use awk to left outer join two csv file based on multiple columns while keeping order of the first file observations - linux

Question

I have two csv files.
File 1
ID,Name,Gender,Salary,DOB
11,Jim,M,200,90
12,David,M,100,89
12,David,M,300,89
13,Lucy,F,150,86
14,Lily,F,200,85
13,Lucy,F,100,86
File 2
DOB,Name,Children
90,Jim,2
88,Michael,4
88,Lily,1
85,Lily,0
What I want to do is to left outer join File 2 into File 1 based on DOB and Name while keeping the order of File 1 observations.
So the output is expected to be
ID,Name,Gender,Salary,DOB,Children
11,Jim,M,200,90,2
12,David,M,100,89,
12,David,M,300,89,
13,Lucy,F,150,86,
14,Lily,F,200,85,0
13,Lucy,F,100,86,
I learned that we need to sort data if we use join command. So I was wondering whether I could use awk to do my work. But I am new with awk. Is there anyone can help me? By the way, if the data is very big, can I drop print command in awk but simply use > *.csv to save into a new csv file? It's because I found solutions to some related questions in this website often used {print ...}. Thank you.

Answer 1

awk to the rescue!
$ awk -F, 'NR==FNR{a[$1,$2]=$3; next} {print $0 FS a[$NF,$2]}' file2 file1
ID,Name,Gender,Salary,DOB,Children
11,Jim,M,200,90,2
12,David,M,100,89,
12,David,M,300,89,
13,Lucy,F,150,86,
14,Lily,F,200,85,0
13,Lucy,F,100,86,
join will require sorted input and you need embellishments to recover initial ordering. You can redirect the output to a file by adding > outputfile.csv