alarsyo/spot
1
0
Fork 0
Code Issues Pull requests Projects Releases Wiki Activity

parity game: compute winning strategy

Browse source 
* spot/misc/game.cc, spot/misc/game.hh: Here.
* bin/ltlsynt.cc: Realizability is now done by checking if the winning
strategy contains the initial state.
This commit is contained in:
Thibaud Michaud 2017-05-02 13:59:00 +02:00
parent f414e9f5f2
commit 601e1405de
3 changed files with 95 additions and 74 deletions
Download patch file Download diff file Expand all files Collapse all files

26
spot/misc/game.hh
View file

@ -22,6 +22,7 @@
#include <algorithm>
#include <memory>
#include <ostream>
#include <unordered_map>
#include <vector>
#include <bddx.h>
@ -94,8 +95,12 @@ public:
/// Print the parity game in PGSolver's format.
void print(std::ostream& os);
// Compute the winner of this game using Zielonka's recursive algorithm.
// False is Even and True is Odd.
typedef std::unordered_set<unsigned> region_t;
// Map state number to index of the transition to take.
typedef std::unordered_map<unsigned, unsigned> strategy_t;
/// Compute the winning strategy and winning region of this game for player
/// 1 using Zielonka's recursive algorithm.
/** \verbatim
@article{ zielonka.98.tcs
title = "Infinite games on finitely coloured graphs with applications to
@ -108,7 +113,7 @@ public:
author = "Wieslaw Zielonka",
}
\endverbatim */
bool winner() const;
std::pair<region_t, strategy_t> solve() const;
/// Whether player 1 has a winning strategy from the initial state.
/// Implements Calude et al.'s quasipolynomial time algorithm.
@ -139,17 +144,16 @@ private:
typedef twa_graph::graph_t::edge_storage_t edge_t;
// Compute (in place) a set of states from which player can force a visit
// through set.
// through set, and a strategy to do it.
// if attr_max is true, states that can force a visit through an edge with
// max parity are also counted in.
void attractor(const std::unordered_set<unsigned>& subgame,
std::unordered_set<unsigned>& set, unsigned max_parity,
bool player, bool attr_max = false) const;
strategy_t attractor(const region_t& subgame, region_t& set,
unsigned max_parity, bool odd,
bool attr_max = false) const;
// Compute the winning region for player Odd.
std::unordered_set<unsigned>
winning_region(std::unordered_set<unsigned>& subgame,
unsigned max_parity) const;
// Compute the winning strategy and winning region for player 1.
std::pair<region_t, strategy_t>
solve_rec(region_t& subgame, unsigned max_parity) const;
};