* src/graph/ngraph.hh: New file.
* src/graph/Makefile.am: Add it.
* src/graphtest/ngraph.cc, src/graphtest/ngraph.test: New files.
* src/graphtest/Makefile.am: Add them
star_normal_form() used to be called under bounded
repetitions like [*0..4], but some of these rewritings
are only correct for [*0..]. For instance
(a*|1)[*] can be rewritten to 1[*]
but (a*|1)[*0..1] cannot be rewritten to 1[*0..1]
it would be correct to rewrite the latter as (a[+]|1)[*0..1],
canceling the empty word in a*.
Also (a*;b*)[*] can be rewritten to (a|b)[*]
but (a*;b*)[*0..1] cannot be rewritten to (a|b)[*0..1]
and it cannot either be rewritten to (a[+]|b[+])[*0..1].
This patch introduces a new function to implement
rewritings under bounded repetition.
* src/ltlvisit/snf.hh, src/ltlvisit/snf.cc (star_normal_form_unbounded):
New function.
* src/ltlvisit/simplify.cc: Use it.
* src/ltltest/reduccmp.test: Add tests.
* doc/tl/tl.tex: Document the rewritings implemented.
star_normal_form() used to be called under bounded
repetitions like [*0..4], but some of these rewritings
are only correct for [*0..]. For instance
(a*|1)[*] can be rewritten to 1[*]
but (a*|1)[*0..1] cannot be rewritten to 1[*0..1]
it would be correct to rewrite the latter as (a[+]|1)[*0..1],
canceling the empty word in a*.
Also (a*;b*)[*] can be rewritten to (a|b)[*]
but (a*;b*)[*0..1] cannot be rewritten to (a|b)[*0..1]
and it cannot either be rewritten to (a[+]|b[+])[*0..1].
This patch introduces a new function to implement
rewritings under bounded repetition.
* src/ltlvisit/snf.hh, src/ltlvisit/snf.cc (star_normal_form_unbounded):
New function.
* src/ltlvisit/simplify.cc: Use it.
* src/ltltest/reduccmp.test: Add tests.
* doc/tl/tl.tex: Document the rewritings implemented.
Fortunately was only enabled with the
ltl_simplifier_options::favor_event_univ option, which cannot yet be
turned on from the command-line tools.
* src/ltlvisit/simplify.cc, doc/tl/tl.tex: Remove the rule.
* src/ltltest/eventuniv.test: Adjust.
* NEWS: Mention the bug.
* src/ltlvisit/simplify.cc: Remove two incorrect rules, and
partially disable another one.
* doc/tl/tl.tex: Reflect the change.
* src/ltltest/reduccmp.test: Likewise.
* src/ltltest/equals.cc: Add safety checks to catch such errors in the
future.
* NEWS: Mention the bug.
* src/taalgos/tgba2ta.cc: Do not assume the input is an sba.
* src/tgbatest/ltl2ta2.test: New file.
* src/tgbatest/Makefile.am: Add it.
* NEWS: Mention the fix.
Also generalize the degen-lcache option.
* src/tgbaalgos/postproc.cc, src/tgbaalgos/postproc.hh: Add the option.
* src/bin/spot-x.cc: Document it.
* src/tgbaalgos/degen.cc, src/tgbaalgos/degen.hh: Implement it.
* src/tgbatest/ltlcross2.test: Add a test configuration.
* src/tgbatest/degenlskip.test: New file.
* src/tgbatest/Makefile.am (TESTS): Add degenlskip.test.
* src/tgba/tgbamask.hh (build_tgba_mask_acc_ignore): New function.
(tgba_mask::wanted): Take an acc argument.
* src/tgba/tgbamask.cc: Implement the above.
* src/tgbatest/maskacc.cc, src/tgbatest/maskacc.test: New files.
* src/tgbatest/Makefile.am: Add them.
nsheap was an horror full of virtual functions required to
customize gtec to implement inclusion-based emptiness-check
in GreatSPN support. Since this support has been removed, we
can remove the nsheap cruft as well. Note that nsheap was
also used in emptinessta for no good reason (the code from
emptinessta was simply copied from gtec without cleanup).
* src/tgbaalgos/gtec/nsheap.cc, src/tgbaalgos/gtec/nsheap.hh:
Delete.
* src/tgbaalgos/gtec/Makefile.am: Adjust.
* src/taalgos/emptinessta.cc, src/taalgos/emptinessta.hh,
src/taalgos/tgba2ta.cc, src/tgbaalgos/gtec/ce.cc,
src/tgbaalgos/gtec/gtec.cc, src/tgbaalgos/gtec/gtec.hh,
src/tgbaalgos/gtec/status.cc, src/tgbaalgos/gtec/status.hh:
Use a simple unordered_map.
It hasn't been tested for several year, may not even compile, has
to be linked with source code that isn't even publicly available,
and its presence was the only reason to keep some inefficient
code in gtec.cc and friends.
* iface/gspn/: Delete this directory.
* iface/Makefile.am, configure.ac, README: Adjust.
* m4/gspnlib.m4: Delete.
* src/sanity/Makefile.am: Do not use LIBGSPN_CPPFLAGS.
This of course concerns push_back and push_front as well.
* src/bin/common_finput.cc, src/bin/dstar2tgba.cc, src/bin/ltl2tgba.cc,
src/bin/ltl2tgta.cc, src/bin/ltlcross.cc, src/bin/ltlfilt.cc,
src/dstarparse/dstarparse.yy, src/kripkeparse/kripkeparse.yy,
src/ltlast/formula.cc, src/ltlparse/ltlparse.yy, src/misc/minato.cc,
src/neverparse/neverclaimparse.yy, src/priv/bddalloc.cc, src/ta/ta.cc,
src/taalgos/emptinessta.cc, src/tgba/taatgba.cc,
src/tgbaalgos/gtec/gtec.cc, src/tgbaalgos/gtec/sccstack.cc,
src/tgbaalgos/magic.cc, src/tgbaalgos/ndfs_result.hxx,
src/tgbaalgos/rundotdec.cc, src/tgbaalgos/scc.cc, src/tgbaalgos/se05.cc,
src/tgbaalgos/simulation.cc, src/tgbaalgos/tau03.cc,
src/tgbaalgos/tau03opt.cc, src/tgbaparse/tgbaparse.yy: Use emplace
to make the code less verbose and avoid creating temporaries.
The returned Boolean indicates whether there is a successor or not.
This way
| for (i->first(); !i->done(); i->next())
| {
| ...
| }
can be replaced by
| if (i->first()) do
| {
| ...
| }
| while (i->next());
avoiding all the virtual calls to done().
* iface/dve2/dve2.cc, src/kripke/kripkeexplicit.cc,
src/kripke/kripkeexplicit.hh, src/ta/ta.hh, src/ta/taexplicit.cc,
src/ta/taexplicit.hh, src/ta/taproduct.cc, src/ta/taproduct.hh,
src/ta/tgtaproduct.cc, src/ta/tgtaproduct.hh, src/tgba/succiter.hh,
src/tgba/succiterconcrete.cc, src/tgba/succiterconcrete.hh,
src/tgba/taatgba.cc, src/tgba/taatgba.hh, src/tgba/tgba.hh,
src/tgba/tgbaexplicit.hh, src/tgba/tgbakvcomplement.cc,
src/tgba/tgbamask.cc, src/tgba/tgbaproduct.cc,
src/tgba/tgbasafracomplement.cc, src/tgba/tgbasgba.cc,
src/tgba/tgbatba.cc, src/tgba/tgbaunion.cc, src/tgba/tgbaunion.hh,
src/tgba/wdbacomp.cc: Implement and adjust to this new interface.
Instead of "delete iter;" we now do "aut->release_iter(iter);" to
give the iterator back to the automaton. The TGBA classes now
reuse a previously returned tgba_succ_iterator to answer a succ_iter()
call, therefore avoiding (1) memory allocation, as well as (2) vtable
and other constant member initialization.
* src/tgba/tgba.hh, src/tgba/tgba.cc (release_iter, iter_cache_):
Implement a release_iter() that stores the released iterator
in iter_cache_.
* src/tgba/succiter.hh (internal::succ_iterable): Move...
* src/tgba/tgba.hh (tgba::succ_iterable): ... here. And use
release_iter().
* iface/dve2/dve2.cc, src/kripke/kripke.cc, src/kripke/kripke.hh,
src/tgba/succiterconcrete.cc, src/tgba/succiterconcrete.hh,
src/tgba/taatgba.hh, src/tgba/tgbabddconcrete.cc,
src/tgba/tgbaexplicit.hh, src/tgba/tgbamask.cc, src/tgba/tgbaproduct.cc,
src/tgba/tgbaproxy.cc, src/tgba/tgbascc.cc, src/tgba/tgbatba.cc,
src/tgba/tgbaunion.cc, src/tgba/tgbaunion.hh, src/tgba/wdbacomp.cc,
src/tgbaalgos/bfssteps.cc, src/tgbaalgos/compsusp.cc,
src/tgbaalgos/cycles.cc, src/tgbaalgos/dtbasat.cc,
src/tgbaalgos/dtgbasat.cc, src/tgbaalgos/gtec/gtec.cc,
src/tgbaalgos/gv04.cc, src/tgbaalgos/isweakscc.cc,
src/tgbaalgos/lbtt.cc, src/tgbaalgos/ltl2tgba_fm.cc,
src/tgbaalgos/magic.cc, src/tgbaalgos/ndfs_result.hxx,
src/tgbaalgos/neverclaim.cc, src/tgbaalgos/reachiter.cc,
src/tgbaalgos/replayrun.cc, src/tgbaalgos/safety.cc,
src/tgbaalgos/scc.cc, src/tgbaalgos/se05.cc,
src/tgbaalgos/simulation.cc, src/tgbaalgos/tau03.cc,
src/tgbaalgos/tau03opt.cc: Use release_iter() instead of deleting
iterators, and used recycle iter_cache_ in implementations of
tgba::succ_iter().
* configure.ac: Enable C++11 mode.
* src/bdd.h: Use noexport, and add a move constructor and
move assignment operator. The move version of these method
do not have to increment the reference counter, saving time.
On a small test run, this change saved 24% of the calls to
bdd_addref_nc().
* src/ltlast/atomic_prop.cc, src/ltlast/atomic_prop.hh,
src/ltlast/automatop.cc, src/ltlast/automatop.hh, src/ltlast/binop.cc,
src/ltlast/binop.hh, src/ltlast/bunop.cc, src/ltlast/bunop.hh,
src/ltlast/multop.cc, src/ltlast/multop.hh, src/ltlast/unop.cc,
src/ltlast/unop.hh: Use std::tuple to replace nested std::pair,
simplify calls to std::map::erase, use auto and std::make_pair with
insert, and simplify the dump() method using a range for.
* src/misc/random.cc, src/misc/random.hh (srand, drand, mrand, rrand,
barand): Simplify using <random> from C++11.
(nrand, bmrand, prand): Remove these unused functions.
* src/tgbaalgos/randomgraph.cc: Adjust the use of barand.
* configure.ac: Do not check for srand48 and drand48.
The bitvect implementation seems a tad faster, but most importantly
this removes the last dependency on Boost.
* src/tgba/tgbasafracomplement.cc: Replace boost::dynamic_bitset by
spot::bitvect.
