Operator &= used to always move Fin to the front, it does not anymore.
The only thing it does now is to merge Inf(x)&Inf(y) as Inf({x,y}).
Operator |= is now symmetrical and merges Fin()s.
Fixes#253.
* spot/twa/acc.cc, spot/twa/acc.hh: Simplify &= and make |= symmetrical.
* spot/twaalgos/cleanacc.cc: Fix conjunction order.
* tests/core/acc.test, tests/core/acc2.test, tests/core/parseaut.test,
tests/core/readsave.test, tests/core/satmin2.test,
tests/core/sccdot.test, tests/python/acc_cond.ipynb,
tests/python/accparse.ipynb, tests/python/automata.ipynb,
tests/python/product.ipynb, tests/python/randaut.ipynb: Adjust test
cases.
* spot/twaalgos/tra2tba.cc: Support Rabin like input
* tests/core/tra2tba.cc: Remove C tests
* tests/core/tra2tba.test: Remove C tests
* tests/python/tra2tba.py: Convert C tests to python
* tests/Makefile.am: Remove C tests and add python tests
* python/spot/impl.i: Add bindings for tra2tba
* spot/twaalgos/Makefile.am: Record tra2tba.cc, tra2tba.hh
* spot/twaalgos/tra2tba.cc: Implement transformation of TRA to TBA
* spot/twaalgos/tra2tba.hh: Introduce declaration of tra_to_tba
* tests/Makefile.am: Record tra2tba tests
* tests/core/tra2tba.cc: Add driver for tests
* tests/core/tra2tba.test: Add tests of tra2tba transformation
These two functions were doing almost identical work, the only
difference was the way to select the SCC to keep. Now we have a more
uniform way to do that. Closes#172.
* bin/autfilt.cc: Offer a unique --decompose-scc option, but keep
--decompose-strength as an alias for backward compatibility.
* spot/twaalgos/strength.cc, spot/twaalgos/strength.hh: Rename
decompose_strength as decompose_scc, and handle a way to list
all SCC numers in the string specifier. This gets rid
of the nearly identical
* tests/core/scc.test, tests/core/strength.test,
tests/python/decompose.ipynb, tests/python/decompose_scc.py: Adjust
test cases.
* NEWS: Adjust.
Fixes#271, reported by Henrich Lauko.
* spot/twa/acc.cc (maybe_accepting): Fix handling in case
of disjunction of Fin.
* tests/core/scc.test, tests/python/accparse2.py: Add more
tests.
Fixes#269.
* spot/twaalgos/stats.cc: Use twa_statistics instead of
twa_sub_statistics when %t is not used.
* bin/common_aoutput.cc: Likewise, also fix %S to use twa_statistics
instead of num_states(), and document that %s,%t,%e all return
statistics about the reachable part of the automaton.
* tests/core/format.test: Add more tests.
* NEWS: Document the issue.
* NEWS: Mention the modification.
* spot/twaalgos/remfin.cc: Adapt to avoid infinite recursion.
* spot/twaalgos/totgba.cc: Work on Streett-like.
* tests/Makefile.am, tests/python/streett_totgba.py: Tests the
modification.
* tests/core/remfin.test: Fix one test case that is now handled by
the modification.
Fixes#262 again. Reported by Maximilien Colange.
* spot/twaalgos/simulation.cc: Use state numbers to order classes, not
their signatures. The problem was that even if two simulation of the
same automaton assign the same signature, the BDD identifier used for
that signature might be different, and therefore the ordering obtained
by using BDDs as keys in a map can be different. A side-effect of
this change is that the order of states in automata produced by
simulation-based reduction may change; many tests had to be updated.
* tests/core/ltl2tgba.test: Add a new test case based on Maximilien's
report.
* tests/core/complement.test, tests/core/det.test,
tests/core/parseaut.test, tests/core/prodor.test, tests/core/scc.test,
tests/python/atva16-fig2a.ipynb, tests/python/automata.ipynb,
tests/python/decompose.ipynb, tests/python/decompose_scc.py,
tests/python/highlighting.ipynb, tests/python/piperead.ipynb,
tests/python/sccinfo.py, tests/python/simstate.py,
tests/python/testingaut.ipynb, tests/python/word.ipynb: Update
test case for new order of states.
Related to #208.
* spot/twaalgos/dot.cc: Fix missing definitions of universal nodes,
and inclusion of universal nodes inside of SCC when none of the
destination comes back to the SCC.
* tests/python/_altscc.ipynb: Adjust and add more test cases.
* tests/core/alternating.test, tests/core/neverclaimread.test,
tests/core/readsave.test, tests/core/sccdot.test,
tests/python/decompose.ipynb: Adjust test cases.
* NEWS: Mention the bug.
* spot/twaalgos/sccinfo.hh, spot/twaalgos/sccinfo.cc: Take
a filter function as optional argument.
* tests/core/sccif.cc, tests/core/sccif.test: New files.
* tests/Makefile.am, tests/core/.gitignore: Adjust.
* NEWS: Mention the new feature.
Fixes#262, reported by Maximilien Colange.
* bin/common_output.cc, bin/common_aoutput.cc, bin/common_aoutput.hh:
Clear the set of atomic propositions if --stats=%[...]x was used.
* spot/twa/bdddict.cc: Release any formula associated to a BDD when it
is unregistered, do not wait for the dictionary's destruction. This
was the main culprit for #262.
* tests/core/ltl2tgba.test: Add test cases.
* NEWS: Mention the bug.
The cache used in formula simplification will keep atomic propositions
defined between several translations, and may impact variable order.
Reported by Maximilien Colange.
* spot/tl/simplify.hh, spot/tl/simplify.cc,
spot/twaalgos/translate.cc, spot/twaalgos/translate.hh (clear_cache):
New method.
* bin/ltl2tgba.cc, bin/ltl2tgta.cc: Call it.
* spot/twaalgos/stats.cc: Do not keep a point to the formula after
printing statistics.
* tests/core/ltl2tgba.test: Add a test case.
* tests/core/readsave.test: Adjust one formula.
* NEWS: Mention the issue.
Fixes#260. Reported by František Blahoudek.
The simplification F(f)|q = F(f|q), where q designates an event_univ
formula, was not always applied because of a couple of issue: (1) the
mospliter was ignoring event_univ unless favor_event_univ was set, (2)
when processing formulas from res_EventUniv they were not put back
into res_F or res_G to be subject to the F/G rules.
* spot/tl/simplify.cc: Improve handling of the above points.
* tests/core/reduccmp.test: Adjust and add test case.
* tests/core/ltl2tgba2.test, tests/python/atva16-fig2a.ipynb: Adjust.
Fixes#255.
* spot/twaalgos/split.cc, spot/twaalgos/split.hh,
tests/core/split.test: New files.
* spot/twaalgos/Makefile.am, tests/Makefile.am: Add them.
* bin/autfilt.cc (--split-edges): New option.
* python/spot/impl.i: Process split.hh.
* tests/python/alternating.py: Test split_edges() on
an alternating automaton.
Fixes#257.
* spot/twaalgos/degen.cc: Implement this optimization.
* tests/core/degendet.test: Add a few test cases.
* tests/core/ltl2tgba2.test: Adjust expected sizes.
* NEWS: Mention the change.
Fixes#233, although more cleanup would be welcome.
* spot/twaalgos/minimize.cc: Replace the uses of twa methods by
twa_graph methods, and simplify some data structures.
* tests/core/acc_word.test, tests/core/readsave.test,
tests/python/automata.ipynb: Adjust changed output due
to different data structures.
* bin/genaut.cc: Use RANGE instead of N, and document it.
Output the FORMAT documentation, and fix handling of %F and %L.
* tests/core/genaut.test: New file.
* tests/Makefile.am: Add it.
* spot/twaalgos/degen.cc, spot/twaalgos/degen.hh,
spot/twaalgos/mask.hh: Store original states in "original-states"
properties.
* spot/twaalgos/dot.cc: Add support for option 'd'.
* bin/common_aoutput.cc: Document it.
* doc/org/concepts.org, NEWS: Document "original-states".
* tests/core/readsave.test: Add some tests.
Reported by Thibaud Michaud
* spot/twaalgos/sbacc.cc: Do not label rejecting SCCs with the empty
mark, as it might be accepting.
* tests/core/sbacc.test: Add test cases.
* NEWS: Mention the bug.
* NEWS: Mention this addition.
* bin/autfilt.cc: Add dualize option
* tests/Makefile.am: Add dualize option test file to the suite.
* tests/core/dualize.test: Test the dualize option.
when fin_alone sets where presents (i.e., not really Rabin condition),
the rabin_to_buchi_maybe() could fail to notice DBA-typeness.
* spot/twaalgos/remfin.cc: Don't set scc_ba_type to false when
fin_alone is present.
* tests/core/remfin.test: Add a test case.
* spot/twaalgos/sum.cc: Fix the sum of automatas having universal
initial transitions.
* tests/core/explsum.test: Add test case testing the handling of
universal initial transitions in sum.
The algorithm had two problems: it was removing only useless
destination from universal destination (instead of removing the entire
edge), and it was not properly iterating over the entire reachable
automaton.
* spot/twa/twagraph.cc: Fix it.
* spot/twa/twagraph.hh: Adjust documentation.
* tests/core/alternating.test: Add more tests.
* tests/python/twagraph.py: Adjust.
* NEWS: Mention the bug.
Fixes#245.
* bin/genltl.cc: Add the option.
* bin/man/genltl.x: Add reference.
* tests/core/ltl2tgba2.test: Use these patterns.
* doc/org/genltl.org, NEWS: Document the options.
Fixes#231.
* NEWS: Mention of implementation of sum, sum_and.
* bin/autfilt.cc: Add --sum, --sum-or and --sum-and options.
* python/spot/impl.i: Add bindings for sum, sum_and.
* spot/twaalgos/Makefile.am: Add sum.cc, sum.hh.
* spot/twaalgos/sum.cc: Implement sum, sum_and.
* spot/twaalgos/sum.hh: Declaration of sum, sum_and.
* tests/Makefile.am: Add sum tests.
* tests/core/explsum.test: Test the sum of two automatons,
false or false, unsatisfied mark propagation, handling of univ.
transitions.
* tests/python/sum.py: Check that two automatons that does not
share their bdd dict are not accepted, then run tests over the
sum of randomly generated LTL formulas.
* spot/twaalgos/degen.cc: Make use_cust_acc_orders as unlikely, as it
was never true in the whole test suite before this patch.
* tests/core/ltlcross2.test: Play with degen-order and degen-lcache
options to test more cases.
