mjhjmtu

A sample input data with 3 columns, 1st and 3rd column has values in duplicates, need to print them uniquely with their repetition count.

sort -u does help in some sense but unable to print occurrence of repeated values relating 1st and 3rd column.

Input :

A 3210 -06:00
A 5172 -06:00
A 3335 -07:00
A 3258 -05:00
B 3322 -05:00
B 5097 -05:00
C 3238 -06:00
C 5364 -05:00
C 3366 -06:00
C 3293 -06:00

Output :

A(2) -06:00 
A(1) -07:00
A(1) -05:00
B(2) -05:00
C(3) -06:00
C(1) -05:00

or

Output :

A 2 -06:00 
A 1 -07:00
A 1 -05:00
B 2 -05:00
C 3 -06:00
C 1 -05:00

1. 
   

   Given Input use cut, sort, uniq and sed:

   
   
   

   cut -d ' ' -f1,3 Input | 
   sort | uniq -c | 
   sed 's/^ *//;s/^([0-9]*) ([^ ]*)/2 1/'
   

   
   


2. 
   

   Using datamash and sed:

   
   
   

   datamash -t ' ' -g1,3 -s countunique 2 < Input | 
   sed 's/(.*) (.*) (.*)/1 3 2/'
   

   
   

Output of either:

A 1 -05:00
A 2 -06:00
A 1 -07:00
B 2 -05:00
C 1 -05:00
C 3 -06:00

Not precisely the format you want, but fits all other requirements:

awk 'print $1" "$3' <inFile> | sort | uniq -c

In english, use awk to print only the first and third columns, then sort, then uniq with count.

$ awk ' count[$1,$3]++ END for (i in count) split(i, field, SUBSEP); printf("%s(%d)%s%sn", field[1], count[i], OFS, field[2]) ' file
A(1) -07:00
B(2) -05:00
A(2) -06:00
A(1) -05:00
C(3) -06:00
C(1) -05:00

Note that the output may not be sorted. Pass it through sort if needed.

The code stores the count for how many times the first and third fields of the input has occurred together as a pair, in the count array (with the first and third fields as the index). At the end, we loop over the indexes of the array, splitting them up into the original first and third fields (as field[1] and field[2] respectively) and output these together with the count in the wanted format.

In the alternative format:

If the input file uses a single space for field separator (otherwise use awk ' print $1,$3 ' instead of the cut):

$ cut -d ' ' -f 1,3 file | sort | uniq -c
 1 A -05:00
 2 A -06:00
 1 A -07:00
 2 B -05:00
 1 C -05:00
 3 C -06:00

To swap the two first columns:

$ cut -d ' ' -f 1,3 file | sort | uniq -c | awk ' print $2, $1, $3 '
A 1 -05:00
A 2 -06:00
A 1 -07:00
B 2 -05:00
C 1 -05:00
C 3 -06:00