[fpc-pascal] Re: StrUtils.RomanToInt oddities

[Marco van de Voort]

In our previous episode, Reinier Olislagers said:
> >> facto behaviour* (return 0 on invalid values) as it is quite sensible
> >> for this kind of numbers.
> > 
> > It is non-orthogonal.
> What is non-orthogonal? 

RomanToInt uses 0 to signal error, other xxxtoint functions throw
exceptions.

> > Because that has an use. The internal FPC compatability, specially in
> > the more fringe areas like this, service no use than fattening maillist
> > archives IMHO.
> 
> If you don't see a use for backward compatibility for existing code...

I see it only as an starting point, not as a rigid unbendable rule.