match patterns update output file uncomment when desired

Question

Need suggestion for following. Have two files myfile and responsefile.

First file

myfile.txt  

user=myname   
user_1=yourname   
group=mygroup   
group_1=yourgroup

second file

responsefile.txt

#Please fill details  
user=  
#user_1=  
#user_2=  

#Please fill details  
group=  
#group_1=  
#group_2=  

Based on myfile.txt data update responsefile.txt as below and the file responsefile.txt is lenghty of about 604L, 16481C.

Result output

responsefile.txt

#Please fill details  
user=myname  
user_1=yourname  
#user_2=  

#Please fill details  
group=mygroup  
group_1=yourgroup  
#group_2=  

If you observe myfile above, I want to match user= in responsefile, then update as user=myname, same applies for group=. Then match user_1= and group_1= which is hashed or commented in responsefile, update as user_1=yourname and group_1=yourgroup. Should not remove hash or uncomment for others in file.

I tried this

awk -F= 'NR==FNR{a[$1]=$0;next}$1 in a{$0=a[$1]}1' myfile.txt responsefile.txt

Please suggest thanks in advance.

Answer 1

Here is a shell + sed script which will pick the variable names and values (in front of the loop) and then edit them into the file (in multiple passes inside the loop):

#!/bin/sh

sed -nr 's/^([^=#][^=]*)=(.*)$/\1\n\2/p' "$1" |
    while read var ; read val ; do
        sed -ri "s/^#?($var=) *$/\1$val/" "$2"
    done

First parameter of the script is the file with values (myfile.txt). Second is the file to fill (responsefile.txt). The second file is edited in place. The script does not treat white space specially (except spaces after = in responsefile.txt) as it was not clear in the question how it should be.

Here is extension of your AWK script completed to be functional in all cases presented in your demo files. First it will fill the associative array from the first file then it will modify the content of the second file and send it to stdout. I am not presenting it as a one-liner to keep it readable:

#!/usr/bin/awk -f

BEGIN   { FS="=" }
NR==FNR { a[$1]=$0; next }
$1 in a { $0=a[$1] }
/^#/    { var=$1; sub(/^#/, "", var); if(var in a) { $0=a[var] } }
1

Your script just needed a special treatment for commented assignments. The parameters of the script are the same as in the first script but the output is sent to stdout and the second file is left unmodified.