Your subject is a bit misleading; you're not converting from 'sed' to Perl - you're converting from a KSH script (which happens to use 'sed') to Perl. As much as I like Perl, and use it for almost every task, for a short script that mostly uses file operations, I'd stick with the standard Unix utils - although I'd do it somewhat differently.

For one thing, since you're only writing your specified string into 'testfile2' - i.e., 'testfile' isn't actually being used in any way - then why not do exactly that? This example copies the action of your script:

#!/usr/bin/ksh

echo 'test ' > testfile2
# echo $(/usr/bin/tput smso)broken$(/usr/bin/tput rmso) > testfile2
[download]

However, since this is a Perl forum:

 
#!/usr/bin/perl -w
use strict;

open F1, "testfile" or die "testfile: $!\n";
open F2, ">testfile2" or die "testfile2: $!\n";

while (<F1>){
    s/.*/test /;
    # s/.*/`tput smso`.'broken'.`tput rmso`/e;
    print F2;
    last;
}
close F1;
close F2;
[download]

-- 
Human history becomes more and more a race between education and catastrophe. -- HG Wells