Uploading and reading file

[Anonymous Monk]

Hi, I'm trying to upload a file, and then process the contents of the file. Mostly this works fine, but in some cases, the results on the page are incorrect. It prints the processed filename correctly, but the processed contents are the contents from a previously uploaded file.

Any ideas where it goes wrong ?

Thanks! Rianne

use CGI qw/:standard -private_tempfiles/;
use Text::CSV;
use Fcntl qw( :DEFAULT :flock );

use constant UPLOAD_DIR     => "/tmp";
use constant BUFFER_SIZE    => 16_384;
use constant MAX_FILE_SIZE  => 50*1_048_576;       # Limit each upload
+ to 50 MB
use constant MAX_DIR_SIZE   => 100 * 1_048_576; # Limit total uploads 
+to 100 MB
use constant MAX_OPEN_TRIES => 100;

$CGI::DISABLE_UPLOADS   = 0;
$CGI::POST_MAX          = MAX_FILE_SIZE;

my $q = new CGI;
$q->cgi_error and error( $q, "Error transferring file: " . $q->cgi_err
+or );

#Some other stuff here

print $q->header( "text/html", -expires => "-1s"  ),                  
+                        
      $q->start_html( -title => "Wordcloud Generator", -bgcolor => "#f
+fffff", -head=>$q->meta({-http_equiv => 'Pragma',
                                          -content    => 'no-cache'}) 
+),
      $q->h2( "Wordcloud Generator" ),
      $q->hr,
      $q->p("Processed file:", $file);
my $in        = $q->upload( "file" );  
my $csv = Text::CSV->new ({ binary => 1, eol => $/, sep_char => "\t" }
+);    
  
     my $header = "1";
      my $numcolumns;
      while(my $line = <$in>)
    {
        if($csv->parse($line))
        {
              my @row = $csv->fields();

#processing the file

[MidLifeXis]

Start with use strict;, use warnings;, and possibly use diagnostics;. You may have just been too aggressive in your trimming code, but I don't see where you declare $file before using it.

See also the section on Embedded newlines -- your line processing is fragile.

It is useful when you ask for help to make it easy for us to replicate your issue. Please include sample input, expected output, actual output, and code that compiles cleanly and illustrates your issue.

[Khen1950fx]

Here's what I found. Make sure that you run it with -T on the cmdline. I left out the csv section to concentrate on CGI.

#!/usr/bin/perl -lT

use strict;
use Fcntl qw/:DEFAULT :flock/;
use CGI::Carp qw/fatalsToBrowser/;
use CGI qw/:standard -private_tempfiles/;

$CGI::POST_MAX = 1024 * 1024 * 50;
$CGI::DISABLE_UPLOADS = 0;

use constant UPLOAD_DIR     => '/tmp';
use constant BUFFER_SIZE    => 16384;
use constant MAX_DIR_SIZE   => 1024 * 1024 * 100;
use constant MAX_OPEN_TRIES => 100;

my $q = new CGI;
print $q->header( "text/html", -expires => "-1s" ),
  $q->start_html(
    -title   => "Wordcloud Generator",
    -bgcolor => "#ffffff",
    -head    => $q->meta(
        {
            -http_equiv => 'Pragma',
            -content    => 'no-cache'
        }
    )
  ),
  $q->h2("Wordcloud Generator"),
  $q->hr,
  $q->p( "Processed file:", my $file ),
  $q->end_html;

my $in = $q->upload('file');
$in = param('upload');
if (not $in && cgi_error()) {
    print $q->header(-status => cgi_error());
    exit 0;
}

[Anonymous Monk]

Thanks for the help. The issue actually had to do with Modperl. Found a solution here: http://perl.apache.org/docs/general/perl_reference/perl_reference.html I changed some of the my declarations to local our, and now it works fine.