find() method in Regex

[Astha Sharma]

import java.util.regex.*;
class Test
{
    public static void main(String[] args)
    {
        Pattern p=Pattern.compile(args[0]);
        Matcher m=p.matcher(args[1]);
        boolean b=false;
        while(b=m.find())
        {
        	System.out.println(m.start()+"\t"+m.group());
        }
    }
}

command line is- "\w*" ab23

output-
0 ab23
4


I estimated the output as-
0 ab23

My doubt is why find() method is searching upto index 4? end of string is at index 3, xo why find() goes one index beyond the length?

[Henry Wong]

Astha Sharma wrote:

import java.util.regex.*;
class Test
{
    public static void main(String[] args)
    {
        Pattern p=Pattern.compile(args[0]);
        Matcher m=p.matcher(args[1]);
        boolean b=false;
        while(b=m.find())
        {
        	System.out.println(m.start()+"\t"+m.group());
        }
    }
}

command line is- "\w*" ab23

output-
0 ab23
4


I estimated the output as-
0 ab23

My doubt is why find() method is searching upto index 4? end of string is at index 3, xo why find() goes one index beyond the length?

The pattern is capable of matching a zero-length string. And you can have a zero length string, at the end of the input string.

Henry

[Matheus Souza]

Henry Wong wrote:

Astha Sharma wrote:

import java.util.regex.*;
class Test
{
    public static void main(String[] args)
    {
        Pattern p=Pattern.compile(args[0]);
        Matcher m=p.matcher(args[1]);
        boolean b=false;
        while(b=m.find())
        {
        	System.out.println(m.start()+"\t"+m.group());
        }
    }
}

command line is- "\w*" ab23

output-
0 ab23
4


I estimated the output as-
0 ab23

My doubt is why find() method is searching upto index 4? end of string is at index 3, xo why find() goes one index beyond the length?

The pattern is capable of matching a zero-length string. And you can have a zero length string, at the end of the input string.

Henry

Complementing what Henry said:
Just put aroung the m.group(), these characters ">" and "<". You'll see that the quantificator '*' prints an empty string too.

[Dan Drillich]

Matheus Souza wrote:Just put aroung the m.group(), these characters ">" and "<". You'll see that the quantificator '*' prints an empty string too.

Right, it prints now -

0       <ab23>
4       <>

Regards,
Dan

[Astha Sharma]

thanks to all for replying. I understand that pattern will search for a zero length string also. But i'm not getting that why it is continuing search beyond the matcher string. Matcher string doesn't have index upto 4.

[Henry Wong]

Astha Sharma wrote:thanks to all for replying. I understand that pattern will search for a zero length string also. But i'm not getting that why it is continuing search beyond the matcher string. Matcher string doesn't have index upto 4.

Regular expressions searches for things other than text strings -- including word boundaries, end of input boundaries, etc. As such, the position at the end of text is a valid position.

Now, a position such as 5 (greater than the length) is beyond the string.

Henry

[Matheus Souza]

Astha Sharma wrote:thanks to all for replying. I understand that pattern will search for a zero length string also. But i'm not getting that why it is continuing search beyond the matcher string. Matcher string doesn't have index upto 4.

The quantificator '*' search for zero, one or many correspondences, so the empty string in the end kind of represents the final of the string and its an correspondence (of 0)

[Dan Drillich]

We can look at -

java Regex "\w*" "123 abc"

0       <123>
3       <>
4       <abc>
7       <>

versus -

java Regex "\w+" "123 abc"

0       <123>
4       <abc>

Regards,
Dan