Re: Regex advice

>> if (/.*simscan~\[\d+\]~([\w|\s]+).*?~(\d+\.\d+)s~.*?~(.*?)~(.*?)~(.*)/) {
> -------------------------^^^^^^^^^^
> This captures the "result", but with a trailing space. You may want to 
> let the capturing parentheses be followed by \s to prevent that.
> 
>>                 $result   = $1 if $1;
>>                 $scantime = $2 if $2;
>>                 $ip       = $3 if $3;
>>                 $from     = $4 if $4;
>>                 $to       = $5 if $5;
> 
> The statement modifiers seem redundant to me.

I don't grasp what you mean, either by the "captures the result" 
statement, or the modifier statement. I thought that the [\w|\s]+  was 
'word or space in any quantity', which would capture either:

"CLEAN"

or

"SPAM REJECT"

Would you please re-write the statement in the context you meant?

> Did you consider using split() ?

...actually, no. Now that I think about it, that is how I do almost all 
other work that is like this.

To be honest, I was following a similar statement in another script that 
I thought I'd try my hand at custom regex with. I'm finding it 
exceptionally tedious and error prone. As I've been going through 
testing, I can't envision myself writing a regex that I can't see a 
potential to fail for :)

> while (<LOG>) {
>     chomp;
>     my ($result, $scantime, $ip, $from, $to)
>       = ( split /~/ )[2,3,5..7] or next;
> 
>     $result =~ s/\s+\(.+//;
>     chop $scantime;

I don't know why this flew right by me today. Thanks for the feedback.

I'm now off to write this in, with some checking so that my $ip lines up 
with an IP, and not with a piece of a subject line where someone threw 
in a rogue ~. (The delimiter is mine, I hacked Simscan to output that as 
it's delimiter for other custom reasons).

Thanks Gunnar,

Steve