I have a file where certain (not all) lines look like so:

Sequence: n

I wish to replace n with the respective line number.

How can I do it with sed?

  • By "do it with sed" do you mean "do it with sed" or "do it with any common tool"? Because people say the former and mean the latter about 98% of the time Aug 25, 2011 at 16:48
  • I am specifically curious about sed.
    – mark
    Aug 25, 2011 at 17:30

1 Answer 1


It is very simple with awk (thanks to @glenn jackman):

awk '/Sequence: n/ { $2 = NR } { print }' input-file

Regarding sed, I've only been able to do this with a two-pass elaboration (tested with GNU sed and with ssed:):

sed '/^Sequence: n$/{s/n$//p;=;d;}' | sed '/^Sequence: $/{N;s/\n//;}' 

The first inserts the line number in the row after "Sequence: ", the second joins the two lines.

  • 2
    I'd write your awk command as: awk '/Sequence: n/ {$2 = NR} {print}' Aug 25, 2011 at 21:22
  • @glenn jackman: well done, my bad.
    – enzotib
    Aug 26, 2011 at 7:22

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .