Re: Efficient Sequential Access to Arrays

[Keean Schupke <keean.schupke@googlemail.com>]

On Thursday, 26 July 2012 13:32:32 UTC+1, Keean Schupke  wrote:
> On Thursday, 26 July 2012 11:10:39 UTC+1, Brian Drummond  wrote:
> > On Thu, 26 Jul 2012 01:38:41 -0700, Keean Schupke wrote:
> > 
> > > This is an equivalent concept to nullable, so two questions:
> > > 
> > > What is the literal for 'invalid', IE:
> > > 
> > > X : Index := #INVALID -- we should be able to initialise to invalid.
> > > 
> > > if X = #INVALID -- I want to be able to test for invalid.
> > 
> > There's an attribute for that.
> > 
> > if X'Valid ...
> > 
> > - Brian
> 
> 
> So if I do
> 
> if X'Valid then
>     Y(X) -- if X is the same type as the index of Y then no need to range check.
> end if
> 
> If I explicitly check 'Valid, the compiler should be able to omit some implicit range checks, as 'Valid is an explicit range check.
> 
> 
> Cheers,
> Keean.


I am wondering if there is a way to use address arithmetic that would be just as safe as array indexing? What about something like:


type Value_Type is limited private;
type Index_Type is range <>;
type Array_Type is array (Index_Type) of Value_Type;

A : Array_Type;
 
type Iterator = Address_Integer range A(A'First)'Address .. A(A'Last)'Address step Value_Type'Size;


This would be safe as it could only contain the address of a valid record in the array, and the syntax could be neat with automatic dereferencing.


Cheers,
Keean.