can sed remove every line that contains 'foo'?

Question

I read somewhere that:

sed -n -e '/foo/d' myinputfile.txt

would remove all occurences of 'foo' from myinputfile.txt.

However this does not seem to work for me. I am a sed noob and cannot seem to work this out. I am basically trying to run a bash script that calls sed on each line to remove a word from the input file and nothing happens when I run it.

Thanks :)

Answer 1

You read incorrectly.

However, while the sed expression itself is correct, the flags are not. sed normally outputs each line as it processes it to stdout, but -n suppresses this. The end result is that no lines are output. You must remove the -n if you want the proper output. You can then redirect this into another file, and then move that file into place.