you have got a lot of help using awk/perl from your previous posts (since 2 years ago), which are strangely similar to this one. you are still doing ldap stuff and formatting the output. What have you learnt this 2 years since 2007? I would expect you to be very familiar with awk/perl already.

Since you aren't getting this, I'll explain it to you. Someone figured out they could write a pretty good computer program by asking endless questions in a newsgroup. It's called "programming by newsgroup" and it takes a long, long time.

Programming by newsgroup (PBN) is sort of like throwing a bunch of proteins into a hot salt water bath, coming back in five years, and seeing what's growing. The most advanced PBN projects use a kind of bot that could almost pass the Turing test, but only if the computer side of the test is allowed to sound totally illiterate -- u knw wht i mn?

So a bot asks a question, someone answers, and the bot inserts the suggestion into a disorganized database of code. Then a genetic algorithm churns for a while, comes up with something new, and the process repeats.

Why do you think there are so many people asking questions who clearly cannot type or think? The answer is obvious: they're -- how shall I say this diplomatically -- synthetic people. As people they're a disaster, but as clone-bots, they're quite impressive.

Some of you may think this is fiction, and today, well, yes, it might just be fiction. But tomorrow? Or a week from now?

This perl version might do you:
#!/usr/bin/perl
my $filename = "a.txt";
open (my $filein, "<", "$filename") or die "can't open $filename";
open (my $fileout, ">b.txt");
my $lc=0;
while (<$filein>) {
my $x=$_;
$x =~ s/\n/ /x;
$line{$lc}=$x;
$lc++;
if ($lc == 3) {
print $fileout $line{$lc-2},$line{$lc-1},$_;
$lc=0;
}
}
# check for unfinished line
if ($lc > 0) {
for (my $x=1;$x<=$lc;$x++) {
print $fileout $line{$x};
}
print "\n";
}

close $filein;
close $fileout;
exit;

Here is a sed script:
In the range from /dn:/ to /created/, the lines are added to the pattern space. On the last line the newlines are replaced with a tab.

This isn't an "improvement", but it does make the sed script produce the "same" output as the original specs
sed '/dn:/,/created/{ /dn/N
/status/N
/created/{s/\n/ /g}
}' test.txt

not too nice!

I'll feed the bot to get a "five minute" lesson on perl/awk that doesn't have a deadline attached.