You can find the line without splitting first:

chomp @file;

my @matches = grep { $file[$_] =~ /^\Q$storeline|\E/ } 0..$#file;
no_match($storeline), return unless @matches;
duped_ids($storeline), return if @matches > 1;

my ($number, $image, $name, $text, $used) = split /\|/, $file[$matches
+[0]];
# do what you do modifying those vars

$file[$matches[0]] = join '|', $number, $image, $name, $text, $used;

rename '/path/to/info/smileset.txt', '/path/to/info/smileset.bak' or d
+ie $!;
open FLE, '>', '/path/to/info/smileset.txt' or die $!;
print FLE $_, $/ for @file;
close FLE or die $!;
[download]

The &no_match and &duped_ids subs are up to you, they would probably log the error.

After Compline,
Zaxo