[YoutubeDL] Ensure dir existence for each requested format (closes #14116)

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

# coding: utf-8\r
from __future__ import unicode_literals\r
\r
import re\r
\r
from .common import InfoExtractor\r
from ..compat import compat_str\r
from ..utils import (\r
    int_or_none,\r
    js_to_json,\r
    try_get,\r
)\r
\r
\r
class JojIE(InfoExtractor):\r
    _VALID_URL = r'''(?x)\r
                    (?:\r
                        joj:|\r
                        https?://media\.joj\.sk/embed/\r
                    )\r
                    (?P<id>[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})\r
                '''\r
    _TESTS = [{\r
        'url': 'https://media.joj.sk/embed/a388ec4c-6019-4a4a-9312-b1bee194e932',\r
        'info_dict': {\r
            'id': 'a388ec4c-6019-4a4a-9312-b1bee194e932',\r
            'ext': 'mp4',\r
            'title': 'NOVÉ BÝVANIE',\r
            'thumbnail': r're:^https?://.*\.jpg$',\r
            'duration': 3118,\r
        }\r
    }, {\r
        'url': 'joj:a388ec4c-6019-4a4a-9312-b1bee194e932',\r
        'only_matching': True,\r
    }]\r
\r
    @staticmethod\r
    def _extract_urls(webpage):\r
        return re.findall(\r
            r'<iframe\b[^>]+\bsrc=["\'](?P<url>(?:https?:)?//media\.joj\.sk/embed/[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})',\r
            webpage)\r
\r
    def _real_extract(self, url):\r
        video_id = self._match_id(url)\r
\r
        webpage = self._download_webpage(\r
            'https://media.joj.sk/embed/%s' % video_id, video_id)\r
\r
        title = self._search_regex(\r
            (r'videoTitle\s*:\s*(["\'])(?P<title>(?:(?!\1).)+)\1',\r
             r'<title>(?P<title>[^<]+)'), webpage, 'title',\r
            default=None, group='title') or self._og_search_title(webpage)\r
\r
        bitrates = self._parse_json(\r
            self._search_regex(\r
                r'(?s)bitrates\s*=\s*({.+?});', webpage, 'bitrates',\r
                default='{}'),\r
            video_id, transform_source=js_to_json, fatal=False)\r
\r
        formats = []\r
        for format_url in try_get(bitrates, lambda x: x['mp4'], list) or []:\r
            if isinstance(format_url, compat_str):\r
                height = self._search_regex(\r
                    r'(\d+)[pP]\.', format_url, 'height', default=None)\r
                formats.append({\r
                    'url': format_url,\r
                    'format_id': '%sp' % height if height else None,\r
                    'height': int(height),\r
                })\r
        if not formats:\r
            playlist = self._download_xml(\r
                'https://media.joj.sk/services/Video.php?clip=%s' % video_id,\r
                video_id)\r
            for file_el in playlist.findall('./files/file'):\r
                path = file_el.get('path')\r
                if not path:\r
                    continue\r
                format_id = file_el.get('id') or file_el.get('label')\r
                formats.append({\r
                    'url': 'http://n16.joj.sk/storage/%s' % path.replace(\r
                        'dat/', '', 1),\r
                    'format_id': format_id,\r
                    'height': int_or_none(self._search_regex(\r
                        r'(\d+)[pP]', format_id or path, 'height',\r
                        default=None)),\r
                })\r
        self._sort_formats(formats)\r
\r
        thumbnail = self._og_search_thumbnail(webpage)\r
\r
        duration = int_or_none(self._search_regex(\r
            r'videoDuration\s*:\s*(\d+)', webpage, 'duration', fatal=False))\r
\r
        return {\r
            'id': video_id,\r
            'title': title,\r
            'thumbnail': thumbnail,\r
            'duration': duration,\r
            'formats': formats,\r
        }\r