postprocessing twice may improve things
This example was sent by Fanda.
% cat p.hoa
HOA: v1 States: 6 Start: 0 AP: 3 "a" "b" "c" acc-name: Buchi
Acceptance: 1 Inf(0) properties: trans-labels explicit-labels trans-acc
semi-deterministic --BODY-- State: 0 [t] 0 [!0&1&2] 1 [0&2 | !1] 2
[0&!1] 3 State: 1 [!0&1&2] 1 [0&2 | !1] 2 [0&!1] 3 State: 2 [!0&1&2]
1 [0&2 | !1] 2 [0&!1] 3 State: 3 [!1] 3 [!1] 4 State: 4 [0&!1] 4 {0}
[!0&!1] 5 {0} State: 5 [!0&!1] 5 [0&!1] 5 {0} --END--
% autfilt --stats=%s p.hoa
6
% autfilt --small --stats=%s p.hoa
4
% autfilt --small p.hoa | autfilt --small --stats=%s
3
It would nice if the first postprocessing would give the smallest reasult already.