Data Comparison in 2 csv files

Question

I have 3 files:

In file1 I've data like this:

id,name,age
1,jj,60
2,kk,45
3,ss,56

In file2 I've data like this:

id,name,age
1,jj,60
2,kk,44
3,ss,55
4,tt,66

In file3 I've only one field i.e. id:

id
1
2
3

Now I want to compare file1 & file2 for those ID which are in file3.

Output should be like this:

id  file1   file2
2   age=45  age=44
3   age=56  age=55

Answer 1

Extended paste + awk solution:

awk -F',' -v ids=$(paste -s -d'|' <(tail -n+2 file3)) \
'BEGIN{ OFS="\t"; print "id", ARGV[1], ARGV[2] }
 FNR==1{ next }
 NR==FNR{ if ($1 ~ "^("ids")$") a[$1$2]=$3; next }
 ($1$2 in a) && a[$1$2] != $3{ 
     printf "%d\tage=%d\tage=%d\n", $1, a[$1$2], $3 
 }' file1 file2

The output:

id  file1   file2
2   age=45  age=44
3   age=56  age=55

Answer 2

Roman's answer is good and gives you the exact output you asked for.

If, however, you only needed the raw data (i.e. without CSV headers and without the printf pretty-printed formatting), you could use paste + comm + grep + process substitution:

$ paste <(comm -1 -3 file1 file2) <(comm -2 -3 file1 file2) | 
    grep -f <(sed -e 's/^/^/; s/$/,/;' file3)
2,kk,44 2,kk,45
3,ss,55 3,ss,56

Explanation:

comm -1 -3 file1 file2 outputs the lines that are unique to file2. comm -2 -3 file1 file2 outputs the lines unique to file1. The output of both these commands are given to paste via process substitution, and joined together.

paste's output is then piped into grep to output only the lines that match the ids listed in file3. Process substitution is also used here to transform the IDs listed in file3 into anchored regular expressions that match the ID numbers ONLY when they are at the start of a line and followed by a comma.

Doing something like this would mostly be useful if you wanted/needed to do further processing before pretty-printing the output.