Shell Programming and Scripting

Hi Guys ,

I have two files.

filea:
12:20:23 4
12:21:23 3
12:22:23 2
12:25:23 3

fileb:
11:20:23 4
12:21:23 3
12:22:23 2
12:26:23 3
12:30:23 3

The result will be:
11:20:23 4
12:20:23 4
12:21:23 6
12:22:23 4
12:25:23 3

it means:
1. if the 1st column same in two files, then plus the value in the 2nd value.
2. the date in fileb if greater than the last line in filea, then ignore.

Thanks

use awk

Code:

awk '{A[$1]+=$2}END{for (i in A){print i" "A[i]}}' filea fileb

just one line awk? it works, i do not understand how to read this awk, but it works so nice.

-----Post Update-----

How about i have a file a.txt.

Thu Jun 11 23:59:58 2009 115
Thu Jun 11 23:59:59 2009 115
Fri Jun 12 00:00:00 2009 115
Fri Jun 12 00:00:01 2009 115

i would like to grep those day jun 11 lines to another file.
i can think about using:
cat a.txt | cut -f3 -d' ' | grep '11' | wc -l

But i feel this is not efficient , as my a.txt contains not only 4 lines. any way can do more efficient by using perl, or others?

Thanks

-----Post Update-----


Maybe below for multiple columns:
awk '{A[$1" "$2" "$3" "$4" "$5"\t"]+=$6}END{for (i in A){print i" "A[i]}}' a.txt b.txt | sort

Code:

nawk 'NR==FNR{
        _[$1]=$2
        gsub(/:/,"",$1)
        max=$1
        next
}
{
        tmp=$1
        gsub(/:/,"",tmp)
        if ( tmp <= max )
                _[$1]+=$2
}
END{
        for (i in _)
                print i" "_[i]
}' f1 f2