tgba.hh 9.62 KB
Newer Older
1
// Copyright (C) 2009, 2011 Laboratoire de Recherche et Développement
Guillaume Sadegh's avatar
Guillaume Sadegh committed
2
3
// de l'Epita (LRDE).
// Copyright (C) 2003, 2004, 2005 Laboratoire d'Informatique de
4
5
// Paris 6 (LIP6), département Systèmes Répartis Coopératifs (SRC),
// Université Pierre et Marie Curie.
Alexandre Duret-Lutz's avatar
Alexandre Duret-Lutz committed
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
//
// This file is part of Spot, a model checking library.
//
// Spot is free software; you can redistribute it and/or modify it
// under the terms of the GNU General Public License as published by
// the Free Software Foundation; either version 2 of the License, or
// (at your option) any later version.
//
// Spot is distributed in the hope that it will be useful, but WITHOUT
// ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
// or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public
// License for more details.
//
// You should have received a copy of the GNU General Public License
// along with Spot; see the file COPYING.  If not, write to the Free
// Software Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA
// 02111-1307, USA.

24
25
26
#ifndef SPOT_TGBA_TGBA_HH
# define SPOT_TGBA_TGBA_HH

27
#include "state.hh"
28
#include "succiter.hh"
29
#include "bdddict.hh"
30
31
32

namespace spot
{
33
  /// \defgroup tgba TGBA (Transition-based Generalized Büchi Automata)
34
35
36
37
38
39
40
41
42
43
44
45
  ///
  /// Spot is centered around the spot::tgba type.  This type and its
  /// cousins are listed \ref tgba_essentials "here".  This is an
  /// abstract interface.  Its implementations are either \ref
  /// tgba_representation "concrete representations", or \ref
  /// tgba_on_the_fly_algorithms "on-the-fly algorithms".  Other
  /// algorithms that work on spot::tgba are \ref tgba_algorithms
  /// "listed separately".

  /// \addtogroup tgba_essentials Essential TGBA types
  /// \ingroup tgba

46
  /// \brief A Transition-based Generalized Büchi Automaton.
47
  /// \ingroup tgba_essentials
48
49
50
51
52
53
54
55
  ///
  /// The acronym TGBA (Transition-based Generalized Büchi Automaton)
  /// was coined by Dimitra Giannakopoulou and Flavio Lerda
  /// in "From States to Transitions: Improving Translation of LTL
  /// Formulae to Büchi Automata".  (FORTE'02)
  ///
  /// TGBAs are transition-based, meanings their labels are put
  /// on arcs, not on nodes.  They use Generalized Büchi acceptance
56
  /// conditions: there are several acceptance sets (of
Alexandre Duret-Lutz's avatar
typos  
Alexandre Duret-Lutz committed
57
  /// transitions), and a path can be accepted only if it traverses
58
59
  /// at least one transition of each set infinitely often.
  ///
Alexandre Duret-Lutz's avatar
typos  
Alexandre Duret-Lutz committed
60
  /// Browsing such automaton can be achieved using two functions:
61
  /// \c get_init_state, and \c succ_iter.  The former returns
Alexandre Duret-Lutz's avatar
typos  
Alexandre Duret-Lutz committed
62
  /// the initial state while the latter lists the
63
64
65
66
67
68
  /// successor states of any state.
  ///
  /// 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.
69
70
  class tgba
  {
71
72
  protected:
    tgba();
73

74
  public:
75
76
    virtual ~tgba();

77
78
79
    /// \brief Get the initial state of the automaton.
    ///
    /// The state has been allocated with \c new.  It is the
80
    /// responsability of the caller to \c destroy it when no
81
82
    /// longer needed.
    virtual state* get_init_state() const = 0;
83

84
    /// \brief Get an iterator over the successors of \a local_state.
85
86
87
88
89
    ///
    /// The iterator has been allocated with \c new.  It is the
    /// responsability of the caller to \c delete it when no
    /// longer needed.
    ///
90
91
92
93
94
95
96
97
98
    /// During synchornized products, additional informations are
    /// passed about the entire product and its state.  Recall that
    /// products can be nested, forming a tree of spot::tgba where
    /// most values are computed on demand.  \a global_automaton
    /// designate the root spot::tgba, and \a global_state its
    /// state.  This two objects can be used by succ_iter() to
    /// restrict the set of successors to compute.
    ///
    /// \param local_state The state whose successors are to be explored.
99
    /// This pointer is not adopted in any way by \c succ_iter, and
100
    /// it is still the caller's responsability to destroy it when
101
102
    /// appropriate (this can be done during the lifetime of
    /// the iterator).
103
104
105
    /// \param global_state In a product, the state of the global
    /// product automaton.  Otherwise, 0.  Like \a locale_state,
    /// \a global_state is not adopted by \c succ_iter.
106
    /// \param global_automaton In a product, the global
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
    /// product automaton.  Otherwise, 0.
    virtual tgba_succ_iterator*
    succ_iter(const state* local_state,
	      const state* global_state = 0,
	      const tgba* global_automaton = 0) const = 0;

    /// \brief Get a formula that must hold whatever successor is taken.
    ///
    /// \return A formula which must be verified for all successors
    ///  of \a state.
    ///
    /// This can be as simple as \c bddtrue, or more completely
    /// the disjunction of the condition of all successors.  This
    /// is used as an hint by \c succ_iter() to reduce the number
    /// of successor to compute in a product.
    ///
    /// Sub classes should implement compute_support_conditions(),
    /// this function is just a wrapper that will cache the
    /// last return value for efficiency.
    bdd support_conditions(const state* state) const;

    /// \brief Get the conjunctions of variables tested by
    ///        the outgoing transitions of \a state.
    ///
    /// All variables tested by outgoing transitions must be
    /// returned.  This is mandatory.
    ///
    /// This is used as an hint by some \c succ_iter() to reduce the
    /// number of successor to compute in a product.
    ///
    /// Sub classes should implement compute_support_variables(),
    /// this function is just a wrapper that will cache the
    /// last return value for efficiency.
    bdd support_variables(const state* state) const;
141
142
143
144
145
146
147
148

    /// \brief 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.
149
    virtual bdd_dict* get_dict() const = 0;
150
151
152
153

    /// \brief Format the state as a string for printing.
    ///
    /// This formating is the responsability of the automata
154
    /// that owns the state.
155
    virtual std::string format_state(const state* state) const = 0;
156

157
158
159
160
161
162
163
164
165
166
    /// \brief Return a possible annotation for the transition
    /// pointed to by the iterator.
    ///
    /// Implementing this function is optional; the default annotation
    /// it the empty string.
    ///
    /// \param t a non-done tgba_succ_iterator for this automata
    virtual std::string
    transition_annotation(const tgba_succ_iterator* t) const;

167
    /// \brief Project a state on an automaton.
168
169
170
171
172
173
174
175
176
177
178
    ///
    /// This converts \a s, into that corresponding spot::state for \a
    /// t.  This is useful when you have the state of a product, and
    /// want restrict this state to a specific automata occuring in
    /// the product.
    ///
    /// It goes without saying that \a s and \a t should be compatible
    /// (i.e., \a s is a state of \a t).
    ///
    /// \return 0 if the projection fails (\a s is unrelated to \a t),
    ///    or a new \c state* (the projected state) that must be
179
    ///    destroyed by the caller.
180
181
    virtual state* project_state(const state* s, const tgba* t) const;

182
    /// \brief Return the set of all acceptance conditions used
183
184
185
186
187
188
189
    /// by this automaton.
    ///
    /// 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.
190
    virtual bdd all_acceptance_conditions() const = 0;
191

192
    /// The number of acceptance conditions.
193
    virtual unsigned int number_of_acceptance_conditions() const;
194

195
    /// \brief Return the conjuction of all negated acceptance
196
    /// variables.
197
    ///
198
    /// For instance if the automaton uses variables <tt>Acc[a]</tt>,
199
    /// <tt>Acc[b]</tt> and <tt>Acc[c]</tt> to describe acceptance sets,
200
    /// this function should return <tt>!Acc[a]\&!Acc[b]\&!Acc[c]</tt>.
201
    ///
202
    /// This is useful when making products: each operand's condition
203
    /// set should be augmented with the neg_acceptance_conditions() of
204
    /// the other operand.
205
    virtual bdd neg_acceptance_conditions() const = 0;
206
207
208
209
210
211
212
213
214
215
216

  protected:
    /// Do the actual computation of tgba::support_conditions().
    virtual bdd compute_support_conditions(const state* state) const = 0;
    /// Do the actual computation of tgba::support_variables().
    virtual bdd compute_support_variables(const state* state) const = 0;
  private:
    mutable const state* last_support_conditions_input_;
    mutable bdd last_support_conditions_output_;
    mutable const state* last_support_variables_input_;
    mutable bdd last_support_variables_output_;
217
    mutable int num_acc_;
218
  };
219

220
221
222
223
224
  /// \addtogroup tgba_representation TGBA representations
  /// \ingroup tgba

  /// \addtogroup tgba_algorithms TGBA algorithms
  /// \ingroup tgba
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242

  /// \addtogroup tgba_on_the_fly_algorithms TGBA on-the-fly algorithms
  /// \ingroup tgba_algorithms

  /// \addtogroup tgba_io Input/Output of TGBA
  /// \ingroup tgba_algorithms

  /// \addtogroup tgba_ltl Translating LTL formulae into TGBA
  /// \ingroup tgba_algorithms

  /// \addtogroup tgba_generic Algorithm patterns
  /// \ingroup tgba_algorithms

  /// \addtogroup tgba_reduction TGBA simplifications
  /// \ingroup tgba_algorithms

  /// \addtogroup tgba_misc Miscellaneous algorithms on TGBA
  /// \ingroup tgba_algorithms
243
}
244
245

#endif // SPOT_TGBA_TGBA_HH