youtube_dl/extractor/vulture.py

   1 from __future__ import unicode_literals
   2
   3 import json
   4 import os.path
   5 import re
   6
   7 from .common import InfoExtractor
   8 from ..utils import (
   9     int_or_none,
  10     parse_iso8601,
  11 )
  12
  13
  14 class VultureIE(InfoExtractor):
  15     IE_NAME = 'vulture.com'
  16     _VALID_URL = r'https?://video\.vulture\.com/video/(?P<display_id>[^/]+)/'
  17     _TEST = {
  18         'url': 'http://video.vulture.com/video/Mindy-Kaling-s-Harvard-Speech/player?layout=compact&read_more=1',
  19         'md5': '8d997845642a2b5152820f7257871bc8',
  20         'info_dict': {
  21             'id': '6GHRQL3RV7MSD1H4',
  22             'ext': 'mp4',
  23             'title': 'kaling-speech-2-MAGNIFY STANDARD CONTAINER REVISED',
  24             'uploader_id': 'Sarah',
  25             'thumbnail': 're:^http://.*\.jpg$',
  26             'timestamp': 1401288564,
  27             'upload_date': '20140528',
  28             'description': 'Uplifting and witty, as predicted.',
  29             'duration': 1015,
  30         }
  31     }
  32
  33     def _real_extract(self, url):
  34         mobj = re.match(self._VALID_URL, url)
  35         display_id = mobj.group('display_id')
  36
  37         webpage = self._download_webpage(url, display_id)
  38         query_string = self._search_regex(
  39             r"queryString\s*=\s*'([^']+)'", webpage, 'query string')
  40         video_id = self._search_regex(
  41             r'content=([^&]+)', query_string, 'video ID')
  42         query_url = 'http://video.vulture.com/embed/player/container/1000/1000/?%s' % query_string
  43
  44         query_webpage = self._download_webpage(
  45             query_url, display_id, note='Downloading query page')
  46         params_json = self._search_regex(
  47             r'(?sm)new MagnifyEmbeddablePlayer\({.*?contentItem:\s*(\{.*?\})\n?,\n',
  48             query_webpage,
  49             'player params')
  50         params = json.loads(params_json)
  51
  52         upload_timestamp = parse_iso8601(params['posted'].replace(' ', 'T'))
  53         uploader_id = params.get('user', {}).get('handle')
  54
  55         media_item = params['media_item']
  56         title = os.path.splitext(media_item['title'])[0]
  57         duration = int_or_none(media_item.get('duration_seconds'))
  58
  59         return {
  60             'id': video_id,
  61             'display_id': display_id,
  62             'url': media_item['pipeline_xid'],
  63             'title': title,
  64             'timestamp': upload_timestamp,
  65             'thumbnail': params.get('thumbnail_url'),
  66             'uploader_id': uploader_id,
  67             'description': params.get('description'),
  68             'duration': duration,
  69         }