol': 'm3u8_native',
 132                     })
 133                 else:
 134                     f.update({
 135                         'container': a.get('container'),
 136                         'ext': aext,
 137                         'filesize': int_or_none(a.get('size')),
 138                     })
 139                 formats.append(f)
 140
 141         self._sort_formats(formats)
 142
 143         subtitles = {}
 144         for caption in data.get('captions', []):
 145             language = caption.get('language')
 146             if not language:
 147                 continue
 148             subtitles[language] = [{
 149                 'url': self._EMBED_BASE_URL + 'captions/' + video_id + '.vtt?language=' + language,
 150             }]
 151
 152         return {
 153             'id': video_id,
 154             'title': title,
 155             'description': data.get('seoDescription'),
 156             'formats': formats,
 157             'thumbnails': thumbnails,
 158             'duration': float_or_none(data.get('duration')),
 159             'timestamp': int_or_none(data.get('createdAt')),
 160             'subtitles': subtitles,
 161         }