How to split a String into variable length tokens ?

Hi All,

Is there any built in library or api provided by Sun or other open source projects, where it would let me split a String into variable length tokens ?

For eg, If I have a string like below

If I want to split this string into tokens, based on variable length tokens, by providing size in some sort of regex (I could not put this in write wordings)

To be more specific, I should be able to split this string based on a copybook structure defined in COBOL. like

01 STRING1 PIC X(3)
02 STRING2 PIC X(5)
etc

I can write my userdefined function, but wondering if anything already existing.
Thanks

If you know the lengths, just use substring(...)

If you want to find a particular token to split on, you can use String#split(java.lang.String).
Or try one of the overloadings of String#indexOf() to find where to start your substrings.

a regex will also let you specify how many characters to match:

[a-z]\{4,8\}

will match 4,5,6,7, or 8 lowercase letters.

fred rosenberger wrote:a regex will also let you specify how many characters to match:

[a-z]\{4,8\}

will match 4,5,6,7, or 8 lowercase letters.

Actually, it'd be:

(No backslash before the braces. I always get confused about what to backslash. It's different among Java, sed, egrep, and vi.)

Kumar Raja wrote:I can write my userdefined function, but wondering if anything already existing.

Seems to me you have 2 tasks:
1. Splitting your string based on predefined lengths.
2. Getting those predefined lengths from a piece of COBOL source.
And of the two, number 2 is going to be a lot harder (the first, as Darryl said, can be done with simple substring()s).

Winston

Jeff Verdegan wrote:
Actually, it'd be:
"[a-z]{4,8}"
(No backslash before the braces. I always get confused about what to backslash. It's different among Java, sed, egrep, and vi.)

Thanks...yeah, I do a lot of trial and error with that myself...

Thank you all for the suggestions. To me regex approach seem to be more easy. Thanks.