mjhjmtu

I have a variable in ksh that can contain no more then 2 comma-separated digits (white spaces allowed). Something like:

a="1,2"
a="1 ,2" 
a="1,"
a="0,0"

I want a loop to

• read through $a
  


• exit if there is more than one ,


• assign m=a[1] and n=a[2]
  

for a in "1,2" "1 ,2" "1," "0,0" "0,0,0" "2,3"; do
 case "$a" in
 *,*,*) printf 'Too many commas: "%s"n' "$a" >&2
 exit 1
 esac
 IFS=', ' read m n <<<"$a"
 printf 'm = %d, n = %dn' "$m" "$n"
done

This produces

m = 1, n = 2
m = 1, n = 2
m = 1, n = 0
m = 0, n = 0
Too many commas: "0,0,0"

when running under ksh93 (or bash or zsh).

Setting IFS to a space and a comma before read makes read split the contents of $a on these two characters (or multiples of them). The remaining data is assigned to m and n.

If a string is something like "2,3 4", then n would be assigned the value 3 4.

Note also that on the third line of output (parsing "1,"), m is shown as zero only because we're using the %d format specifier with printf. If interpreted as a string, the variable m is empty.