UP | HOME

Creating an automaton by adding states and transitions

Table of Contents

This example demonstrates how to create an automaton and then print it.

Transition-based Generalized Büchi automaton

For historical reasons, TGBAs are the more commonly used type of automata in Spot.

C++

#include <iostream>
#include <spot/twaalgos/hoa.hh>
#include <spot/twa/twagraph.hh>

int main(void)
{
  // The bdd_dict is used to maintain the correspondence between the
  // atomic propositions and the BDD variables that label the edges of
  // the automaton.
  spot::bdd_dict_ptr dict = spot::make_bdd_dict();
  // This creates an empty automaton that we have yet to fill.
  spot::twa_graph_ptr aut = make_twa_graph(dict);

  // Since a BDD is associated to every atomic proposition, the
  // register_ap() function returns a BDD variable number
  // that can be converted into a BDD using bdd_ithvar().
  bdd p1 = bdd_ithvar(aut->register_ap("p1"));
  bdd p2 = bdd_ithvar(aut->register_ap("p2"));

  // Set the acceptance condition of the automaton to Inf(0)&Inf(1)
  aut->set_generalized_buchi(2);

  // States are numbered from 0.
  aut->new_states(3);
  // The default initial state is 0, but it is always better to
  // specify it explicitely.
  aut->set_init_state(0U);

  // new_edge() takes 3 mandatory parameters: source state,
  // destination state, and label.  A last optional parameter can be
  // used to specify membership to acceptance sets.
  aut->new_edge(0, 1, p1);
  aut->new_edge(1, 1, p1 & p2, {0});
  aut->new_edge(1, 2, p2, {1});
  aut->new_edge(2, 1, p1 | p2, {0, 1});

  // Print the resulting automaton.
  print_hoa(std::cout, aut);
  return 0;
}
HOA: v1
States: 3
Start: 0
AP: 2 "p1" "p2"
acc-name: generalized-Buchi 2
Acceptance: 2 Inf(0)&Inf(1)
properties: trans-labels explicit-labels trans-acc
--BODY--
State: 0
[0] 1
State: 1
[0&1] 1 {0}
[1] 2 {1}
State: 2
[0 | 1] 1 {0 1}
--END--

Python

import spot
import buddy

# The bdd_dict is used to maintain the correspondence between the
# atomic propositions and the BDD variables that label the edges of
# the automaton.
bdict = spot.make_bdd_dict();
# This creates an empty automaton that we have yet to fill.
aut = spot.make_twa_graph(bdict)

# Since a BDD is associated to every atomic proposition, the register_ap()
# function returns a BDD variable number that can be converted into a BDD
# using bdd_ithvar() from the BuDDy library.
p1 = buddy.bdd_ithvar(aut.register_ap("p1"))
p2 = buddy.bdd_ithvar(aut.register_ap("p2"))

# Set the acceptance condition of the automaton to Inf(0)&Inf(1)
aut.set_generalized_buchi(2)

# States are numbered from 0.
aut.new_states(3)
# The default initial state is 0, but it is always better to
# specify it explicitely.
aut.set_init_state(0)

# new_edge() takes 3 mandatory parameters: source state, destination state,
# and label.  A last optional parameter can be used to specify membership to
# acceptance sets.  In the Python version, the list of acceptance sets the
# transition belongs to should be specified as a list.
aut.new_edge(0, 1, p1)
aut.new_edge(1, 1, p1 & p2, [0])
aut.new_edge(1, 2, p2, [1]);
aut.new_edge(2, 1, p1 | p2, [0, 1]);

# Print the resulting automaton.
print(aut.to_str('hoa'))
HOA: v1
States: 3
Start: 0
AP: 2 "p1" "p2"
acc-name: generalized-Buchi 2
Acceptance: 2 Inf(0)&Inf(1)
properties: trans-labels explicit-labels trans-acc
--BODY--
State: 0
[0] 1
State: 1
[0&1] 1 {0}
[1] 2 {1}
State: 2
[0 | 1] 1 {0 1}
--END--

Büchi automaton, with state-based acceptance

Spot does not really support state-based acceptance condition; this is faked as follows:

  • instead of marking states as accepting, we mark all outgoing edges of accepting states,
  • additionally, we set prop_state_acc(true) to indicate that the automaton should output as if it were state-based.

Some algorithm recognize the prop_state_acc() properties and trigger some special handling of the automaton, maybe to preserve its "fake state-based nature".

The print_hoa() function will check that automata marked with prop_state_acc() are actually using the same acceptance sets on all outgoing transitions of each state, and raise an exception otherwise.

In the following example, we are going to use Büchi acceptance, but prop_state_acc() can be used with any acceptance. Again, this property should be read as "this automaton with transition-based acceptance is actually encoding a state-based acceptance, so treat it as follows whenever possible".

C++

#include <iostream>
#include <spot/twaalgos/hoa.hh>
#include <spot/twa/twagraph.hh>

int main(void)
{
  // The bdd_dict is used to maintain the correspondence between the
  // atomic propositions and the BDD variables that label the edges of
  // the automaton.
  spot::bdd_dict_ptr dict = spot::make_bdd_dict();
  // This creates an empty automaton that we have yet to fill.
  spot::twa_graph_ptr aut = make_twa_graph(dict);

  // Since a BDD is associated to every atomic proposition, the
  // register_ap() function returns a BDD variable number
  // that can be converted into a BDD using bdd_ithvar().
  bdd p1 = bdd_ithvar(aut->register_ap("p1"));
  bdd p2 = bdd_ithvar(aut->register_ap("p2"));

  // Set the acceptance condition of the automaton to Inf(0)
  aut->set_buchi();
  // Pretend this is state-based acceptance
  aut->prop_state_acc(true);

  // States are numbered from 0.
  aut->new_states(3);
  // The default initial state is 0, but it is always better to
  // specify it explicitely.
  aut->set_init_state(0U);

  // new_edge() takes 3 mandatory parameters: source state,
  // destination state, and label.  A last optional parameter can be
  // used to specify membership to acceptance sets.
  aut->new_edge(0, 1, p1);
  // All edges leaving the same state must belong to the same
  // acceptance sets.
  aut->new_edge(1, 1, p1 & p2, {0});
  aut->new_edge(1, 2, p2, {0});
  aut->new_edge(2, 1, p1 | p2);
  // Print the resulting automaton.
  print_hoa(std::cout, aut);
  return 0;
}
HOA: v1
States: 3
Start: 0
AP: 2 "p1" "p2"
acc-name: Buchi
Acceptance: 1 Inf(0)
properties: trans-labels explicit-labels state-acc
--BODY--
State: 0
[0] 1
State: 1 {0}
[0&1] 1
[1] 2
State: 2
[0 | 1] 1
--END--

Python

import spot
import buddy

# The bdd_dict is used to maintain the correspondence between the
# atomic propositions and the BDD variables that label the edges of
# the automaton.
bdict = spot.make_bdd_dict();
# This creates an empty automaton that we have yet to fill.
aut = spot.make_twa_graph(bdict)

# Since a BDD is associated to every atomic proposition, the register_ap()
# function returns a BDD variable number that can be converted into a BDD
# using bdd_ithvar() from the BuDDy library.
p1 = buddy.bdd_ithvar(aut.register_ap("p1"))
p2 = buddy.bdd_ithvar(aut.register_ap("p2"))

# Set the acceptance condition of the automaton to Inf(0)
aut.set_buchi()
# Pretend this is state-based acceptance
aut.prop_state_acc(True);

# States are numbered from 0.
aut.new_states(3)
# The default initial state is 0, but it is always better to
# specify it explicitely.
aut.set_init_state(0)

# new_edge() takes 3 mandatory parameters: source state, destination state,
# and label.  A last optional parameter can be used to specify membership to
# acceptance sets.  In the Python version, the list of acceptance sets the
# transition belongs to should be specified as a list.
aut.new_edge(0, 1, p1)
# All edges leaving the same state must belong to the same acceptance sets.
aut.new_edge(1, 1, p1 & p2, [0])
aut.new_edge(1, 2, p2, [0]);
aut.new_edge(2, 1, p1 | p2);

# Print the resulting automaton.
print(aut.to_str('hoa'))
HOA: v1
States: 3
Start: 0
AP: 2 "p1" "p2"
acc-name: Buchi
Acceptance: 1 Inf(0)
properties: trans-labels explicit-labels state-acc
--BODY--
State: 0
[0] 1
State: 1 {0}
[0&1] 1
[1] 2
State: 2
[0 | 1] 1
--END--

Automaton with arbitrary acceptance condition

Generalized Büchi, and Büchi are common enough to warrant a dedicated method for setting the acceptance condition in the twa_graph class. Arbitrary acceptance condition can be set with set_acceptance.

C++

#include <iostream>
#include <spot/twaalgos/hoa.hh>
#include <spot/twa/twagraph.hh>

int main(void)
{
  spot::bdd_dict_ptr dict = spot::make_bdd_dict();
  spot::twa_graph_ptr aut = make_twa_graph(dict);
  bdd p1 = bdd_ithvar(aut->register_ap("p1"));
  bdd p2 = bdd_ithvar(aut->register_ap("p2"));

  // The acceptance condition can be parsed from a string,
  // or built from parts using spot::acc_cond.
  aut->set_acceptance(3, "(Inf(0) & Fin(1)) | Fin(2)");

  aut->new_states(3);
  aut->set_init_state(0U);
  aut->new_edge(0, 1, p1);
  aut->new_edge(1, 1, p1 & p2, {0});
  aut->new_edge(1, 2, p2, {1});
  aut->new_edge(2, 1, p1 | p2, {0, 2});

  print_hoa(std::cout, aut);
  return 0;
}
HOA: v1
States: 3
Start: 0
AP: 2 "p1" "p2"
Acceptance: 3 (Inf(0) & Fin(1)) | Fin(2)
properties: trans-labels explicit-labels trans-acc
--BODY--
State: 0
[0] 1
State: 1
[0&1] 1 {0}
[1] 2 {1}
State: 2
[0 | 1] 1 {0 2}
--END--

Python

import spot
import buddy

bdict = spot.make_bdd_dict();
aut = spot.make_twa_graph(bdict)
p1 = buddy.bdd_ithvar(aut.register_ap("p1"))
p2 = buddy.bdd_ithvar(aut.register_ap("p2"))

# The acceptance condition can be parsed from a string,
# or built from parts using spot.acc_cond.
aut.set_acceptance(3, "(Inf(0) & Fin(1)) | Fin(2)");

aut.new_states(3)
aut.set_init_state(0)
aut.new_edge(0, 1, p1)
aut.new_edge(1, 1, p1 & p2, [0])
aut.new_edge(1, 2, p2, [1]);
aut.new_edge(2, 1, p1 | p2, [0, 2]);

print(aut.to_str('hoa'))
HOA: v1
States: 3
Start: 0
AP: 2 "p1" "p2"
Acceptance: 3 (Inf(0) & Fin(1)) | Fin(2)
properties: trans-labels explicit-labels trans-acc
--BODY--
State: 0
[0] 1
State: 1
[0&1] 1 {0}
[1] 2 {1}
State: 2
[0 | 1] 1 {0 2}
--END--