YES(O(1),O(n^1)) 2.62/1.08 YES(O(1),O(n^1)) 2.62/1.08 2.62/1.08 We are left with following problem, upon which TcT provides the 2.62/1.08 certificate YES(O(1),O(n^1)). 2.62/1.08 2.62/1.08 Strict Trs: 2.62/1.08 { perfectp(0()) -> false() 2.62/1.08 , perfectp(s(x)) -> f(x, s(0()), s(x), s(x)) 2.62/1.08 , f(0(), y, 0(), u) -> true() 2.62/1.08 , f(0(), y, s(z), u) -> false() 2.62/1.08 , f(s(x), 0(), z, u) -> f(x, u, minus(z, s(x)), u) 2.62/1.08 , f(s(x), s(y), z, u) -> 2.62/1.08 if(le(x, y), f(s(x), minus(y, x), z, u), f(x, u, z, u)) } 2.62/1.08 Obligation: 2.62/1.08 innermost runtime complexity 2.62/1.08 Answer: 2.62/1.08 YES(O(1),O(n^1)) 2.62/1.08 2.62/1.08 We add the following weak dependency pairs: 2.62/1.08 2.62/1.08 Strict DPs: 2.62/1.08 { perfectp^#(0()) -> c_1() 2.62/1.08 , perfectp^#(s(x)) -> c_2(f^#(x, s(0()), s(x), s(x))) 2.62/1.08 , f^#(0(), y, 0(), u) -> c_3() 2.62/1.08 , f^#(0(), y, s(z), u) -> c_4() 2.62/1.08 , f^#(s(x), 0(), z, u) -> c_5(f^#(x, u, minus(z, s(x)), u)) 2.62/1.08 , f^#(s(x), s(y), z, u) -> 2.62/1.08 c_6(f^#(s(x), minus(y, x), z, u), f^#(x, u, z, u)) } 2.62/1.08 2.62/1.08 and mark the set of starting terms. 2.62/1.08 2.62/1.08 We are left with following problem, upon which TcT provides the 2.62/1.08 certificate YES(O(1),O(n^1)). 2.62/1.08 2.62/1.08 Strict DPs: 2.62/1.08 { perfectp^#(0()) -> c_1() 2.62/1.08 , perfectp^#(s(x)) -> c_2(f^#(x, s(0()), s(x), s(x))) 2.62/1.08 , f^#(0(), y, 0(), u) -> c_3() 2.62/1.08 , f^#(0(), y, s(z), u) -> c_4() 2.62/1.08 , f^#(s(x), 0(), z, u) -> c_5(f^#(x, u, minus(z, s(x)), u)) 2.62/1.08 , f^#(s(x), s(y), z, u) -> 2.62/1.08 c_6(f^#(s(x), minus(y, x), z, u), f^#(x, u, z, u)) } 2.62/1.08 Strict Trs: 2.62/1.08 { perfectp(0()) -> false() 2.62/1.08 , perfectp(s(x)) -> f(x, s(0()), s(x), s(x)) 2.62/1.08 , f(0(), y, 0(), u) -> true() 2.62/1.08 , f(0(), y, s(z), u) -> false() 2.62/1.08 , f(s(x), 0(), z, u) -> f(x, u, minus(z, s(x)), u) 2.62/1.08 , f(s(x), s(y), z, u) -> 2.62/1.08 if(le(x, y), f(s(x), minus(y, x), z, u), f(x, u, z, u)) } 2.62/1.08 Obligation: 2.62/1.08 innermost runtime complexity 2.62/1.08 Answer: 2.62/1.08 YES(O(1),O(n^1)) 2.62/1.08 2.62/1.08 No rule is usable, rules are removed from the input problem. 2.62/1.08 2.62/1.08 We are left with following problem, upon which TcT provides the 2.62/1.08 certificate YES(O(1),O(n^1)). 2.62/1.08 2.62/1.08 Strict DPs: 2.62/1.08 { perfectp^#(0()) -> c_1() 2.62/1.08 , perfectp^#(s(x)) -> c_2(f^#(x, s(0()), s(x), s(x))) 2.62/1.08 , f^#(0(), y, 0(), u) -> c_3() 2.62/1.08 , f^#(0(), y, s(z), u) -> c_4() 2.62/1.08 , f^#(s(x), 0(), z, u) -> c_5(f^#(x, u, minus(z, s(x)), u)) 2.62/1.08 , f^#(s(x), s(y), z, u) -> 2.62/1.08 c_6(f^#(s(x), minus(y, x), z, u), f^#(x, u, z, u)) } 2.62/1.08 Obligation: 2.62/1.08 innermost runtime complexity 2.62/1.08 Answer: 2.62/1.08 YES(O(1),O(n^1)) 2.62/1.08 2.62/1.08 The weightgap principle applies (using the following constant 2.62/1.08 growth matrix-interpretation) 2.62/1.08 2.62/1.08 The following argument positions are usable: 2.62/1.08 Uargs(c_2) = {1}, Uargs(c_5) = {1}, Uargs(c_6) = {2} 2.62/1.08 2.62/1.08 TcT has computed the following constructor-restricted matrix 2.62/1.08 interpretation. 2.62/1.08 2.62/1.08 [0] = [0] 2.62/1.08 [0] 2.62/1.08 2.62/1.08 [s](x1) = [0] 2.62/1.08 [0] 2.62/1.08 2.62/1.08 [minus](x1, x2) = [1 0] x1 + [0] 2.62/1.08 [0 0] [0] 2.62/1.08 2.62/1.08 [perfectp^#](x1) = [0] 2.62/1.08 [0] 2.62/1.08 2.62/1.08 [c_1] = [1] 2.62/1.08 [0] 2.62/1.08 2.62/1.08 [c_2](x1) = [1 0] x1 + [0] 2.62/1.08 [0 1] [0] 2.62/1.08 2.62/1.08 [f^#](x1, x2, x3, x4) = [1] 2.62/1.08 [0] 2.62/1.08 2.62/1.08 [c_3] = [0] 2.62/1.08 [0] 2.62/1.08 2.62/1.08 [c_4] = [0] 2.62/1.08 [0] 2.62/1.08 2.62/1.08 [c_5](x1) = [1 0] x1 + [0] 2.62/1.08 [0 1] [0] 2.62/1.08 2.62/1.08 [c_6](x1, x2) = [1 0] x2 + [0] 2.62/1.08 [0 1] [0] 2.62/1.08 2.62/1.08 The order satisfies the following ordering constraints: 2.62/1.08 2.62/1.08 [perfectp^#(0())] = [0] 2.62/1.08 [0] 2.62/1.08 ? [1] 2.62/1.08 [0] 2.62/1.08 = [c_1()] 2.62/1.08 2.62/1.08 [perfectp^#(s(x))] = [0] 2.62/1.08 [0] 2.62/1.08 ? [1] 2.62/1.08 [0] 2.62/1.08 = [c_2(f^#(x, s(0()), s(x), s(x)))] 2.62/1.08 2.62/1.08 [f^#(0(), y, 0(), u)] = [1] 2.62/1.08 [0] 2.62/1.08 > [0] 2.62/1.08 [0] 2.62/1.08 = [c_3()] 2.62/1.08 2.62/1.08 [f^#(0(), y, s(z), u)] = [1] 2.62/1.08 [0] 2.62/1.08 > [0] 2.62/1.08 [0] 2.62/1.08 = [c_4()] 2.62/1.08 2.62/1.08 [f^#(s(x), 0(), z, u)] = [1] 2.62/1.08 [0] 2.62/1.08 >= [1] 2.62/1.08 [0] 2.62/1.08 = [c_5(f^#(x, u, minus(z, s(x)), u))] 2.62/1.08 2.62/1.08 [f^#(s(x), s(y), z, u)] = [1] 2.62/1.08 [0] 2.62/1.08 >= [1] 2.62/1.08 [0] 2.62/1.08 = [c_6(f^#(s(x), minus(y, x), z, u), f^#(x, u, z, u))] 2.62/1.08 2.62/1.08 2.62/1.08 Further, it can be verified that all rules not oriented are covered by the weightgap condition. 2.62/1.08 2.62/1.08 We are left with following problem, upon which TcT provides the 2.62/1.08 certificate YES(O(1),O(n^1)). 2.62/1.08 2.62/1.08 Strict DPs: 2.62/1.08 { perfectp^#(0()) -> c_1() 2.62/1.08 , perfectp^#(s(x)) -> c_2(f^#(x, s(0()), s(x), s(x))) 2.62/1.08 , f^#(s(x), 0(), z, u) -> c_5(f^#(x, u, minus(z, s(x)), u)) 2.62/1.08 , f^#(s(x), s(y), z, u) -> 2.62/1.08 c_6(f^#(s(x), minus(y, x), z, u), f^#(x, u, z, u)) } 2.62/1.08 Weak DPs: 2.62/1.08 { f^#(0(), y, 0(), u) -> c_3() 2.62/1.08 , f^#(0(), y, s(z), u) -> c_4() } 2.62/1.08 Obligation: 2.62/1.08 innermost runtime complexity 2.62/1.08 Answer: 2.62/1.08 YES(O(1),O(n^1)) 2.62/1.08 2.62/1.08 We estimate the number of application of {1} by applications of 2.62/1.08 Pre({1}) = {}. Here rules are labeled as follows: 2.62/1.08 2.62/1.08 DPs: 2.62/1.08 { 1: perfectp^#(0()) -> c_1() 2.62/1.08 , 2: perfectp^#(s(x)) -> c_2(f^#(x, s(0()), s(x), s(x))) 2.62/1.08 , 3: f^#(s(x), 0(), z, u) -> c_5(f^#(x, u, minus(z, s(x)), u)) 2.62/1.08 , 4: f^#(s(x), s(y), z, u) -> 2.62/1.08 c_6(f^#(s(x), minus(y, x), z, u), f^#(x, u, z, u)) 2.62/1.08 , 5: f^#(0(), y, 0(), u) -> c_3() 2.62/1.08 , 6: f^#(0(), y, s(z), u) -> c_4() } 2.62/1.08 2.62/1.08 We are left with following problem, upon which TcT provides the 2.62/1.08 certificate YES(O(1),O(n^1)). 2.62/1.08 2.62/1.08 Strict DPs: 2.62/1.08 { perfectp^#(s(x)) -> c_2(f^#(x, s(0()), s(x), s(x))) 2.62/1.08 , f^#(s(x), 0(), z, u) -> c_5(f^#(x, u, minus(z, s(x)), u)) 2.62/1.08 , f^#(s(x), s(y), z, u) -> 2.62/1.08 c_6(f^#(s(x), minus(y, x), z, u), f^#(x, u, z, u)) } 2.62/1.08 Weak DPs: 2.62/1.08 { perfectp^#(0()) -> c_1() 2.62/1.08 , f^#(0(), y, 0(), u) -> c_3() 2.62/1.08 , f^#(0(), y, s(z), u) -> c_4() } 2.62/1.08 Obligation: 2.62/1.08 innermost runtime complexity 2.62/1.08 Answer: 2.62/1.08 YES(O(1),O(n^1)) 2.62/1.08 2.62/1.08 The following weak DPs constitute a sub-graph of the DG that is 2.62/1.08 closed under successors. The DPs are removed. 2.62/1.08 2.62/1.08 { perfectp^#(0()) -> c_1() 2.62/1.08 , f^#(0(), y, 0(), u) -> c_3() 2.62/1.08 , f^#(0(), y, s(z), u) -> c_4() } 2.62/1.08 2.62/1.08 We are left with following problem, upon which TcT provides the 2.62/1.08 certificate YES(O(1),O(n^1)). 2.62/1.08 2.62/1.08 Strict DPs: 2.62/1.08 { perfectp^#(s(x)) -> c_2(f^#(x, s(0()), s(x), s(x))) 2.62/1.08 , f^#(s(x), 0(), z, u) -> c_5(f^#(x, u, minus(z, s(x)), u)) 2.62/1.08 , f^#(s(x), s(y), z, u) -> 2.62/1.08 c_6(f^#(s(x), minus(y, x), z, u), f^#(x, u, z, u)) } 2.62/1.08 Obligation: 2.62/1.08 innermost runtime complexity 2.62/1.08 Answer: 2.62/1.08 YES(O(1),O(n^1)) 2.62/1.08 2.62/1.08 Due to missing edges in the dependency-graph, the right-hand sides 2.62/1.08 of following rules could be simplified: 2.62/1.08 2.62/1.08 { f^#(s(x), s(y), z, u) -> 2.62/1.08 c_6(f^#(s(x), minus(y, x), z, u), f^#(x, u, z, u)) } 2.62/1.08 2.62/1.08 We are left with following problem, upon which TcT provides the 2.62/1.08 certificate YES(O(1),O(n^1)). 2.62/1.08 2.62/1.08 Strict DPs: 2.62/1.08 { perfectp^#(s(x)) -> c_1(f^#(x, s(0()), s(x), s(x))) 2.62/1.08 , f^#(s(x), 0(), z, u) -> c_2(f^#(x, u, minus(z, s(x)), u)) 2.62/1.08 , f^#(s(x), s(y), z, u) -> c_3(f^#(x, u, z, u)) } 2.62/1.08 Obligation: 2.62/1.08 innermost runtime complexity 2.62/1.08 Answer: 2.62/1.08 YES(O(1),O(n^1)) 2.62/1.08 2.62/1.08 Consider the dependency graph 2.62/1.08 2.62/1.08 1: perfectp^#(s(x)) -> c_1(f^#(x, s(0()), s(x), s(x))) 2.62/1.08 -->_1 f^#(s(x), s(y), z, u) -> c_3(f^#(x, u, z, u)) :3 2.62/1.08 2.62/1.08 2: f^#(s(x), 0(), z, u) -> c_2(f^#(x, u, minus(z, s(x)), u)) 2.62/1.08 -->_1 f^#(s(x), s(y), z, u) -> c_3(f^#(x, u, z, u)) :3 2.62/1.08 -->_1 f^#(s(x), 0(), z, u) -> c_2(f^#(x, u, minus(z, s(x)), u)) :2 2.62/1.08 2.62/1.08 3: f^#(s(x), s(y), z, u) -> c_3(f^#(x, u, z, u)) 2.62/1.08 -->_1 f^#(s(x), s(y), z, u) -> c_3(f^#(x, u, z, u)) :3 2.62/1.08 -->_1 f^#(s(x), 0(), z, u) -> c_2(f^#(x, u, minus(z, s(x)), u)) :2 2.62/1.08 2.62/1.08 2.62/1.08 Following roots of the dependency graph are removed, as the 2.62/1.08 considered set of starting terms is closed under reduction with 2.62/1.08 respect to these rules (modulo compound contexts). 2.62/1.08 2.62/1.08 { perfectp^#(s(x)) -> c_1(f^#(x, s(0()), s(x), s(x))) } 2.62/1.08 2.62/1.08 2.62/1.08 We are left with following problem, upon which TcT provides the 2.62/1.08 certificate YES(O(1),O(n^1)). 2.62/1.08 2.62/1.08 Strict DPs: 2.62/1.08 { f^#(s(x), 0(), z, u) -> c_2(f^#(x, u, minus(z, s(x)), u)) 2.62/1.08 , f^#(s(x), s(y), z, u) -> c_3(f^#(x, u, z, u)) } 2.62/1.08 Obligation: 2.62/1.08 innermost runtime complexity 2.62/1.08 Answer: 2.62/1.08 YES(O(1),O(n^1)) 2.62/1.08 2.62/1.08 We use the processor 'Small Polynomial Path Order (PS,1-bounded)' 2.62/1.08 to orient following rules strictly. 2.62/1.08 2.62/1.08 DPs: 2.62/1.08 { 1: f^#(s(x), 0(), z, u) -> c_2(f^#(x, u, minus(z, s(x)), u)) 2.62/1.08 , 2: f^#(s(x), s(y), z, u) -> c_3(f^#(x, u, z, u)) } 2.62/1.08 2.62/1.08 Sub-proof: 2.62/1.08 ---------- 2.62/1.08 The input was oriented with the instance of 'Small Polynomial Path 2.62/1.08 Order (PS,1-bounded)' as induced by the safe mapping 2.62/1.08 2.62/1.08 safe(0) = {}, safe(s) = {1}, safe(minus) = {1, 2}, 2.62/1.08 safe(f^#) = {2, 3, 4}, safe(c_2) = {}, safe(c_3) = {} 2.62/1.08 2.62/1.08 and precedence 2.62/1.08 2.62/1.08 empty . 2.62/1.08 2.62/1.08 Following symbols are considered recursive: 2.62/1.08 2.62/1.08 {f^#} 2.62/1.08 2.62/1.08 The recursion depth is 1. 2.62/1.08 2.62/1.08 Further, following argument filtering is employed: 2.62/1.08 2.62/1.08 pi(0) = [], pi(s) = [1], pi(minus) = [], pi(f^#) = [1], 2.62/1.08 pi(c_2) = [1], pi(c_3) = [1] 2.62/1.08 2.62/1.08 Usable defined function symbols are a subset of: 2.62/1.08 2.62/1.08 {f^#} 2.62/1.08 2.62/1.08 For your convenience, here are the satisfied ordering constraints: 2.62/1.08 2.62/1.08 pi(f^#(s(x), 0(), z, u)) = f^#(s(; x);) 2.62/1.08 > c_2(f^#(x;);) 2.62/1.08 = pi(c_2(f^#(x, u, minus(z, s(x)), u))) 2.62/1.08 2.62/1.08 pi(f^#(s(x), s(y), z, u)) = f^#(s(; x);) 2.62/1.08 > c_3(f^#(x;);) 2.62/1.08 = pi(c_3(f^#(x, u, z, u))) 2.62/1.08 2.62/1.08 2.62/1.08 The strictly oriented rules are moved into the weak component. 2.62/1.08 2.62/1.08 We are left with following problem, upon which TcT provides the 2.62/1.08 certificate YES(O(1),O(1)). 2.62/1.08 2.62/1.08 Weak DPs: 2.62/1.08 { f^#(s(x), 0(), z, u) -> c_2(f^#(x, u, minus(z, s(x)), u)) 2.62/1.08 , f^#(s(x), s(y), z, u) -> c_3(f^#(x, u, z, u)) } 2.62/1.08 Obligation: 2.62/1.08 innermost runtime complexity 2.62/1.08 Answer: 2.62/1.08 YES(O(1),O(1)) 2.62/1.08 2.62/1.08 The following weak DPs constitute a sub-graph of the DG that is 2.62/1.08 closed under successors. The DPs are removed. 2.62/1.08 2.62/1.08 { f^#(s(x), 0(), z, u) -> c_2(f^#(x, u, minus(z, s(x)), u)) 2.62/1.08 , f^#(s(x), s(y), z, u) -> c_3(f^#(x, u, z, u)) } 2.62/1.08 2.62/1.08 We are left with following problem, upon which TcT provides the 2.62/1.08 certificate YES(O(1),O(1)). 2.62/1.08 2.62/1.08 Rules: Empty 2.62/1.08 Obligation: 2.62/1.08 innermost runtime complexity 2.62/1.08 Answer: 2.62/1.08 YES(O(1),O(1)) 2.62/1.08 2.62/1.08 Empty rules are trivially bounded 2.62/1.08 2.62/1.08 Hurray, we answered YES(O(1),O(n^1)) 2.62/1.08 EOF