RE: [sv-bc] illegal priority if

>From: "Michael (Mac) McNamara" <mcnamara@cadence.com>

>Anyway, where are we going with this? 
>
>   Is there any consensus towards making a modification of the language? 

I haven't heard any practical suggestions yet, so there certainly can't be
consensus.


>   Issuing an interpertation of the language that codifies Brad's understanding 
of the language as the committee's official view?

That depends on how you mean the term "interpretation".

I don't see any other way that the LRM text can be correctly interpreted.
The text may be hard to follow, but if you follow it, there is only one
possible meaning.  I don't believe it is ambiguous.

So this would not be making a decision on something that is not
specified.  It would be explaining the implications of what the
specification says, to anyone having trouble following the logic.


>   Issuing an interpertation that does something codifies something different 
than Brad's view?

I don't think you could call that an interpretation.  That would be a
modification to the language.


>   Leaving this alone for now?

I am inclined to leave this one alone.  It is a crazy corner case, for
which nobody has offered a better solution that is practical.

On the other hand, I think there is a real issue with the text that
allows checking the uniqueness *after* executing the statement for
one of the cases.  That is an incorrect check.  It could produce false
errors in common coding situations.  Perfectly valid state machines
using unique could be declared illegal with this allowed.  And fixing
it is just a matter of modifying the text to require checking all the
conditions before executing the selected statement.  This is not
difficult.

Steven Sharp
sharp@cadence.com