Hello Coin

Francis ANDRE francis.andre.kampbell at orange.fr
Wed Nov 20 12:47:26 PST 2013

Hi Coin List

I did not receive any answer on my previous request which means certainly that 
there is no more open call for proposals. Anyway, I will bring my 2 cents to the 
coin distributor!...

I would like Java to offer "true" array/arrays... And you would say: Hey guy, 
what is a "true" array, Java gets already the array concept... Well, let me 
explain a little bit. In java one declares an array of int for example as

int[] myInts;   // myInts, an array of int

and the allocation of the array by

myints = new array[10];   // myints is an array of 10 int
myints[0] = 90;  // the int at index 0 is now 90. The index of an array starts 
at 0 and ends at myints.length - 1
myints[9] = 99;  // the int at index 9 is now 99

This view of an array starting at 0 is more an implementer view that is 
considering the first element of the array is starting at the offset 0 of the 
base of the array....But in mathematics, an array has usually an lower bound and 
an upper bound which are not always 0 and length - 1. Thus from my point of 
view, an "true" array should have a lower bound and an upper bound as in many 
others computer languages (as Algol, PL/1 for example)

Thus, I would propose a more abstract view of an array like

int[] myints;
myints = new array[-100:100];
myints[-100] = 0;
myints[0] = 0;

with obviously all adapted controls and exception like an index out of array 
range and a length = upperbound - lowerbound + 1

myints[500] = 0; // throw an index out of array bound exception
myints.length == upperbound - lowerbound + 1 // this is true

That's all folks!


Le 23/10/2013 08:03, Francis ANDRE a écrit :
> Hi Coin list
> I discovered this discussion list recently -- coin is not an obvious slang for a
> non US guy, so the discovery was more by curiosity than understanding -- and I
> am wondering if there will be another open call for proposals in near future?
> Francis

More information about the coin-dev mailing list