子字符串或完全匹配的正则表达式

Question

Given the example string "hello", I need an expression for validating user input (any combination of the existing letters; without repeating a used letter).

In this context valid and invalid input examples are as follows:

valid: "hello", "hell", "lol" .... etc.

invalid: "heel", "loo"... etc.

I have tried the likes of ...

(.)*([hello])(.)*

and

[hello]+

But, they don't sort the invalid ones.

Any help would be appreciated thank you.

NOTE: This is not just substring or exact match, per the examples, combinations of letters are valid.

Answer 1

Regular expressions is not the right tool...they should be used for matching left to right, not counting various characters in a random order. You're better off having a validation string hello , looping through each character from the input string, and checking to see if the character exists (if it does, remove that character from the validation string and continue. otherwise, the input fails).

Here is a quick example I whipped up in Java :

public static boolean testString(String testString)
{
    String allowedCharacters = "hello";

    for(int i = 0; i < testString.length(); i++) {
        int position = allowedCharacters.indexOf(testString.charAt(i));

        if(position == -1) {
            System.out.println(testString + " - fail");
            return false;
        } else {
            allowedCharacters = allowedCharacters.substring(0, position)
                              + allowedCharacters.substring(position + 1);
        }
    }


    System.out.println(testString + " - success");
    return true;
}

Calling the function with example output:

testString("hello"); // hello - success
testString("hell");  // hell - success
testString("lol");   // lol - success

testString("heel");  // heel - fail
testString("loo");   // loo - fail