f1(X) -> if3(X, c, n__f1(n__true))
if3(true, X, Y) -> X
if3(false, X, Y) -> activate1(Y)
f1(X) -> n__f1(X)
true -> n__true
activate1(n__f1(X)) -> f1(activate1(X))
activate1(n__true) -> true
activate1(X) -> X
↳ QTRS
↳ DependencyPairsProof
f1(X) -> if3(X, c, n__f1(n__true))
if3(true, X, Y) -> X
if3(false, X, Y) -> activate1(Y)
f1(X) -> n__f1(X)
true -> n__true
activate1(n__f1(X)) -> f1(activate1(X))
activate1(n__true) -> true
activate1(X) -> X
IF3(false, X, Y) -> ACTIVATE1(Y)
ACTIVATE1(n__f1(X)) -> F1(activate1(X))
ACTIVATE1(n__f1(X)) -> ACTIVATE1(X)
F1(X) -> IF3(X, c, n__f1(n__true))
ACTIVATE1(n__true) -> TRUE
f1(X) -> if3(X, c, n__f1(n__true))
if3(true, X, Y) -> X
if3(false, X, Y) -> activate1(Y)
f1(X) -> n__f1(X)
true -> n__true
activate1(n__f1(X)) -> f1(activate1(X))
activate1(n__true) -> true
activate1(X) -> X
↳ QTRS
↳ DependencyPairsProof
↳ QDP
↳ DependencyGraphProof
IF3(false, X, Y) -> ACTIVATE1(Y)
ACTIVATE1(n__f1(X)) -> F1(activate1(X))
ACTIVATE1(n__f1(X)) -> ACTIVATE1(X)
F1(X) -> IF3(X, c, n__f1(n__true))
ACTIVATE1(n__true) -> TRUE
f1(X) -> if3(X, c, n__f1(n__true))
if3(true, X, Y) -> X
if3(false, X, Y) -> activate1(Y)
f1(X) -> n__f1(X)
true -> n__true
activate1(n__f1(X)) -> f1(activate1(X))
activate1(n__true) -> true
activate1(X) -> X
↳ QTRS
↳ DependencyPairsProof
↳ QDP
↳ DependencyGraphProof
↳ QDP
↳ QDPOrderProof
IF3(false, X, Y) -> ACTIVATE1(Y)
ACTIVATE1(n__f1(X)) -> F1(activate1(X))
ACTIVATE1(n__f1(X)) -> ACTIVATE1(X)
F1(X) -> IF3(X, c, n__f1(n__true))
f1(X) -> if3(X, c, n__f1(n__true))
if3(true, X, Y) -> X
if3(false, X, Y) -> activate1(Y)
f1(X) -> n__f1(X)
true -> n__true
activate1(n__f1(X)) -> f1(activate1(X))
activate1(n__true) -> true
activate1(X) -> X
The following pairs can be oriented strictly and are deleted.
The remaining pairs can at least be oriented weakly.
ACTIVATE1(n__f1(X)) -> ACTIVATE1(X)
Used ordering: Polynomial interpretation [21]:
IF3(false, X, Y) -> ACTIVATE1(Y)
ACTIVATE1(n__f1(X)) -> F1(activate1(X))
F1(X) -> IF3(X, c, n__f1(n__true))
POL(ACTIVATE1(x1)) = x1
POL(F1(x1)) = 1
POL(IF3(x1, x2, x3)) = x3
POL(activate1(x1)) = 0
POL(c) = 0
POL(f1(x1)) = 0
POL(false) = 0
POL(if3(x1, x2, x3)) = 0
POL(n__f1(x1)) = 1 + x1
POL(n__true) = 0
POL(true) = 0
↳ QTRS
↳ DependencyPairsProof
↳ QDP
↳ DependencyGraphProof
↳ QDP
↳ QDPOrderProof
↳ QDP
↳ QDPOrderProof
IF3(false, X, Y) -> ACTIVATE1(Y)
ACTIVATE1(n__f1(X)) -> F1(activate1(X))
F1(X) -> IF3(X, c, n__f1(n__true))
f1(X) -> if3(X, c, n__f1(n__true))
if3(true, X, Y) -> X
if3(false, X, Y) -> activate1(Y)
f1(X) -> n__f1(X)
true -> n__true
activate1(n__f1(X)) -> f1(activate1(X))
activate1(n__true) -> true
activate1(X) -> X
The following pairs can be oriented strictly and are deleted.
The remaining pairs can at least be oriented weakly.
ACTIVATE1(n__f1(X)) -> F1(activate1(X))
Used ordering: Polynomial interpretation [21]:
IF3(false, X, Y) -> ACTIVATE1(Y)
F1(X) -> IF3(X, c, n__f1(n__true))
POL(ACTIVATE1(x1)) = 1 + x1
POL(F1(x1)) = x1
POL(IF3(x1, x2, x3)) = x1 + x3
POL(activate1(x1)) = x1
POL(c) = 0
POL(f1(x1)) = x1
POL(false) = 1
POL(if3(x1, x2, x3)) = x1 + x2 + x3
POL(n__f1(x1)) = x1
POL(n__true) = 0
POL(true) = 0
if3(false, X, Y) -> activate1(Y)
f1(X) -> if3(X, c, n__f1(n__true))
activate1(n__f1(X)) -> f1(activate1(X))
true -> n__true
f1(X) -> n__f1(X)
activate1(n__true) -> true
if3(true, X, Y) -> X
activate1(X) -> X
↳ QTRS
↳ DependencyPairsProof
↳ QDP
↳ DependencyGraphProof
↳ QDP
↳ QDPOrderProof
↳ QDP
↳ QDPOrderProof
↳ QDP
↳ DependencyGraphProof
IF3(false, X, Y) -> ACTIVATE1(Y)
F1(X) -> IF3(X, c, n__f1(n__true))
f1(X) -> if3(X, c, n__f1(n__true))
if3(true, X, Y) -> X
if3(false, X, Y) -> activate1(Y)
f1(X) -> n__f1(X)
true -> n__true
activate1(n__f1(X)) -> f1(activate1(X))
activate1(n__true) -> true
activate1(X) -> X