<FORM ENCTYPE="multipart/form-data" ACTION="the.cgi" METHOD=POST>
Var1 : <input type=text name=var1><br>
File : <INPUT NAME="file" TYPE="file"><p> 
<INPUT TYPE="submit" VALUE="Send File"> 
</FORM>
[download]

#!/usr/bin/perl
use CGI;

#SET SOME DEFAULT STUFF
###################################################
use constant BUFFER_SIZE   => 16_384;      
use constant MAX_FILE_SIZE => 48_576;       
$CGI::DISABLE_UPLOADS      = 0;             
$CGI::POST_MAX             = MAX_FILE_SIZE;                 

#GET THE VARIABLE AND FILE INFORMATION
###################################################
my $query         = new CGI;
my $filehandle    = $query->upload('file');
my $filename      = $query->param('file');
my $var1          = $query->param('var1');

#GET FILE NAME BY REMOVING FULL DIRECTORY INFO
#eg C:\windows\blah\file.txt
##############################################
@pathz      = (split(/\\/,$filename));
$fileb      = $pathz[$#pathz];
@pathza     = (split('/',$fileb));
$filename   = $pathza[$#pathza];



#UPLOAD THE FILE TO SELECTED DESTINATION
##############################################
open OUTPUT, ">/absolute/path/to/$filename" or die "Can't open: $!";
while (<$filehandle>) {
    print OUTPUT;
}
close OUTPUT or die "Can't close: $!";


#DONE
##############################################
print "Content-type: text/html\n\n";
print "Sorted!<p>\n";
print "Passed Variable = $var1<p>";
print "File name uploaded was <i>$filename</i><p>\n";
exit;
[download]

Your suspicion, mentioned in the CB, was that the file's size was causing the problem. That's pretty likely, because you'll notice that CGI::POST_MAX is set to a pretty low number (that value is in *bytes*, by the way ... so as things stand, you'll only allow 48K total in all POSTed data.)

Other non-miscellany : rewrite this script to work under use strict, you'll thank yourself for it later.

There are also some modules that can help you make this script more portable; your "filename parsing" routine in fact only handles Win32-style delimiters => \ , but the standard (i.e. already installed) module File::Basename has robust routines that will handle the delimiters used on most operating systems. You'll need to check the $ENV{USER_AGENT} string to make an educated guess about the operating system the client is using. I do believe KM <-- follow that, by the way ... it's my way of indirectly plugging his book -- has a different strategy for dealing with handling uploads for multiple OSes, but I don't recall off the top of my head what it is.

As far as controlling the error messages, the strategy I like to use is to define a subroutine (I call it bail usually) that will get called in place of die. That way, if something fails I can intercept the error message and give the user something prettier to look at than a 500 error page. If your script is dying *within* calls to functions that are built-in or imported from modules, the usual strategy is to wrap those calls in a block-style eval, and then to check the value of the special variable $@, which will be set to the value of any error thrown by something within the eval block. A toy example:

eval {
    die "You're ugly, and your President dresses funny.";
}
if ($@) {
   print "Hmm, got an error message that read \"$@\" ... I wonder why?
+\n";
}
[download]

Do a super search on "Exception Handling" and read the eval manual page for more info.

HTH

perl -e 'print "How sweet does a rose smell? "; chomp ($n = <STDIN>); 
+$rose = "smells sweet to degree $n"; *other_name = *rose; print "$oth
+er_name\n"'
[download]

thx man, nice scipt and it works