Lazy grounding bug for definition

[Broes De Cat]

{ 
    Equiv(m,pid,pid,t).
}

can be true also if arg 2 and 3 differ.

[Broes De Cat]

Another bug:

vocabulary V{ type Value isa int

Knows(Value,Value) Equiv(Value,Value) } theory T:V{ { Knows(v1,v2) <- (Knows(v11,v2) & Equiv(v11,v1)) | (Knows(v1,v21)). Equiv(pid,pid2) <- pid=pid2. } ?x y: Knows(x, y). } structure S:V{ Value = {1..2} } procedure main(tsdelay, satdelay){ stdoptions.groundverbosity=0 stdoptions.satverbosity=20 stdoptions.groundwithbounds=false stdoptions.groundlazily=true stdoptions.tseitindelay=tsdelay stdoptions.satdelay=satdelay print(modelexpandpartial(T, S)) }

has models but shouldn't.

[Broes De Cat]

Removing milestone: Later (automated comment)