Hash problem

[Nesh]

Hello Perl Monks I have a text file as

[17/May/2005:18:15:59 -0700]sdjhsj USERID=t70xvcy; jdhdh JSESSIONID=00
+00RDnYTABcRWOx0UN2Zq--sZB:10f408hdg
[17/May/2005:18:15:59 -0700]sdjhsj USERID=t70xvcy; jdhdh JSESSIONID=00
+00RDnYTABcRWOx0UN2Zq--sZB:10f408hdg
[17/May/2005:18:15:59 -0700]sdjhsj USERID=t70xvcy; jdhdh JSESSIONID=00
+00RDnYTABcRWOx0UN2Zq--sZB:10f408g13

I have the code as

my $FileName = "test.txt";
my ($sec,$min,$hour,$mday,$mon,$year,$wday,$yday,$isdst);
   ($sec,$min,$hour,$mday,$mon,$year,$wday,$yday,$isdst)=localtime(tim
+e);  
   
open ACCESS,"< $FileName" || die "Can't open the file....";

my %User_Lst;
while (<ACCESS>) {
 if (/:(..):(..):(..).*USERID=([a-z0-9].+; ).*JSESSIONID=([a-z0-9].+)/
+){
  $l_hr=$1;  $l_min=$2;  $l_sec=$3;
  $UserID=$4;
  $session=$5;
  $UserID=~s/;.*//g;
  $session=~s/.*://g;

   
  $User_Lst{$UserID}=join(":",$l_hr,$l_min,$l_sec);
#  $User_Lst{$sessionid}=$session;
 }
}

my $l_Counter=0;
#while ( (($UserID,$sessionid $TStamp) = each (%User_Lst)) and (($sess
+ionid) = each (%Session_Lst))) {
while ( ($UserID, $TStamp) = each (%User_Lst)) {
      
     print sprintf("Server Name :%s,\t User Name: %s ,\t Time:%s \n", 
+$sessionid, $UserID, $TStamp); $l_Counter++; }  


close ACCESS;

Now the problem I am facing is that how do I push session on to the hash so that I can extract it later in order to populate the result. Appreciate your help.

[rupesh]

use strict; 
my %hash;

foreach(<DATA>)
{
    $hash{$1}=() if /JSESSIONID=(.*)$/;
}
foreach (sort keys %hash) { print $_."\n"; }

__DATA__

[17/May/2005:18:15:59 -0700]sdjhsj USERID=t70xvcy; jdhdh JSESSIONID=00
+00RDnYTABcRWOx0UN2Zq--sZB:10f408hdg
[17/May/2005:18:15:59 -0700]sdjhsj USERID=t70xvcy; jdhdh JSESSIONID=00
+00RDnYTABcRWOx0UN2Zq--sZB:10f408hdg
[17/May/2005:18:15:59 -0700]sdjhsj USERID=t70xvcy; jdhdh JSESSIONID=00
+00RDnYTABcRWOx0UN2Zq--sZB:10f408g13

[bart]

I am amazed that

$hash{$1}=()

works without warning, on 5.6.1 and on 5.8.4, as does

my $x = ();

Can anybody explain, are my expectations so off?

[jasonk]

From perldata:

If you evaluate an array in scalar context, it returns the length
of the array. (Note that this is not true of lists, which return the last
value, like the C comma operator, nor of built-in functions, which
return whatever they feel like returning.)

An empty list doesn't have any values, so trying to get the last value gives you undef.

[jdporter]

So you want to report the user, session id, and timestamp of each record grouped by user? Could be tricky, since even in your very brief sample data, none of those fields is uniquely occurring. So, let's try listing by user, primarily, and by time, secondarily. Any complete duplicates are not shown; just one instance will be printed. (However, the count is known, if you decide to print that too.)

my %User_Lst;

open ACCESS,"< $FileName" or die "Can't open $FileName for read: $!\n"
+;
while (<ACCESS>)
{
    /:(..:..:..).*USERID=([a-z0-9].+); .*JSESSIONID=[a-z0-9].*?:(.+)/
        and $User_Lst{$2}{$1}{$3}++;
}
close ACCESS;

for my $UserID ( sort keys %User_Lst )
{
    for my $TStamp ( sort keys %{ $User_Lst{$UserID} } )
    {
        for my $session ( sort keys %{ $User_Lst{$UserID}{$TStamp} } )
        {
            printf "Server Name :%s,\t User Name: %s ,\t Time:%s \n",
                $session, $UserID, $TStamp;
        }
    }
}