CODEWITHSUNDEEP
linuxawk
Muhammad
Muhammad

Reputation: 631

how to combine contents of two files based on key columns in linux

I've two files with 3 identical columns and 4th a different one.

File A

a b c 100
e f g 50
h i j 25

File B

a b c 200
e f g 20
h i j 15

How can files A and B be combined to look like file C?

File C

a b c 100 200
e f g 50 20
h i j 25 15

--UPDATE--

I've used the solutions provided by Jotne an Kent but both of the script's output have . (dot) instead of comma. it looks like

  a,b,c,100.200 
  e,f,g,50.20

Upvotes: 0

Views: 91

Answers (2)

fedorqui
fedorqui

Reputation: 289535

If they had just a column in common, join could make it. But let's use it and then parse the output:

$ join <(sort f1) <(sort f2)
a b c 100 b c 200
e f g 50 f g 20
h i j 25 i j 15

This joined based on the first column. Now, let's use cut to get everything but columns 5 and 6:

$ join <(sort f1) <(sort f2) | cut -d' ' -f1-4,7
a b c 100 200
e f g 50 20
h i j 25 15

Note the usage of sort to sort the files, because join needs files to be sorted to work. With the sample data given it worked without sort, but added for consistency.

Upvotes: 0

Jotne
Jotne

Reputation: 41456

Here is one awk

awk 'FNR==NR {a[$1,$2,$3]=$4;next} {print $0,a[$1,$2,$3]}' B A > C
cat C
a b c 100 200
e f g 50 20
h i j 25 15

Upvotes: 2

Related Questions