open(USEDPW, "> $usedpw") or die "crap on me for writing $!";
flock USEDPW, 2;
[download]

You need to open the file for reading and writing without altering it, get the flock, then proceed. Many correct examples exist. This is not one of them.

-- Randal L. Schwartz, Perl hacker
Be sure to read my standard disclaimer if this is a reply.

I think this is what you are trying to achieve. I've commented each line so that you can understand exactly what's going on there.

#!/usr/bin/perl

use strict;
use warnings;
use CGI;
use CGI::Carp 'fatalsToBrowser';

my $query = CGI->new;
print $query->header;

my $usedpw = "logfile.txt" ;
my @chars = ( "A" .. "Z", "a" .. "z", 0 .. 9, qw(! @ $ % ^ & *) );

# opening for reading
open(USEDPW, "<", $usedpw) or die $!;
# shared lock
flock USEDPW, 1;
# read entire file into array, one line per element
my @usedpw = <USEDPW>;
# close the file
close(USEDPW);

# remove new line chars from passwords
chomp @usedpw;

my $pw;
my $used;

# loop generating passwords until we create one that's unique
do {
    # generate a password
    $pw = join @chars[map{rand @chars} (1..17)];
    # assume it's not used
    $used = 0;
    # compare generated password to each existing password
    for (@usedpw) {
    # if the generated password has been used before,
    # set the used flag and abort the for loop
    $used=1,last if $_ eq $pw;
    }
} while ($used);

# display the password
print "Your unique password is: $pw\n";

# open for append (>>) and save the new password
open(USEDPW, ">>", $usedpw) or die $!;
# exclusive lock
flock USEDPW, 2;
# write new password to end of file
print USEDPW "$pw\n";
# close the file
close(USEDPW);
[download]

---
print map { my ($m)=1<<hex($_)&11?' ':'';
$m.=substr('AHJPacehklnorstu',hex($_),1) }
split //,'2fde0abe76c36c914586c';
[download]

while (<USEDPW>)
    {
        my $pw = @chars[map{rand @chars} (1..17)];
    }
unless ($_ ne my $usedpw) {
        print "Your unique ID is: $pw\n";
    }

my $pw;
open(USEDPW, "> $usedpw") or die "crap on me for writing $!";
flock USEDPW, 2;
print USEDPW "$pw\n";
[download]

#notice we declare it OUTSIDE the loop
my $pw;
while (<USEDPW>)
    {
        $pw = @chars[map{rand @chars} (1..17)];
    }
unless ($_ ne my $usedpw) {
        print "Your unique ID is: $pw\n";
    }

open(USEDPW, "> $usedpw") or die "crap on me for writing $!";
flock USEDPW, 2;
print USEDPW "$pw\n";
[download]

#!/usr/bin/perl

use CGI::Carp 'fatalsToBrowser';
use strict;
use warnings;
use CGI;
my $query = CGI->new;
my $usedpw = "logfile.txt";
my @chars;
print $query->header;

# opening for reading
open(USEDPW, "< $usedpw") or die "poo poo on reading $!";
flock USEDPW, 1;
# Storing the file handle, not a scalar, into an array for full readin
+g
my @used_pw = <USEDPW>;
close(USEDPW);
# remove trailing whitespace
chomp my $used_pw;

while (<USEDPW>)
    {
        my $pw = @chars[map{rand @chars} (1..17)];
    }
unless ($_ ne my $usedpw) {
        print "Your unique ID is: $pw\n";
    }

my $pw;
open(USEDPW, "> $usedpw") or die "crap on me for writing $!";
flock USEDPW, 2;
print USEDPW "$pw\n";
close(USEDPW);
[download]

Almost ... try this instead:

while (<USEDPW>) {
   my $pw = @chars[map{rand @chars} (1..17)];
}
unless ($_ ne $usedpw) {
   print "Your unique ID is: $pw\n";
}
[download]

or this:

while (<USEDPW>)
{
   my $pw = @chars[map{rand @chars} (1..17)];
}
unless ($_ ne $usedpw)
{
   print "Your unique ID is: $pw\n";
}
[download]

Whether you like to hug the open brace (example 1) or put it's on it's own line (example 2) is a matter of personal preference - both are correct. However, you really need to line the closing brace up with what started it. Also, i recommend using tabs instead of spaces - you can always use a tool like expand to convert each tab to any number of spaces. For comments to the left of code, well, i try to avoid them, but when i have to have to have them i use spaces to line them up. Keep up the good work. :)

Oh yeah ... you can simplify the unless block like so:

print "Your unique ID is: $pw\n" if $_ eq $usedpw;
[download]

... and double negatives are bad habits, notice that i replaced unless with if and ne with eq.

jeffa

L-LL-L--L-LL-L--L-LL-L--
-R--R-RR-R--R-RR-R--R-RR
B--B--B--B--B--B--B--B--
H---H---H---H---H---H---
(the triplet paradiddle with high-hat)

Whether you like to hug the open brace (example 1) or put it's on it's own line (example 2) is a matter of personal preference - both are correct. However, you really need to line the closing brace up with what started it.

Lining up the braces with keywords is also a matter of personal style. I do it and I wish every one did. Unfortunately, some people don't. Really, it's all a matter of style and there is no correct or incorrect way to do it. I think the real important thing is consistency.

Also, i recommend using tabs instead of spaces

I strongly disagree with this. Use spaces. Tabs just end up being an additional hassle for the programmer. Tabstops have to be set correctly in order for the code to look good. That's not a huge problem but it is an annoyance. The real problem occurs when a tab-using programmer comes along and modifies source code written by a space-using coder (or vice-versa.)

So, once again the answer is consistency. The reason I advocate spaces is that they are inherently consistent and do not rely on an editor or terminal variable as tabs do. A space is a space is a space whereas a tab is eight spaces or maybe four or perhaps two or possibly three and sometimes six... etc.

I know some people advocate using tabs for indentation and spaces for alignment. In my estimation, that's only half a solution. Worse, it creates its own problems later when the code needs to be maintained.

I'd break my tab key off if I didn't use it for completion in my shell.

-sauoq
"My two cents aren't worth a dime.";