youtube_dl/extractor/pearvideo.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     qualities,
   9     unified_timestamp,
  10 )
  11
  12
  13 class PearVideoIE(InfoExtractor):
  14     _VALID_URL = r'https?://(?:www\.)?pearvideo\.com/video_(?P<id>\d+)'
  15     _TEST = {
  16         'url': 'http://www.pearvideo.com/video_1076290',
  17         'info_dict': {
  18             'id': '1076290',
  19             'ext': 'mp4',
  20             'title': '小浣熊在主人家玻璃上滚石头：没砸',
  21             'description': 'md5:01d576b747de71be0ee85eb7cac25f9d',
  22             'timestamp': 1494275280,
  23             'upload_date': '20170508',
  24         }
  25     }
  26
  27     def _real_extract(self, url):
  28         video_id = self._match_id(url)
  29
  30         webpage = self._download_webpage(url, video_id)
  31
  32         quality = qualities(
  33             ('ldflv', 'ld', 'sdflv', 'sd', 'hdflv', 'hd', 'src'))
  34
  35         formats = [{
  36             'url': mobj.group('url'),
  37             'format_id': mobj.group('id'),
  38             'quality': quality(mobj.group('id')),
  39         } for mobj in re.finditer(
  40             r'(?P<id>[a-zA-Z]+)Url\s*=\s*(["\'])(?P<url>(?:https?:)?//.+?)\2',
  41             webpage)]
  42         self._sort_formats(formats)
  43
  44         title = self._search_regex(
  45             (r'<h1[^>]+\bclass=(["\'])video-tt\1[^>]*>(?P<value>[^<]+)',
  46              r'<[^>]+\bdata-title=(["\'])(?P<value>(?:(?!\1).)+)\1'),
  47             webpage, 'title', group='value')
  48         description = self._search_regex(
  49             (r'<div[^>]+\bclass=(["\'])summary\1[^>]*>(?P<value>[^<]+)',
  50              r'<[^>]+\bdata-summary=(["\'])(?P<value>(?:(?!\1).)+)\1'),
  51             webpage, 'description', default=None,
  52             group='value') or self._html_search_meta('Description', webpage)
  53         timestamp = unified_timestamp(self._search_regex(
  54             r'<div[^>]+\bclass=["\']date["\'][^>]*>([^<]+)',
  55             webpage, 'timestamp', fatal=False))
  56
  57         return {
  58             'id': video_id,
  59             'title': title,
  60             'description': description,
  61             'timestamp': timestamp,
  62             'formats': formats,
  63         }