Noob Question - not sure what's wrong

vishi has asked for the wisdom of the Perl Monks concerning the following question:

Hi all,

I'm trying to do some input validation and I'm pretty sure I'm doing something really stupid that's making it not work the way I want to.

I want to accept values from users that are of the form:

SB0
SB00
[download]

So, basically, it is a string "SB" or "sb" (case insensitive)followed by either 1 or two digits. Everything else should fail, and prompt the user to enter the right string. Here's the code I have right now. Not sure where I am going wrong:


our $sbNumHost1='';

while ($sbNumHost1 eq "")
{
    print "Enter the SB## \t: ";
    chomp($sbNumHost1 = ReadLine(0));
    if ($sbNumHost1 !~ m/(SB)(\d{1,2})/i)
    {
        print "matched >>>>>>>$&<<<<<<<\n";
        $sbNumHost1 = '';
    }
}
print "done\n";
[download]

Unfortunately, this code gladly accepts inputs like SB111 which should not be a valid input.

Helllllllppppuh

Comment on Noob Question - not sure what's wrong Select or Download Code

Replies are listed 'Best First'.
Re: Noob Question - not sure what's wrong by ikegami (Patriarch) on May 11, 2011 at 07:02 UTC
"SB111" does indeed match the pattern. You want `/^SB[0-9]{1,2}\z/i`. Then there's the issue that you use `$&` when no match occur, even though it's only set on match. I think you use `!~` when you meant to use `=~`. Then there's the issue that you use `$&`, which causes all non-capturing regex matches in your program to become slower. `$sbNumHost1` would be a suitable replacement for `$&` here.	[reply] [d/l] [select]
Re^2: Noob Question - not sure what's wrong by vishi (Beadle) on May 11, 2011 at 07:11 UTC
Thanks ikegami, But I think I found what I was doing wrong. I simplified the code to this and it now works... `while ($sbNumHost1 !~ m/^(SB)(\d{1,2})$/i) { print "Enter the SB## \t: "; chomp($sbNumHost1 = ReadLine(0)); } print "done\n";` [download]	[reply] [d/l]
Re^3: Noob Question - not sure what's wrong by ikegami (Patriarch) on May 11, 2011 at 15:52 UTC
That matches "SB00\n" and you said it shouldn't. Replace /`$`/ with /`\z`/ to get the stated desired behaviour. `\d` can match more than 0-9. I presume you don't want to match arab digits. That uses needless captures. Get rid of them. `m/^(SB)(\d{1,2})$/i` [download] should be `m/^SB[0-9]{1,2}\z/i` [download]	[reply] [d/l] [select]