When something like XFa & FXa is reduced, the subformulae XFa and FXa
are both rewritten separately to XFa, and then the vector of arguments
of the And operators, [XFa,XFa], is passed through a specialized loop
that searches of the form X(...) that can potentially be simplified with
some other terms. This loop converts the vector [XFa,XFa] into the set
{XFa,XFa}={XFa} and forgot to deal with the case where the insertion
would actually not add an existing subformula.
* src/ltlvisit/simplify.cc: Fix the code for Or, and And.
* src/ltltest/reduc0.test: New file, to test it.
* src/ltltest/Makefile.am (TESTS): Add it.
* src/ltltest/reduccmp.test: Add an extra test that does not
trigger the bug (because reduccmp.test uses more than basic
optimizations, and the implication-based simplifications are
already able to detect that XFa and FXa are equivalent).
* src/tgbaalgos/postproc.hh (run): Rename the first argument as
input_disown to help Swig.
* wrap/python/spot.i: Wrap spot::postprocessor.
* wrap/python/ajax/ltl2tgba.html, wrap/python/ajax/protocol.txt:
Add an option for nondeterministic monitor.
* wrap/python/ajax/spot.in: Honor the new option, and rewrite the
monitor production using postprocessor.
* src/tgbatest/ltl2tgba.cc: Calling tgbatest/ltl2tgba -M -O (which
makes no sense, but that is no reason) used the "minimized" variable
for two automata, overwriting one.
* wrap/python/spot.i: The python bindings did not know about
sba_explicit automata, causing memory leaks, and complaints from the
bdd_dict.
Enable LTL simplifications by default for ltl2tgba & ltl2tgta, and make
sure the ltl_simplifier_options are all false initially. Before this
patch --low/-r1 had the same effect as --medium/-r2 with respect to LTL
simplification.
* src/bin/ltl2tgba.cc, src/bin/ltl2tgta.cc (simplification_level): Set
to 3 by default.
* src/bin/common_r.cc: Disable all ltl_simplifier options initially.
This follows up on a mail from Sonali Dutta.
* src/tgba/bdddict.hh (assert_emptiness, ~bdd_dict): Better
documentation.
* src/tgba/formula2bdd.hh (formula_to_bdd): Mention
unregister_all_my_variables().
(bdd_to_formula): Complete the documentation.
* THANKS: Add Sonali Dutta.
* src/misc/formater.cc, src/misc/formater.hh (scan): New method.
(prime): Use it.
* src/bin/ltlcross.cc (translator_runner::translator_runner): Scan each
specification string, and report those missing an input or output
%-sequence.
* NEWS: Mention it.
* src/ltlparse/ltlscan.ll: Accept as a proposition any
alphanumeric string that is not an operator.
* NEWS: Mention it.
* src/ltltest/lbt.test: New file. Also tests previous patch.
* src/ltltest/Makefile.am: Add it.
* src/tgba/tgbakvcomplement.cc
(tgba_kv_complement_succ_iterator::current_state_),
src/ta/taexplicit.hh (state_ta_explicit::source_):
Remove useless private member.
* src/ta/taexplicit.cc: Adjust constructors.
* src/ta/tgta.cc, src/ta/taexplicit.hh: Also fix
copyright banner.
* src/bin/ltlcross.cc (exec_with_timeout): Work
around warning about status not being set in the
error path.
* bench/wdba/defs.in, bench/ltlclasses/defs.in,
bench/ltlcounter/defs.in: Delete.
* bench/wdba/run, bench/ltlclasses/run, bench/ltlcounter/run: Adjust not
to use them.
* configure.ac: Do not output the associated defs files.
* bench/ltl2tgba/sum.py: New file.
* bench/ltl2tgba/.gitignore, bench/ltl2tgba/Makefile.am,
bench/ltl2tgba/README, bench/ltl2tgba/algorithms, bench/ltl2tgba/big,
bench/ltl2tgba/defs.in, bench/ltl2tgba/known, bench/ltl2tgba/small:
Rewrite this benchmark completely. Also drop support of Wring and
Modella, as we cannot get them to work reliably.
* bench/ltl2tgba/formulae.ltl: Rewrite in Spot's syntax.
* bench/ltl2tgba/lbtt2csv.pl, bench/ltl2tgba/ltl2baw.in,
bench/ltl2tgba/parseout.pl: Delete these scripts, no
longer needed.
* configure.ac: Do not output ltl2baw.pl anymore.
