UNIX for Dummies Questions & Answers

Hi all,

I would like to replace some columns from file1 with columns from file2. Currently, I'm able to do it with the following command:


awk 'NR==FNR{a[NR]=$1;b[NR]=$2;c[NR]=$3;next;}
{$2=a[FNR];$4=b[FNR];$5=c[FNR];print}' file2 file1 > temp
mv -f temp file1

First, i make the changes and save it as a temp file. Then, i replace file1 with the temp file.
Is there any way i can write the changes to file1 directly without storing it in a temp file first.
How can i perform the same operation using sed?
Pls advice.
Thanks and Regards.

Example:
--------
File 1:
LAYOUT Label TEXT x1 x2 46 10 FOLLI018ULL
LAYOUT Label TEXT x1 x2 46 10 FOLLI018ULL
LAYOUT Label TEXT x1 x2 46 10 FOLLI018ULL
LAYOUT Label TEXT x1 x2 46 10 FOLLI018ULL
LAYOUT Label TEXT x1 x2 46 10 FOLLI018ULL

File 2:
V1 10 20
V5 100 20
V7 50 20
V15 20 50
V20 50 100

Output File 1:
LAYOUT V1 TEXT 10 20 46 10 FOLLI018ULL
LAYOUT V5 TEXT 100 20 46 10 FOLLI018ULL
LAYOUT V7 TEXT 50 20 46 10 FOLLI018ULL
LAYOUT V15 TEXT 20 50 46 10 FOLLI018ULL
LAYOUT V20 TEXT 50 100 46 10 FOLLI018ULL

I don't see the problem to use a temporary file but you can embed the mv command in your awk command in the END block:

Code:

awk 'NR==FNR{a[NR]=$1;b[NR]=$2;c[NR]=$3;next;}
{$2=a[FNR];$4=b[FNR];$5=c[FNR];print}
END{system("mv temp file1")}' file2 file1 > temp

Regards