yt_dlp/extractor/pearvideo.py

   1 import re
   2
   3 from .common import InfoExtractor
   4 from ..utils import (
   5     qualities,
   6     traverse_obj,
   7     unified_timestamp,
   8 )
   9
  10
  11 class PearVideoIE(InfoExtractor):
  12     _VALID_URL = r'https?://(?:www\.)?pearvideo\.com/video_(?P<id>\d+)'
  13     _TEST = {
  14         'url': 'http://www.pearvideo.com/video_1076290',
  15         'info_dict': {
  16             'id': '1076290',
  17             'ext': 'mp4',
  18             'title': '小浣熊在主人家玻璃上滚石头：没砸',
  19             'description': 'md5:01d576b747de71be0ee85eb7cac25f9d',
  20             'timestamp': 1494275280,
  21             'upload_date': '20170508',
  22         },
  23     }
  24
  25     def _real_extract(self, url):
  26         video_id = self._match_id(url)
  27
  28         webpage = self._download_webpage(url, video_id)
  29
  30         quality = qualities(
  31             ('ldflv', 'ld', 'sdflv', 'sd', 'hdflv', 'hd', 'src'))
  32
  33         formats = [{
  34             'url': mobj.group('url'),
  35             'format_id': mobj.group('id'),
  36             'quality': quality(mobj.group('id')),
  37         } for mobj in re.finditer(
  38             r'(?P<id>[a-zA-Z]+)Url\s*=\s*(["\'])(?P<url>(?:https?:)?//.+?)\2',
  39             webpage)]
  40         if not formats:
  41             info = self._download_json(
  42                 'https://www.pearvideo.com/videoStatus.jsp', video_id=video_id,
  43                 query={'contId': video_id}, headers={'Referer': url})
  44             formats = [{
  45                 'format_id': k,
  46                 'url': v.replace(info['systemTime'], f'cont-{video_id}') if k == 'srcUrl' else v,
  47             } for k, v in traverse_obj(info, ('videoInfo', 'videos'), default={}).items() if v]
  48
  49         title = self._search_regex(
  50             (r'<h1[^>]+\bclass=(["\'])video-tt\1[^>]*>(?P<value>[^<]+)',
  51              r'<[^>]+\bdata-title=(["\'])(?P<value>(?:(?!\1).)+)\1'),
  52             webpage, 'title', group='value')
  53         description = self._search_regex(
  54             (r'<div[^>]+\bclass=(["\'])summary\1[^>]*>(?P<value>[^<]+)',
  55              r'<[^>]+\bdata-summary=(["\'])(?P<value>(?:(?!\1).)+)\1'),
  56             webpage, 'description', default=None,
  57             group='value') or self._html_search_meta('Description', webpage)
  58         timestamp = unified_timestamp(self._search_regex(
  59             r'<div[^>]+\bclass=["\']date["\'][^>]*>([^<]+)',
  60             webpage, 'timestamp', fatal=False))
  61
  62         return {
  63             'id': video_id,
  64             'title': title,
  65             'description': description,
  66             'timestamp': timestamp,
  67             'formats': formats,
  68         }