To satisfy the mathematician in me, I have to tell you that you're never going to get the "correct" answer, the division produces a number that has an infinite decimal, but I get what you're after, I think. The number you want is:
1 272 750 412 266 999 760.463 917 525 ...
but the number you're getting is:
1 272 750 412 266 999 800
This is due to the lack of precision in the number format used by the language. The loss of precision doesn't occur when you do the division, it happens much sooner than that, as soon as you assign the constant to the variable. The number you want to store is:
123 456 789 989 898 976 765
but the number you're actually storing is:
123 456 789 989 898 980 000
This is what results in the wrong answer.
Since I don't know the Lotus Script environment, I will do two things; first, give you some code that will fix this particular problem, like so:
var num = [12345678998, 9898976765];
var num1 = num[0] / 97;
var num2 = Math.floor(num1);
num2 = num1 - num2;
num2 *= 97;
num2 = Math.round(num2)
num2 *= Math.pow(10, num[1].toString().length);
num2 = (num2 + num[1]) / 97;
alert(Math.floor(num1).toString() + num2.toString());
that you can then generalize to fit your needs. This code splits the division into two smaller divisions that the number storage format CAN handle and adds the remainder of the first division to the second, producing this result:
1 272 750 412 266 999 760.463 917 7
which isn't exact, but probably close enough, right? How you split the bigger number into fragments without loosing precision is left up two you. (Hint: use strings)
Second, I will point you to BigInt.js, a library for doing math with arbitrarily large integers in JavaScript. If you can include this library into your code, it is definitely the more economical way to go.
I hope one of these helps.