I have a text file which contains like below:
{"userId":"f1fcab","count":"3","type":"Stack"}
{"userId":"fcab","count":"2","type":"Stack"}
{"userId":"abcd","count":"5","type":"Stack"}
I want to get sum of the value of count.
I am using awk to achive this like below:
$ awk -F "," '{print $4}' test.txt
How can I get only the integer type using awk and add them all.
My script should give me as
sum=10
You could try the below,
$ awk -F'"' '{sum = sum + $8;}END{print "sum="sum+0}' file
sum=10
-F'"' Sets the double quotes as FS value. Awk splits the row into colunms according to the value of FS variable.sum = sum + $8 Calculate the sum of all the values in column no 8 and store it into a variable called sumsum at the end will give you the desired output.
sum+=$8 instead of sum = sum + $8. Also, it is sometimes handy to print sum + 0, so that if sum wasn't set, you'll get a 0.+0 solves is when the sum wasn't executed at all. That is, awk '{sum+=$2} END {print sum}' file will fail if the file is empty; also, if the sum is within a condition that is never matched.You can get the value of count key using double quotes (") as delimiter so that the eighth column will be the value to count on:
$ awk -F"\"" 'BEGIN {sum=0} {sum+=$8} END {print sum}' fd
10
Assuming consistent use of double quote characters, you can use:
awk -F\" '{s += $8} END{print "sum=" s+0}' inputFile
This will generate:
sum=10
This works because a quote delimiter gives you the fields:
1 2 3 4 5 6 7 8 ...
{"userId":"f1fcab","count":"3","type":"Stack"}
awk -F'[:"]' '{sum+=$10} END{print "sum=" sum}' File
Setting ':' and '"' as delimiters. Then taking the 10th field, which is the count value. add then up to sum and print at the end.
Example:
sdlcb@ubuntu:~/AMD_C/SO$ cat File
{"userId":"f1fcab","count":"3","type":"Stack"}
{"userId":"fcab","count":"2","type":"Stack"}
{"userId":"abcd","count":"5","type":"Stack"}
sdlcb@ubuntu:~/AMD_C/SO$ awk -F'[:"]' '{sum+=$10} END{print "sum=" sum}' File
sum=10
awk 'match($0,"count\":\"([^\"])",a){x+=a[1]}END{print "sum="x}' file