CODEWITHSUNDEEP
bashshellawksolaris
nainometer
nainometer

Reputation: 454

How to compare first column of two files and print the difference of second and third column?

I want to compare first column of two different files and if there is a match, put the difference of second and third column of both files into a new file with first column as the matched entry.

Output File = Matched,Difference of c2 and c3 of file1,Difference of c2 and c3 of file 2

Example:

File 1

12,1,3 
13,2,4 
14,5,7

File 2

12,4,5
13,4,7
15,3,9

Desired Output File

12 -2 -1
13 -2 -3

Please tell me how may I do this. File length is varying. File1 is of length 100 and File2 is of length 20

Upvotes: 0

Views: 2002

Answers (4)

Inian
Inian

Reputation: 85895

Use awk which is more suited for this.

awk 'BEGIN{FS=","}FNR==NR{array1[$1]=$2-$3; next}($1 in array1){array2[$1]=$2-$3}END{for (i in array2){print i,array1[i],array2[i]}}' file1 file2
12 -2 -1
13 -2 -3

If awk is not native in Solaris, Can you try nawk as, put the below contents in a file called nawk_script.awk

BEGIN{FS=","}
FNR==NR{array1[$1]=$2-$3; next}($1 in array1){array2[$1]=$2-$3}
END{for (i in array2){print i,array1[i],array2[i]}}

and run it as

nawk -f nawk_script.awk file1 file2

Upvotes: 2

codeforester
codeforester

Reputation: 43109

This will do it - it is pure bash and doesn't fork any sub-processes:

#!/usr/bin/env bash

# the following loop will end when an EOF is encountered on
# either of the two input files, whichever happens first.
while IFS=, read -r m1 m2 m3 <&3 && IFS=, read -r n1 n2 n3 <&4; do
  if [[ $m1 == $n1 ]]; then
    printf "%d %d %d\n" $m1 $((m2 - m3)) $((n2 - n3))
  fi
done 3< file1.txt 4< file2.txt

Output:

12 -2 -1
13 -2 -3

Upvotes: 1

James Brown
James Brown

Reputation: 37464

In awk. If $1 is not yet in array a, subtract and store the value to the array. If it already is, print the key, value in a, and the value of subtraction. Space before 13 was already in your data:

$ awk -F, '!($1 in a){ a[$1]=$2-$3; next } { $3=$2-$3; $2=a[$1] } 1' f1 f2
12 -2 -1
 13 -2 -3

Upvotes: 0

Chem-man17
Chem-man17

Reputation: 1770

One way to do it is using paste and awk-

paste file1 file2 | awk '{gsub(",", " " , $0)} {if ($1 == $4) print $1, $2-$3, $5-$6}'

Output-

12 -2 -1
13 -2 -3

Explanation-

The paste command pastes the files side by side. The awk gsub command changes the commas to spaces. The second brace in the awk checks if the first and fourth columns are the same and if they are, does the required subtraction.

Upvotes: 0

Related Questions