[Oberon] Compiler behaviour for a32 := a8

[August Karlstrom]

On 2021-02-02 00:31, Chris Burrows wrote:
> A message I posted here last year is somewhat relevant:
> 
> http://lists.inf.ethz.ch/pipermail/oberon/2020/014095.html

Niklaus Wirth says that

"It is the privilege of a report to leave certain constructs undefined. 
In this case a user should simply avoid using the (undefined) feature."

To me this makes perfect sense for the "gray areas." However, I didn't 
know it also applies to the rules of type compatibility. The conclusion 
then is that although the PO compiler supports structural equivalence of 
arrays and procedures, and access to intermediate constants, types and 
procedures we shouldn't use these features.


-- August