finding middle character in string using regex only

Question

How can I find middle character with regex only

For example,this shows the expected output

Hello -> l

world -> r

merged -> rg (see this for even number of occurances)

hi -> hi

I -> I

I tried

(?<=\w+).(?=\w+)

I'm curious as to why you would want the overhead of RegEx for this. — BLaZuRE, Jun 26 '13 at 06:44
@NAMO This is not a good way to learn regexes. Learn how to use them on situations where you need them. — Maroun, Jun 26 '13 at 06:46
@NAMO you can read about Regex here http://www.regular-expressions.info/wordboundaries.html — Ruchira Gayan Ranaweera, Jun 26 '13 at 06:51
Regexes don't solve every problem. In general they are very bad at solving problems that require counting, or remembering numbers. — Patashu, Jun 26 '13 at 06:52
The right way to learn about regexes is to read a good book on the subject; e.g. Aho, Sethi & Ullman, or Freidl. Freidl is more practical ... but the former (and similar) covers the theory. — Stephen C, Jun 26 '13 at 09:05

score 5 · Accepted Answer · answered Jun 26 '13 at 06:45

5

Regular expressions cannot count in the way that you are looking for. This looks like something regular expressions cannot accomplish. I suggest writing code to solve this.

answered Jun 26 '13 at 06:45

Pieter van Ginkel

29,160
8
71
111

1

+1 - So far, this is the only Answer that actually answers the question. The OP specifically wants to do it using a regex ... to understand the limits of regexes. – Stephen C Jun 26 '13 at 08:44

score 1 · Answer 2 · answered Jun 26 '13 at 06:47

1

String str="Hello";
String mid="";
int len = str.length();
if(len%2==1)
    mid= Character.toString(str.getCharAt(len/2));
else
    mid= Character.toString(str.getChatAt(len/2))+ Character.toStringstr.getCharAt((len/2)-1));

This should probably work.

answered Jun 26 '13 at 06:47

Veer Shrivastav

5,434
11
53
83

1

Adding to chars together doesn't result in concatenation (even in reverse order). Anyway OP needs regex-only way. – default locale Jun 26 '13 at 06:49

score 1 · Answer 3 · answered Jun 26 '13 at 06:55

1

public static void main(String[] args) {
        String s = "jogijogi";
        int size = s.length() / 2;
        String temp = "";
        if (s.length() % 2 == 0) {
            temp = s.substring(size - 1, (s.length() - size) + 1);
        } else if (s.length() % 2 != 0) {
            temp = s.substring(size, (s.length() - size));
        } else {
            temp = s.substring(1);

        }

        System.out.println(temp);
    }

answered Jun 26 '13 at 06:55

shreyansh jogi

2,082
12
20

1

Does not answer the question ... especially if you consider the OP's stated reason he asked it. – Stephen C Jun 26 '13 at 08:45

bobble bubble · Answer 4 · 2022-12-06T22:53:15.900

Related: How to match the middle character in a string with regex?

The following regex is based on @jaytea's approach and works well with e.g. PCRE, Java or C#.

^(?:.(?=.+?(.\1?$)))*?(^..?$|..?(?=\1$))

Here is the demo at regex101 and a .NET demo at RegexPlanet (click the green ".NET" button)

Middle character(s) will be found in the second capturing group. The goal is to capture two middle characters if there is an even amount of characters, else one. It works by a growing capture towards the end (first group) while lazily going through the string until it ends with the captured substring that grows with each repitition. ^..?$ is used to match strings with one or two characters length.

This "growing" works with capturing inside a repeated lookahead by placing an optional reference to the same group together with a freshly captured character into that group (further reading here).

A PCRE-variant with \K to reset and full matches: ^(?:.(?=.+?(.\1?$)))+?\K..?(?=\1$)|^..?

Curious about the "easy solution using balancing groups" that @Qtax mentions in his question.

finding middle character in string using regex only

4 Answers4

Linked