Fixed with string manipulation with sed or awk ?

[Exie]

Hi,

I've got a fixed with data file, and I need to alter 1 field before it gets processed.

The last field has ( and ) and _ in it, and we need to take them out.

I started with

echo "word1word2wo(r)d3_  " | tr -d '()_'

This has 2 problems (apart from been somewhat risky if it affects other fields). Firstly this altered my record length and secondly I actually want to change to size of the last field from 10 to 8

how can I get the output to be:
"word1word2word3 "

I've tried SED and AWK, but feel like my head is going to explode.

 

[Annihilannic]

Something like this perhaps, although it poses the same risk to the other fields:

echo "word1word2wo(r)d3_  " | awk '{gsub("[()_]","");printf "%-18s\n",$0}'

To avoid that risk you might need to separate that field from the rest first before performing the replacement, but the printf() usage should be useful for you.

Annihilannic.