$_= ~/^\w+\s*=\s*\w+\s*\,\?\w+/ ;
[download]

First thing: if you don't supply a variable, the regex will automatically work on $_, so you shouldn't include that. It's confusing and may lead to bugs. Just put the /regex/.

Second, your regex doesn't actually do anything. It checks the pattern against your chomped line, but it doesn't do anything to it. To manipulate it, you'd need to add another section and make it a s/// operator, with the replacement portion between the second and third delimiters. I don't know what you do want to do to it, so for now, here's what your pattern is checking for. I'm going to use the /x modifier so that I can include whitespace and comments, which I'd suggest you do in the future unless you're dealing with a simple enough regex that it's obvious at a glance what it does.

/^       # anchored to the beginning of the string
 \w+     # one or more word characters (a word)
 \s*     # zero or more whitespace characters
 =       # an equals sign
 \s*     # zero or more whitespace characters
 \w+     # one or more word characters (a word)
 \s*     # zero or more whitespace characters
 \,      # a comma (you actually don't need to backslash this)
 \?      # a question mark
 \w+     # one or more word characters (a word)
    /x;
[download]

As you can see now, this will require that a matching line include a comma followed by a question mark. That's probably not what you want. It also requires a third "word" following the question mark. A good method would be to write out exactly what you want to match, as I've done with my comments above, and then figure out the regex pattern piece by piece from there. Then explain what changes you want to make to it, and we can figure out the second part.

$_= ~/^\w+\s*=\s*\w+\s*\,\?\w+/ ;

Actually, it's worse than that. Because the  = ~ is screwed up (should have been =~), the statement (if that's the statement that's actually in perlynewby's code (if that's actually perlynewby)) is assigning the 1s-complement of the numeric value of the result of the  m// against  $_ to  $_ instead:

c:\@Work\Perl\monks>perl -wMstrict -le
"$_ = 'x=x,?x';
 print qq{'$_'};
 ;;
 $_= ~/^\w+\s*=\s*\w+\s*\,\?\w+/ ;
 print qq{'$_'};
 ;;
 $_ = '----';
 print qq{'$_'};
 ;;
 $_= ~/^\w+\s*=\s*\w+\s*\,\?\w+/ ;
 print qq{'$_'};
"
'x=x,?x'
'4294967294'
'----'
'4294967295'
[download]

---

Give a man a fish:  <%-(-(-(-<

ok, now, I am stuck here. I spend time testing/reading on split with various delimiters.I learned how to remove stuff I don't want...

BUT

now I get some error I DON"T understand after 1.5 weeks of playing with perl. I'm so FRUSTRATED with my prgress!

can you help me out by explaining my error. I am too inexperience to SEE it myself.

first data key has random values of 2 or less numbers

uno = uno,eins
due = dos,zwei 
tre = tres,drei
quattro = quatro
cinque = cinco,funf
sei = seis,
sette = siete,sechs
otto = ocho
nouve = nueve, neun
dieci = diez,   zehn
undici = once, elf
dodici = doce
tredici = trece, dreizehn
[download]

2nd data may have 3 or less vaules

due =deux, two,   
tre = trois,drei,    three 
quattro = quatre,four
cinque = cinq,funf, five
sei = six  , six
sette = sept    ,   seven   ,sechs
dieci =dix
undici = onze,eleven
tredici = treize,thirteen, dreizehn
[download]

use strict;
use warnings;

#declare variables
my %hash;
my $data;
#opening Files
open my $in, '<',"./Test_Data_RanNumbers.txt" or die ("can't open the 
+file:$!\n");
open my $in1,'<',"./Test_Data_More_RanNumbers.txt" or die ("can't open
+ file : $!\n");
open my $out ,'>' ,"./Test_Data_Out_Ita_SpanFrenEngGer.txt" or die "ca
+n't open the file for write:$!\n";
open my $out1,'>',"./Test_data_out_Ita_SpanFren.txt" or die "can't ope
+n the file :$!\n";
open my $out2,'>' , "./Test_Data_Out_None_Match.txt" or die "can't ope
+n file for write:$!\n";

while ($in){
    #data manipulation to clean up ='s and ,'s
    #dieci = diez,   zehn  -->worse case, remove spaces and = and comm
+a;
    #quattro = quatro -->only one number with spaces or not in from of
+ =...
    chomp($data);
    my ($ita,$spa,$num)= split(/[=\s,]+/,$data); # removes '=' or 's' 
+or ',' & '+' to match 1 or more these characters
    $hash{$ita}[0]=$spa;
    $hash{$ita}[1]=$num; #keeping it for check later if "defined" | wi
+ll be undef if not
    }
close $in;

while ($in1){
    chomp($data);
    my ($ita,$fren,$num)= split(/[=\s,]+/,$data); 
    $hash{$ita}[0]=$fren;
    $hash{$ita}[1]=$num; #keeping for check later checks programming
    }
close $in1;

foreach my $ita (keys %hash){
    if($hash{$ita}[0] and $hash{$ita}[1]){
        print "$ita =>", join(',',@{$hash{$ita}}),"\n";
        }else {
            print "$ita =>",join(',',@{$hash{$ita}}),"\n";
            }
    }
[download]

never mind...forget about it ;-)

I am back at it again with clear mind and eyes and I see what I need to do...I think my frustration disadvantage the best of me.

I still like Perl ;-) until next bug in my code...but practicing

Don't feel bad or get too frustrated, just keep plugging away and ask when you need help. Your problem may not seem that complicated, but anything involving multi-dimensional hashes is beyond what I'd consider beginner-level.

When I started learning Perl 20 years ago, it was with a book called something like Learn Perl in 21 Days; and I already had some experience with C, awk, sed, and shell programming; but I probably still wasn't doing multi-level hashes in 1.5 weeks. It seems like people get thrown into the deep end a lot more these days -- maybe due to bosses not knowing what's hard and what's easy and not wanting to give people the proper time for training in the age of outsourcing -- so they don't get to follow a more natural learning curve. You're doing fine.

Aaron B.
Available for small or large Perl jobs and *nix system administration; see my home node.