Question

My question might not be quite clear… so lets get to samples datas…
fileA
162,400
174,402
175,404
187,406

fileB
E-J1-N5,319319,240248,364364,162162
F-J6-N1,380380,250250,360360,162174
E-J6-N2,380380,240240,364374,175187

Wanted output
E-J1-N5,319319,240248,364364,400400
F-J6-N1,380380,250250,360360,400402
E-J6-N2,380380,240240,364374,404406

So I need to substitute all occurrences of field1 (fileA) by field2 (fileA) but in another file, fileB (field $5).

Here is what I though could work. Store in two variables: recordx (field1) and recordx (field2). My NOT working code:

#!/bin/bash
for record in fileA
   do
        field1=`cut -d, -f 1 $record`
        field2=`cut -d, -f 2 $record`
        awk -v var1=$field1 -v var2=$field2 -F, '{sub(/var1/,"var2",$5);print}' fileB

   done

I get an error :
awk: cannot open 175 (No such file or directory)

Why is awk trying to open a variable ?
Am I heading in a good direction ?

Any help would be greatly appreciated !

Answer 1

awk to the rescue

awk 'NR==FNR{
       a[$1]=$2
       next
     }
     {
       match($5,/(...)(...)/,b)
       $5=a[b[1]] a[b[2]]
     }1' FS="," OFS="," fileA fileB

Output

$ awk 'NR==FNR{a[$1]=$2;next}{match($5,/(...)(...)/,b);$5=a[b[1]] a[b[2]]}1' FS="," OFS="," fileA >
E-J1-N5,319319,240248,364364,400400
F-J6-N1,380380,250250,360360,400402
E-J6-N2,380380,240240,364374,404406