spot
1.2.6
|
A Testing Automaton. More...
#include <ta/ta.hh>
Public Types | |
typedef std::set< state *, state_ptr_less_than > | states_set_t |
Public Member Functions | |
virtual const states_set_t | get_initial_states_set () const =0 |
Get the initial states set of the automaton. More... | |
virtual spot::state * | get_artificial_initial_state () const |
Get the artificial initial state set of the automaton. Return 0 if this artificial state is not implemented (in this case, use get_initial_states_set ) The aim of adding this state is to have an unique initial state. This artificial initial state have one transition to each real initial state, and this transition is labeled by the corresponding initial condition. (For more details, see the paper cited above) More... | |
virtual ta_succ_iterator * | succ_iter (const spot::state *state) const =0 |
Get an iterator over the successors of state. More... | |
virtual ta_succ_iterator * | succ_iter (const spot::state *state, bdd changeset) const =0 |
Get an iterator over the successors of state filtred by the changeset on transitions. More... | |
virtual bdd_dict * | get_dict () const =0 |
Get the dictionary associated to the automaton. More... | |
virtual std::string | format_state (const spot::state *s) const =0 |
Format the state as a string for printing. More... | |
virtual bool | is_accepting_state (const spot::state *s) const =0 |
Return true if s is a Buchi-accepting state, otherwise false. More... | |
virtual bool | is_livelock_accepting_state (const spot::state *s) const =0 |
Return true if s is a livelock-accepting state , otherwise false. More... | |
virtual bool | is_initial_state (const spot::state *s) const =0 |
Return true if s is an initial state, otherwise false. More... | |
virtual bdd | get_state_condition (const spot::state *s) const =0 |
Return a BDD condition that represents the valuation of atomic propositions in the state s. More... | |
virtual void | free_state (const spot::state *s) const =0 |
Release a state s. More... | |
virtual bdd | all_acceptance_conditions () const =0 |
Return the set of all acceptance conditions used by this automaton (for Generalized form: Transition-based Generalized Testing Automata). More... | |
A Testing Automaton.
The Testing Automata (TA) were introduced by Henri Hansen, Wojciech Penczek and Antti Valmari in "Stuttering-insensitive automata for on-the-fly detection of livelock properties" In Proc. of FMICSÃ02, vol. 66(2) of Electronic Notes in Theoretical Computer Science.Elsevier.
While a TGBA automaton observes the value of the atomic propositions, the basic idea of TA is to detect the changes in these values; if a valuation does not change between two consecutive valuations of an execution, the TA stay in the same state. A TA transition (s,k,d) is labeled by a "changeset"
k:
i.e. the set of atomic propositions that change between states s
and d
, if the changeset is empty then the transition is called stuttering transition. To detect execution that ends by stuttering in the same TA state, a new kind of acceptance states is introduced: "livelock-acceptance states" (in addition to the standard Buchi-acceptance states).
Browsing such automaton can be achieved using two functions: get_initial_states_set
or get_artificial_initial_state
, and succ_iter
. The former returns the initial state(s) while the latter lists the successor states of any state (filtred by "changeset").
Note that although this is a transition-based automata, we never represent transitions! Transition informations are obtained by querying the iterator over the successors of a state.
|
pure virtual |
Return the set of all acceptance conditions used by this automaton (for Generalized form: Transition-based Generalized Testing Automata).
The goal of the emptiness check is to ensure that a strongly connected component walks through each of these acceptiong conditions. I.e., the union of the acceptiong conditions of all transition in the SCC should be equal to the result of this function.
Implemented in spot::ta_product, and spot::ta_explicit.
|
pure virtual |
Format the state as a string for printing.
This formating is the responsability of the automata that owns the state.
Implemented in spot::ta_product, and spot::ta_explicit.
|
pure virtual |
Release a state s.
Implemented in spot::ta_product, and spot::ta_explicit.
|
inlinevirtual |
Get the artificial initial state set of the automaton. Return 0 if this artificial state is not implemented (in this case, use get_initial_states_set
) The aim of adding this state is to have an unique initial state. This artificial initial state have one transition to each real initial state, and this transition is labeled by the corresponding initial condition. (For more details, see the paper cited above)
Reimplemented in spot::ta_explicit.
|
pure virtual |
Get the dictionary associated to the automaton.
State are represented as BDDs. The dictionary allows to map BDD variables back to formulae, and vice versa. This is useful when dealing with several automata (which may use the same BDD variable for different formula), or simply when printing.
Implemented in spot::ta_product, and spot::ta_explicit.
|
pure virtual |
Get the initial states set of the automaton.
Implemented in spot::ta_product, and spot::ta_explicit.
|
pure virtual |
Return a BDD condition that represents the valuation of atomic propositions in the state s.
Implemented in spot::ta_product, and spot::ta_explicit.
|
pure virtual |
Return true if s is a Buchi-accepting state, otherwise false.
Implemented in spot::ta_product, and spot::ta_explicit.
|
pure virtual |
Return true if s is an initial state, otherwise false.
Implemented in spot::ta_product, and spot::ta_explicit.
|
pure virtual |
Return true if s is a livelock-accepting state , otherwise false.
Implemented in spot::ta_product, and spot::ta_explicit.
|
pure virtual |
Get an iterator over the successors of state.
The iterator has been allocated with new
. It is the responsability of the caller to delete
it when no longer needed.
Implemented in spot::ta_product, and spot::ta_explicit.
|
pure virtual |
Get an iterator over the successors of state filtred by the changeset on transitions.
The iterator has been allocated with new
. It is the responsability of the caller to delete
it when no longer needed.
Implemented in spot::ta_product, and spot::ta_explicit.