Cool! Ruby encoding detector [Universal encoding detector in Ruby]

I used to want to use Java for something similar. Today I see Ruby already has
-----
Gem install Chardet-y
-----

Require 'rubygems'
Require 'universaldetector'
Require 'net/http'
Net: http. version_1_2
Net: http. Start ('www .sina.com.cn ') {| HTTP |
Data = http. Get ("/"). Body
P universaldetector: chardet (data)
}
It is a Python version transplant, and there is a confidence parameter. I tested several sites in China and it is very accurate. However, I found that as long as it is utf8 or gb2312, all the returned confidence is 0.99, only some foreign pure ASCII web pages will return 1.0