Regular Expressions

Garibaldi · 01-02-2006 07:30PM #1

Evenin' all,

Any regular expressions experts out there? I've got this:

^(([\w]*\d+[a-zA-Z]+)|([a-zA-Z]+\d+[\w]*))\d*$

which checks a string to see if it contains at least one letter and at least one number, but what I also need is to ensure that the string being searched is between 6 and 12 characters in length {6,12}. I've tried all sorts of combinations of brackets and replacing the *s with {6,12}, and adding:

\A[\w]{6,12}$

to the end of the existing line, all to no avail. If there's anyone who can put this out of my misery, I'd really appreciate it. While you're at it, if you can recommend a good book with in-depth explanations of regexps, I'd appreciate that too.

Thanks for any help you can give!

G

bpmurray · 02-02-2006 01:18AM

The bible: http://www.oreilly.com/catalog/regex/

daymobrew · 02-02-2006 10:01AM

Maybe a single regular expression can do all that you need but that would be beyond me. For easier maintenance in the future its best to try to keep it simple, even if it ends up being longer.
I just added a regex to check the length.

#!/usr/bin/perl -w
use strict;

my @strs = qw/ a1b2c abcedfg a1q2w3w abcd 1234 12345678901q  abcdefghijklmn1 /;

foreach my $str ( @strs )
{
  # "[a-zA-Z]" abbreviated to "\w" and redundant [] removed.
  if ( $str =~ /^((\w*\d+\w+)|(\w+\d+\w*))\d*$/ && $str =~ /^.{6,12}$/ )
  {
    print "$str = MATCHES\n";
  }
  else
  {
    print "$str = No match\n";
  }
}

Script output wrote:

a1b2c = No match
abcedfg = No match
a1q2w3w = MATCHES
abcd = No match
1234 = No match
12345678901q = MATCHES
abcdefghijklmn1 = No match

Garibaldi · 02-02-2006 10:54AM

Cheers, gents. My bacon has been saved.

daymobrew · 02-02-2006 01:34PM

# "[a-zA-Z]" abbreviated to "\w" and redundant [] removed.

I was just reading through a regex book - \w matches digits too, so cannot be used as a substitute for "[a-zA-Z]". So, use your original regex with the length check regex.

Regular Expressions

Comments