I want to calculate the sum of a specific column using bash without using the print that specific column (I want to keep all the output columns of my pipeline and only sum one of them!)
1
-
5Show example, a few lines of input and desired output.grebneke– grebneke2014-01-24 20:41:39 +00:00Commented Jan 24, 2014 at 20:41
Add a comment
|
4 Answers
If you wanted to sum over, say, the second column, but print all columns in some pipeline:
cat data | awk '{sum+=$2 ; print $0} END{print "sum=",sum}'
If the file data looks like:
1 2 3
4 5 6
7 8 9
Then the output would be:
1 2 3
4 5 6
7 8 9
sum= 15
Comments
Do you want to continuously sum one column, step by step?
Does is have to be bash or can you use awk:
# file 'fields.txt':
1 foo
2 bar
10 baz
8 boz
# Step by step sum the first column:
awk '{s+=$1; print s, $2}' < fields.txt
# Output:
1 foo
3 bar
13 baz
21 boz
Comments
Assuming the same input data as @John1024, you can use good ol' cut and paste and some bash arithmetic:
$ cat data | echo $(( $( cut -d' ' -f2 | paste -s -d+ - ) ))
15
$
The trick here is to tell paste to insert + as a delimiter, then perform bash arithmetic using $(( )) on the resulting expression.
Note I am just cating the input data for illustrative purposes - it could be piped from another source, or the data file passed directly to cut as well.
Comments
awk '{sum+=$1;} END { print "Total of 1st Column:" sum }1' abc.t6RrMm
Given a file like:
12 12 12
1 1 1
2 2 1
0 1 2
Total of 1st Column is 15.
