Closed astefano closed 7 years ago
I've added an example to the README file.
i get bin/mlsolver: unknown option '-stenv'.
what are, in fact, the arguments to mlsolver? (-h gives the ones for pgsolver, i assume).
bin/mlsolver --help Modal Logic Solver 1.3 Authors: Oliver Friedmann (University of Munich) and Martin Lange (University of Kassel), 2008-2017
Usage: mlsolver [options] [formula]
Options are
--validity, -val
strange, as i initially wrote, i get:
lacra@lacra-VirtualBox:~/tools/mlsolver$ bin/mlsolver -help
PGSolver Collection Ver. 4.0: Parity Game Solver
Authors: Oliver Friedmann (University of Munich) and Martin Lange (University of Kassel), 2008-2016 http://tcsprojects.org
Usage: pgsolver [options] [infile]
Solves the parity game given in <infile>. If this argument is omitted it reads a game from STDIN.
Options are
-v <level> sets the verbosity level, valid arguments are 0 (quiet), 1 (default), 2 (verbose) and 3 (very verbose)
--printsolonly causes the program to simply output a parsable solution
-d <filename> output the solution as a coloured graph in dotty format into <filename>
--printsolvedgame, -pg outputs the solved (!) game on STDOUT
...
That is definitely PGSolver. Did you run make in the pgsolver subdirectory instead of the mlsolver directory?
no, no, i followed your instructions precisely (make from mlsolver). so i get in bin:
lacra@lacra-VirtualBox:~/tools/mlsolver$ ls bin
ctlstarsudokugenerator elevatortsgenerator guarded_trafo_worst_case ltmcparitybuechigenerator mlsolver mucalcsudokugenerator pdlsudokugenerator philosopherstsgenerator
I've adjusted some ambiguously named files - please try one more time. Not sure why we get different results, but it might be related to the order in which our system are enumerating files.
thanks a lot, now it works! one last question: what's the grammar for transition systems? (can't find where Tcstransitionparser.parse... is implemented...)
Apologies for the lack of documentation, this is definitely something we need to work on.
In the meantime, formula parsing is in TCSLib: https://github.com/tcsprojects/tcslib/tree/master/src/formula/parser
ok, thanks! yep, some documentation would be great :) Please bear with me a little bit more... How should one interpret the output? For instance, there is no difference between checking true or [a]true.
lacra@lacra-VirtualBox:~/tools/mlsolver$ bin/mlsolver -mc pdl -ts ex.ts "<a>true"
Modal Logic Solver 1.3
Authors: Oliver Friedmann (University of Munich) and Martin Lange (University of Kassel), 2008-2017
Evaluating formula... 0.00 sec
Using Modelchecking Procedure For PDL
Starting procedure...
Modelchecking Procedure For PDL
lacra@lacra-VirtualBox:~/tools/mlsolver$ bin/mlsolver -mc pdl -ts ex.ts "[a]true"
Modal Logic Solver 1.3
Authors: Oliver Friedmann (University of Munich) and Martin Lange (University of Kassel), 2008-2017
Evaluating formula... 0.00 sec
Using Modelchecking Procedure For PDL
Starting procedure...
Modelchecking Procedure For PDL
Shouldn't it output (un)sat?
Just for the reference, i just took a simple TS: lts 2; start 0; 0 a : 1; 1 b : 0;
True is "tt" (see parser). And you always have to specify a solver, e.g. -pgs recursive
ok, thanks a lot for all your help!
Hi again,
Could we please have a running example? And how to run it, currently, i get:
lacra@lacra-VirtualBox:~/tools/mlsolver/bin$ ./mlsolver --help
PGSolver Collection Ver. 4.0: Parity Game Solver
Authors: Oliver Friedmann (University of Munich) and Martin Lange (University of Kassel), 2008-2016 http://tcsprojects.org
Usage: pgsolver [options] [infile]
Solves the parity game given in <infile>. If this argument is omitted it reads a game from STDIN.
Options are
-v <level>
sets the verbosity level, valid arguments are 0 (quiet), 1 (default), 2 (verbose) and 3 (very verbose)
--printsolonly
causes the program to simply output a parsable solution
-d <filename>
output the solution as a coloured graph in dotty format into <filename>
--printsolvedgame, -pg
outputs the solved (!) game on STDOUT
--parsesolution, -ps <filename>
parses the solution to the game from FILE
--justheatCPU, -jh
suppress the printing of the strategies and the winning regions
--verify, -ve
verify solution using the universal solver (fast)
--disableglobalopt, -dgo
disable global optimization
--disablesccdecomposition, -dsd
disable scc decomposition
--disablelocalopt, -dlo
disable local optimization
--disablespecialgames, -dsg
disable optimized solving of special games
--solverinfo
output information about all available solvers
--globallysolve, -global <solver>
solves globally, valid solvers are bigstep dominiondec fpiter genetic guessstrategy modelchecker optstratimprov policyiter recursive smallprog stratimprdisc stratimprloc2 stratimprlocal stratimprove
--locallysolve, -local <solver> <node>
solves locally, valid solvers are modelchecker stratimprloc2 stratimprlocal
--args, -x
pass args to the solver (write '-x "--help"' to learn about available args (if there are any))
--generator, -gen <generator> "<args>"
use generator, valid ones are cliquegame clusteredrg elevatorvergm jurdzinskigame laddergame langincl modelcheckerladder randomgame recursiveladder steadygame towersofhanoi
I tried to see if maybe the generators can be seen as examples, but not sure how to run correctly:
lacra@lacra-VirtualBox:~/tools/mlsolver/bin$ ./mlsolver -gen modelcheckerladder
./mlsolver: option '-gen' needs an argument.
Cheers, lacramioara