Implication/Exclusion graph for dead features

[FAMILIAR-project]

fml> fm1
fm1: (FEATURE_MODEL) A: [B] [E] [C] [F] ;
!E;
!F;
fml> i1 = computeImplies fm1
i1: (SET) {(C -> A);(E -> B);(B -> A);(E -> A);(F -> B);(E -> C);(E -> F);(F -> C);(F -> A);(F -> E)}

E -> F should not be included !?

fml> e1 = computeExcludes fm1
e1: (SET) {(B -> !F);(A -> !F);(C -> !E);(A -> !E);(B -> !E);(C -> !F)}

E -> !F should be included !?

[FAMILIAR-project]

In FMLDisjunctiveSATMergerTest

test1()

_shell.parse("// with explicity negated variables\n" +
"fm1 = FM (A : [B] [C] [E] [F] ; !E ; !F ; ) \n" +
"fm2 = FM (A : [B] [E] [F] ; B : C ; !E ; )\n" +
"fm3 = FM (A : B [F] [E] [C] ; !F ; !C ; )");

    List<FeatureModelVariable> fmvs = new ArrayList<FeatureModelVariable>() ;
    FeatureModelVariable fm1 = getFMVariable("fm1") ; 
    System.err.println("eg fm1 = " + fm1.computeExcludesEdge());

eg fm1 = [(C -> !E), (B -> !E), (A -> !F), (B -> !F), (A -> !E), (C -> !F)]

why E -> !F does not appear is strange

[FAMILIAR-project]

System.err.println("eg fm1 (bis)" + new SATFMLFormula(fm1).computeExclusionGraph(fm1.features().names()));

eg fm1 (bis)(E -> !F)
(E -> !A)
(E -> !B)
(E -> !C)
(F -> !A)
(F -> !B)
(F -> !C)

so SAT-based is OK