youtube_dl/extractor/joj.py

   1 # coding: utf-8\r
   2 from __future__ import unicode_literals\r
   3 \r
   4 import re\r
   5 \r
   6 from .common import InfoExtractor\r
   7 from ..compat import compat_str\r
   8 from ..utils import (\r
   9     int_or_none,\r
  10     js_to_json,\r
  11     try_get,\r
  12 )\r
  13 \r
  14 \r
  15 class JojIE(InfoExtractor):\r
  16     _VALID_URL = r'''(?x)\r
  17                     (?:\r
  18                         joj:|\r
  19                         https?://media\.joj\.sk/embed/\r
  20                     )\r
  21                     (?P<id>[^/?#^]+)\r
  22                 '''\r
  23     _TESTS = [{\r
  24         'url': 'https://media.joj.sk/embed/a388ec4c-6019-4a4a-9312-b1bee194e932',\r
  25         'info_dict': {\r
  26             'id': 'a388ec4c-6019-4a4a-9312-b1bee194e932',\r
  27             'ext': 'mp4',\r
  28             'title': 'NOVÉ BÝVANIE',\r
  29             'thumbnail': r're:^https?://.*\.jpg$',\r
  30             'duration': 3118,\r
  31         }\r
  32     }, {\r
  33         'url': 'https://media.joj.sk/embed/9i1cxv',\r
  34         'only_matching': True,\r
  35     }, {\r
  36         'url': 'joj:a388ec4c-6019-4a4a-9312-b1bee194e932',\r
  37         'only_matching': True,\r
  38     }, {\r
  39         'url': 'joj:9i1cxv',\r
  40         'only_matching': True,\r
  41     }]\r
  42 \r
  43     @staticmethod\r
  44     def _extract_urls(webpage):\r
  45         return [\r
  46             mobj.group('url')\r
  47             for mobj in re.finditer(\r
  48                 r'<iframe\b[^>]+\bsrc=(["\'])(?P<url>(?:https?:)?//media\.joj\.sk/embed/(?:(?!\1).)+)\1',\r
  49                 webpage)]\r
  50 \r
  51     def _real_extract(self, url):\r
  52         video_id = self._match_id(url)\r
  53 \r
  54         webpage = self._download_webpage(\r
  55             'https://media.joj.sk/embed/%s' % video_id, video_id)\r
  56 \r
  57         title = self._search_regex(\r
  58             (r'videoTitle\s*:\s*(["\'])(?P<title>(?:(?!\1).)+)\1',\r
  59              r'<title>(?P<title>[^<]+)'), webpage, 'title',\r
  60             default=None, group='title') or self._og_search_title(webpage)\r
  61 \r
  62         bitrates = self._parse_json(\r
  63             self._search_regex(\r
  64                 r'(?s)(?:src|bitrates)\s*=\s*({.+?});', webpage, 'bitrates',\r
  65                 default='{}'),\r
  66             video_id, transform_source=js_to_json, fatal=False)\r
  67 \r
  68         formats = []\r
  69         for format_url in try_get(bitrates, lambda x: x['mp4'], list) or []:\r
  70             if isinstance(format_url, compat_str):\r
  71                 height = self._search_regex(\r
  72                     r'(\d+)[pP]\.', format_url, 'height', default=None)\r
  73                 formats.append({\r
  74                     'url': format_url,\r
  75                     'format_id': '%sp' % height if height else None,\r
  76                     'height': int(height),\r
  77                 })\r
  78         if not formats:\r
  79             playlist = self._download_xml(\r
  80                 'https://media.joj.sk/services/Video.php?clip=%s' % video_id,\r
  81                 video_id)\r
  82             for file_el in playlist.findall('./files/file'):\r
  83                 path = file_el.get('path')\r
  84                 if not path:\r
  85                     continue\r
  86                 format_id = file_el.get('id') or file_el.get('label')\r
  87                 formats.append({\r
  88                     'url': 'http://n16.joj.sk/storage/%s' % path.replace(\r
  89                         'dat/', '', 1),\r
  90                     'format_id': format_id,\r
  91                     'height': int_or_none(self._search_regex(\r
  92                         r'(\d+)[pP]', format_id or path, 'height',\r
  93                         default=None)),\r
  94                 })\r
  95         self._sort_formats(formats)\r
  96 \r
  97         thumbnail = self._og_search_thumbnail(webpage)\r
  98 \r
  99         duration = int_or_none(self._search_regex(\r
 100             r'videoDuration\s*:\s*(\d+)', webpage, 'duration', fatal=False))\r
 101 \r
 102         return {\r
 103             'id': video_id,\r
 104             'title': title,\r
 105             'thumbnail': thumbnail,\r
 106             'duration': duration,\r
 107             'formats': formats,\r
 108         }\r