2. More On Arrays #

Created Tuesday 12 November 2019

• We do and will only use an integer literal as the array length. So no variables allowed. Reason will be discussed later.
• Use a big number like 100, I know this wastes space. We will deal with this in dynamic memory allocation.

Q) Find the maximum integer in an integer array of size n. A) Why not? If we take A[0] as optimum, two possibilities arise for the value of ‘max’:

1. It is the optimum. Hence, nothing will be greater or lesses than it (< or >) conditions for greatest and least respectively.[Note we don’t need to update if duplicate optimums are present, no equality in comparisons].
2. It is not the optimum, here. The value will be updated to a relative optimum of the comparison, but if we hit an optimum, possibility no.2 becomes true, hence no change will occur.
3. But what if n = 0, then, we get max as A[0] which is not even there. In such a case, i.e if we have a function called max, it has to return some integer, but it has nothing to work with. We are stuck. So for consistency, we initialize max as INT_MIN (in ) or INT32_MIN(). So it returns a hugely negative integer in the case of n=0. For The general case, this is simply case 2(or rarely case 1, i.e if all are really INT_MIN). Which eventually becomes case 2. Problem Solved.

• INT_MIN or INT_MAX was only for a formality of consistency for the return value. It is a convention. **It is better to use INT_MIN or MAX for no confusion at all. **
• In the method max = A[0], we had to check if n>0 first.

Algorithmically(Assuming n>0):