Shell Programming and Scripting

I was using the following code successfully on a linux system.

Code:

while read pattern; do
  grep -q $pattern en.pep
  if [ $? -ne 0 ]; then continue; fi
  line_number=`grep -m 1 -n $pattern en.pep | cut -f 1 -d :`
  echo ">$pattern"
  sed -n "$line_number { n; p; q}" en.pep
done < file1 > en_peps

But when I run it on unix I get the following error repeated many times with different line (?) numbers. Needless to say this error is not giving me much of a clue (I'm not much of a programmer), and all forum posts I've found were resolved in specific ways that don't apply to my situation.

Code:

sed: 1: "3163 { n; p; q}": extra characters at the end of q command

If someone could shed some light for me I'd appreciate it!

I think you're missing a semi-colon. Try this:

Code:

sed -n "$line_number { n; p; q;}" en.pep

Many Unix sed have bugs with semicolon - newline works better

Code:

sed -n "$line_number {
n
p
q
}
" en.pep

Thanks, both of these solutions worked.

Those are not a bugs:
sed: rationale

Anyway, the OP's sed apparently does support the semicolon (as most seds do nowadays), but since } is a command, there should be a semicolon in front of it.