How to count total quantity in a list with multiple occurences

Question

I'm quite new to bash and its syntax, therefore I would like to know how to sort a list to get the total number for an inventory of our hard drives.

Example is below. We have many servers and many hard drives from different vendors and models.

How would you count them together based on the numbers and not the occurrence of their name.

ST8000NC0002: 8
ST900MM0168: 1
ST900MM0168: 1
ST9300603SS: 3
ST9300605SS: 4
ST9500620SS: 3
WD1003FBYX: 7
WD1500HLFS: 4
WD9001BKHG: 3
WD9001BKHG: 3
WD9001BKHG: 4
WD9001BKHG: 4

---

Desired output:

ST8000NC0002: 8 
ST900MM0168: 2 
ST9300603SS: 3 
ST9300605SS: 4 
ST9500620SS: 3 
WD1003FBYX: 7 
WD1500HLFS: 4 
WD9001BKHG: 14

Answer 1

awk is made for this. Populate an array (data) indexed by the first field ($1), accumulating values from the second field ($2). Once the array is populated, output the data using a simple for loop. Pipe the output through sort for convenience.

awk '{data[$1]+=$2} END { for( d in data ) { print d, data[d] } }' input | sort
ST8000NC0002: 8
ST900MM0168: 2
ST9300603SS: 3
ST9300605SS: 4
ST9500620SS: 3
WD1003FBYX: 7
WD1500HLFS: 4
WD9001BKHG: 14

Answer 2

Using GNU datamash:

datamash -W -g 1 sum 2 < file

• -W use whitspace (spaces/tabs) as field delimiter (instead of tabs)
• -g 1 group on the first field
• sum 2 sum values on the second field (per group)

Output:

ST8000NC0002:   8
ST900MM0168:    2
ST9300603SS:    3
ST9300605SS:    4
ST9500620SS:    3
WD1003FBYX:     7
WD1500HLFS:     4
WD9001BKHG:     14

This works fine since your input is already sorted on the first field. For unsorted input, add option -s to sort the input before grouping.