CGI perl reading input from webform in order to split

[neemie]

I'm trying to split a $string that is inputted into a webform made using CGI.pm

The split is on '>' and I have it working from the command line using shift to get the file into the script however using the webform doesn't work - i.e. I get no output. This works from the command line using shift to get the read in the file as follows:

my $inFile = shift; 
open (IN, "$inFile");
$/ = ">";
while ( my $record = <IN> ) {
chomp $record; 
my ($defLine, @seqLines) = split /\n/, $record;
my $sequence = join('',@seqLines);

however, using the code below in a cgi script does not work - I guess the cgi script is forcing a $string? but I'm not sure how to proceed

use CGI qw(:cgi-lib :standard);
print "Content-type: text/html\n\n";
my $seq = param('sequence');
$/ = ">";
chomp $seq; 
my ($defLine, @seqLines) = split /\n/, $seq;

any advice, greatly appreciated

[GrandFather]

Don't guess. Add a print statement to your code and see what you are actually getting. '>' is special in HTML and is generally encoded as >. Maybe you are getting an encoded string where you expect an decoded string?

[CountZero]

Are you sure that $seq contains anything?

Why don't you do a split />/, $seq directly? Globally changing $/ is not advised as it can have unexpected side-effects.

[Anonymous Monk]

#!/usr/bin/perl --
use strict;
use warnings;
use CGI::Carp qw(fatalsToBrowser);
use CGI qw/ param escapeHTML header /;

Main( @ARGV );
exit( 0 );

sub Main {
   my( $headers, $body ) = SequencePage();
   print $headers, $body;
}
sub SequencePage {
   my( $defLine, @seqLines ) = split />/, param('sequence');
   my $body = escapeHTML( $defLine );
   $body .= "<br>".escapeHTML( $_ ) for @seqLines;
   return header(), $body;
}

[wazat]

I've never been a heavy CGI.pm user, but one question that comes to mind is if 'sequence' is actually a file upload instead of a normal form parameter. If so, I think you need to treat param('sequence') as a filehandle.