from __future__ import unicode_literals

from .common import InfoExtractor

from ..compat import compat_urlparse
from ..utils import qualities


class TheSceneIE(InfoExtractor):
    _VALID_URL = r'https?://thescene\.com/watch/[^/]+/(?P<id>[^/#?]+)'

    _TEST = {
        'url': 'https://thescene.com/watch/vogue/narciso-rodriguez-spring-2013-ready-to-wear',
        'info_dict': {
            'id': '520e8faac2b4c00e3c6e5f43',
            'ext': 'mp4',
            'title': 'Narciso Rodriguez: Spring 2013 Ready-to-Wear',
            'display_id': 'narciso-rodriguez-spring-2013-ready-to-wear',
        },
    }

    def _real_extract(self, url):
        display_id = self._match_id(url)

        webpage = self._download_webpage(url, display_id)

        player_url = compat_urlparse.urljoin(
            url,
            self._html_search_regex(
                r'id=\'js-player-script\'[^>]+src=\'(.+?)\'', webpage, 'player url'))

        player = self._download_webpage(player_url, display_id)
        info = self._parse_json(
            self._search_regex(
                r'(?m)video\s*:\s*({.+?}),$', player, 'info json'),
            display_id)

        qualities_order = qualities(('low', 'high'))
        formats = [{
            'format_id': '{0}-{1}'.format(f['type'].split('/')[0], f['quality']),
            'url': f['src'],
            'quality': qualities_order(f['quality']),
        } for f in info['sources']]
        self._sort_formats(formats)

        return {
            'id': info['id'],
            'display_id': display_id,
            'title': info['title'],
            'formats': formats,
            'thumbnail': info.get('poster_frame'),
        }