difficult error checking

Anonymous Monk has asked for the wisdom of the Perl Monks concerning the following question:

I don't know what to call this, but I'm confused! I have a form that uploads images (upload#). Since I have the option of uploading more than 1 file at a time, I pass the work to a subroutine with an increment on $num.

Everything works fine except the new error checking I tried to make. Since each image filename is stored into a hash, I need them to be different. I tried using a foreach(keys.. to check, and this DOES work...kind of.

The problem is, if upload1 and upload2 are unique, but upload3 isn't, images 1 and 2 will be sucessful and image 3 will die. Likewise, if image 1 is a duplicate and 2 and 3 are different, because image1 dies, the rest can't be successfull.

Can someone help me figure out a way to set it up to print "$filename was uploaded" for ALL images that are unique, even if the image before it was a duplicate?

An example:

___DATA___
dog.gif
cat.gif
bird.gif

Upload form attempts:
upload1 = bat.gif
upload2 = dog.gif
upload3 = gopher.gif

Print out results:
bat.gif was successfull
dog.gif was a duplicate, not uploaded
gopher.gif was successfull
[download]

SCRIPT:

if ( param('upload1') ) {

my $num = 1;

&dirty_work($num);

}

if ( param('upload2') ) {

my $num = 2;

&dirty_work($num);
}

if ( param('upload3') ) {

my $num = 3;

&dirty_work($num);
}

if ( param('upload4') ) {

my $num = 4;
&dirty_work($num);
}


sub dirty_work {
my $num = shift;
    # take form data
    my $remotefile = param("upload$num");
# make new variable to prevent overwriting of form data
    my $filename = $remotefile;
    
    my $title = param("title$num");
    $title =~ s/&/&\#38/g;
    my $comments = param("comments$num");
    $comments =~ s/&/&\#38/g;
    # remove all directories in the file name path
    $filename =~ s/^.*[\\\/]//;

    my $localfile = "$uploaddir/$filename";
    my $type = uploadInfo($remotefile)->{'Content-Type'};
    unless ( $type eq 'image/pjpeg' || $type eq 'image/gif' || $type e
+q 'image/bmp') {
        print "Wrong!  This is not a supported file type.";
        exit;
}

##################################
# Check for duplicate filenames
##################################
foreach (keys %upload)
{
  my ($filename_check, $title_check, $comments_check, $w_check, $h_che
+ck, $count_check) = split(/::/, $upload{$_});
  if($filename eq "$filename_check") {
    print "<center><font color=red>Duplicate filename found: $filename
+_check</font></center>";
    exit;
  }
}
 
print "<font color=blue><center>Uploaded $filename</font></center>";
[download]

Comment on difficult error checking Select or Download Code

Replies are listed 'Best First'.
Re: difficult error checking by Sol-Invictus (Scribe) on Jan 28, 2004 at 21:48 UTC
use a hash to filter doubles `#! perl -w use strict; my($file); my(%upload,%seen); %upload=(a=> 'dog', b=> 'cat', c=> 'llama', d => 'dog' ); for (keys %upload){ $file=$upload{$_}; if(defined $seen{$file}) { print "<center><font color=red>Duplicate filename found: $ +file</font></center>\n"; exit; } $seen{$file}=$file; print "<font color=blue><center>Uploaded $file</font></center>"; + }` [download] Instead of the exit call you could just make the filename more unique by tagging on a number (much like your OS does when you make try to make a file with a non unique name) `#! perl -w use strict; my(%upload,%seen); my($counter,$file,$old); %upload=(a=> 'dog.gif', b=> 'cat.gif', c=> 'llama.gif', d => 'dog.gif' ); for (keys %upload){ $file=$upload{$_}; if(defined $seen{$file}) { $old=$file; while (defined $seen{$file}){ $file=$old; $counter++; $file=join('.',$file,$counter); } } $seen{$file}=$file; print "<font color=blue><center>Uploaded ",$file,"</font></cente +r>\n"; }` [download]	[reply] [d/l] [select]
Re: difficult error checking by fireartist (Chaplain) on Jan 29, 2004 at 09:47 UTC
It may not be suitable for your application, but I normally assign my own filename to any uploaded file. Something like, `sub new_filename { my $filename = shift; my $ext; my $newname; if ($filename =~ /.*(\.[a-zA-Z0-9]{2,4})$/) { $ext = $1; } else { return; } do { $newname = rand_string() . $ext; } while (-e $newname); return $newname; } sub rand_string { my @char = (0..9, 'a'..'z', 'A'..'Z'); my $string = ''; for (1 .. 15) { $string .= $char[rand 64]; } return $string; }` [download]	[reply] [d/l]


good chemistry is complicated, and a little bit messy -LW
	PerlMonks