occurences

[Murugs]

I have a text file for column a and b which looks like this

a b
18772 2
18324 4
18009 2
18772 3
18772 3
18772 2
18009 2
18772 3
18009 3
17654 2
18772 3

I need the output
a b c
18772 2 2
18772 3 4
18009 2 2
18009 3 1
18324 4 1
17654 2 1

where c means
no of occurences for value a against value b.

Regards
MP

 

[vgersh99]

BEGIN {
FS=OFS=" ";
}

{
arr[$1,$2]++;
}

END {
for (i in arr) {
split(i, ind, SUBSEP);
printf("%s %s %s %s %s\n", ind[1], OFS, ind[2], OFS, arr);
}
}
vlad
+---------------------------+
|#include<disclaimer.h> |
+---------------------------+

 

[Murugs]

Hello vlad
BIG thanks for ur script.
It did the job..
But if possible shall you explain the script line by line.

Thanks
MP

 

[vgersh99]

BEGIN {
FS=OFS=&quot; &quot;;
}

{
#
# make the FIRST and the SECOND columns indecies to the
# associative array &quot;arr&quot; - the 'arr' is a two-dimentional
# array. Increment [++] the value of the array's cell
# [indexed by &quot;$1,$2&quot;].
#
arr[$1,$2]++;
}

END {
#
# iterate through array 'arr' - current index is 'i'
for (i in arr) {
# split the current index 'i' into an array 'ind'.
# the FIRST cell of 'ind' is used to be column ONE
# the SECOND cell of 'ind' is used to be column TWO

split(i, ind, SUBSEP);
# The value of arr is an accumulated count of
# occurances.
# Print out the statistics.
printf(&quot;%s %s %s %s %s\n&quot;, ind[1], OFS, ind[2], OFS, arr);
}
} vlad
+---------------------------+
|#include<disclaimer.h> |
+---------------------------+