Re: in_range
Gabriel Boehme wrote:
> Why not just define index like this:
>
> type index_value(integer x)
> return x > 0
> end type
Type is fine for debugging, but in a final application I like to leave some
error checks - especially checks like this - in the code, so I can attempt
to gracefully recover.
In addition, the value not being in range isn't always an error. For an
obscure (and bad) example, I use this test to determine if a value is an
index (relatively small) or a handle (typically large). The code ran into
trouble when the handles became so large, the values were negative.
> I don't see the overriding necessity for an in_range() function.
I don't see an overriding necessity, either. But I think that an inclusion
of something like this might help create safer code.
-- David Cuny
|
Not Categorized, Please Help
|
|
