added an IE for criterion.com

[yt-dlp.git] / youtube_dl / extractor / criterion.py

# -*- coding: utf-8 -*-

import re

from .common import InfoExtractor

class CriterionIE(InfoExtractor):
    _VALID_URL = r'http://www.criterion.com/films/(.*)'

    def _real_extract(self, url):
        mobj = re.match(self._VALID_URL, url)
        video_id = mobj.group(1).split('-')[0]
        webpage = self._download_webpage(url, video_id)

        final_url = self._search_regex(r'so.addVariable\("videoURL", "(.+?)"\)\;',
                                webpage, 'video url')
        title = self._search_regex(r'<meta content="(.+?)" property="og:title" />',
                                webpage, 'video title')
        description = self._search_regex(r'<meta name="description" content="(.+?)" />',
                                webpage, 'video description')
        thumbnail = self._search_regex(r'so.addVariable\("thumbnailURL", "(.+?)"\)\;',
                                webpage, 'thumbnail url')
        ext = final_url.split('.')[-1]

        return {'id': video_id,
                'url' : final_url,
                'title': title,
                'ext': ext,
                'description': description,
                'thumbnail': thumbnail,
                }