Merge branch '_stream' of https://github.com/pulpe/youtube-dl into pulpe-_stream

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

# -*- coding: utf-8 -*-
from __future__ import unicode_literals

import re
import json

from .common import InfoExtractor


class StreamCZIE(InfoExtractor):
    _VALID_URL = r'https?://www\.stream\.cz/((?P<category>.+)/)?(?P<videogroup>.+)/(?P<videoid>.+)'

    _TESTS = [{
        'url': 'http://www.stream.cz/peklonataliri/765767-ecka-pro-deti',
        'md5': '6d3ca61a8d0633c9c542b92fcb936b0c',
        'info_dict': {
            'id': '765767',
            'ext': 'mp4',
            'title': 'Peklo na talíři: Éčka pro děti',
            'description': 'md5:49ace0df986e95e331d0fe239d421519',
            'thumbnail': 'http://im.stream.cz/episode/52961d7e19d423f8f06f0100',
        },
    },
    ]

    def _real_extract(self, url):
        mobj = re.match(self._VALID_URL, url)
        video_id = mobj.group('videoid')

        webpage = self._download_webpage(url, video_id)

        data = self._html_search_regex(r'Stream\.Data\.Episode\((.+?)\);', webpage, 'stream data')

        jsonData = json.loads(data)

        formats = []
        for video in jsonData['instances']:
            format_id = video['instances'][0]['quality']

            if format_id == '240p':
                quality = 0
            elif format_id == '360p':
                quality = 1
            elif format_id == '480p':
                quality = 2
            elif format_id == '720p':
                quality = 3

            formats.append({
                'format_id': format_id,
                'url': video['instances'][0]['source'],
                'quality': quality,
                'ext': 'mp4',
            })

        self._sort_formats(formats)

        return {
            'id': str(jsonData['id']),
            'title': self._og_search_title(webpage),
            'thumbnail': jsonData['episode_image_original_url'].replace('//', 'http://'),
            'formats': formats,
            'description': self._og_search_description(webpage),
        }