Java字符串分割长度

Question

System.out.println("neon.mems.cmu.edu/people/".split("/").length); // output is 2

I was doing some url processing. To my surprise I just got the result above. I thought the number of elements could be the number of splitters plus one.

I didn't realize the last empty string(or just null ) is cut off from the splitted array until now. I wonder if this is the case with every programming language.

Answer 1

No that's not the case for every programming language and there is no universal specification so there is no reason it should be.

Go

a := strings.Split("neon.mems.cmu.edu/people/", "/")
fmt.Println(len(a)) // prints 3

Javascript

Type this in the console of your browser :

"neon.mems.cmu.edu/people/".split('/')

The result is

["neon.mems.cmu.edu", "people", ""]

What you should do when a match is empty isn't something obvious or inherent to the split concept. A proof of that is that old Internet Explorer versions did remove those empty matches .

Answer 2

why it is discarded empty string?

String#split(regx) internally call String#split(regx,0) which execute Pattern.compile(regex).split(this, limit); actually - code snippet

if (limit == 0)
        while (resultSize > 0 && matchList.get(resultSize-1).equals(""))
            resultSize--;

where empty string has been discarded from resultSize if limit is 0.

How to get desired result?

String#split(regx,limit) use get desired result.

System.out.println("neon.mems.cmu.edu/people/".split("/",3).length);

Result :

 3

And about language specification I am agree with @dystroy