[Oberon] Type compatibility in Oberon

[Chris Burrows]

> -----Original Message-----
> From: Oberon [mailto:oberon-bounces at lists.inf.ethz.ch] On Behalf Of
> Andreas Pirklbauer
> Sent: Friday, 1 May 2020 10:15 PM
> To: Oberon List
> Subject: [Oberon] Type compatibility in Oberon
> 
>    > Not exactly. This amended statement is more accurate:
>    >
>    > "It is not guaranteed that a program THAT USES UNDEFINED
>    > FEATURES compiled with two different implementations of the
>    > same report behave in the same way.
> 
> 
> If you view (just as an example) the maximum case label in the numeric
> case statement as an  undefined feature , I am O.K.
> with the amended statement above.
> 
> So, if compiler A implements the numeric case stmt with max case label =
> 1024 and compiler B with max case label = 255, then the statement
> 
>   CASE k OF 1000: (*do something*) END
> 
> will compile with compiler A, but not with compiler B. Imagine a user who
> has written many legacy programs with label >255 using compiler A and now
> switches to compiler B. Now what?
> 
> Should he HAVE to edit all his legacy programs?
> 

Yes - in a hypothetical case. However, from my experience it is highly unlikely to be a problem in the real world.