Find and replace specific entries in awk

[Eric33]

I am newbie writing a awk script that read dynamically assigned variables from shell, and perform the following logic:
1. if Field1=var1 and Field2=var2, extract Field3
(i)if Field3 contains "%1" or "%2"..."%n" (not determinate), replace with corresponding fields from var3 before output Field3
(ii)if no "%n" found, output Field 3

2. if no match, output var3

I written a partial script to match the fields but have difficulty with the replacing the "%n" porttion.

Any help is much appreciated. Thanks in advance! This is what I have so far.

ie

var1=\"young\"
var2=\"girl\"
var3=\"shells/hawiian sea/shore\"

nawk -F, 'BEGIN {Counter = 0} {if (($1 == tolower('"$var1"')) && ($2 == tolower('"$var2"'&quot)) {print $3; Counter = 1}} END {if (Counter ==0) print '"$var3"'} input_file

Input_file:
young,girl,she sells sea %1 by the %2 shore\r\n

Output:
she sells sea shells by the hawiian sea shore\r\n

 

[CaKiwi]

This may do what you want. I did it under DOS so I hard-coded the variables instead of using variable substitution. You could use the -v flag to pass the shell variables

nawk -v v1=$var1 -v v2=$var2 -v v3=$var3 -F,

BEGIN {
if (!v1) v1 = tolower("young&quot
if (!v2) v2 = tolower("girl&quot
if (!v3) v3 = "shells/hawiian sea/shore"
n3 = split(v3,a3,"/&quot
}
{
if (($1 == v1) && ($2 == v2)) {
while (match($3,"%&quot) {
j = RSTART+1
k = match(substr($3,j),/[0-9]*/)
m = substr($3,j,k)+0
sub(/%[0-9]*/,a3[m],$3)
}
print $3
}
else
print v3
}

CaKiwi

"I love mankind, it's people I can't stand" - Linus Van Pelt

 

[Eric33]

Thanks CaKiwi. The replacement algorithm works great! However, I have problem passing the variables using the -v option. Somehow, the variables is not passed in properly. Anyway, it is easily workaround using the in-line substitution.

Thanks!