def test():
input = '(Ga -> Gb) W c'
spot.setup()
f = spot.formula(input)
a = f.translate('tgba')
output = a.to_str('dot')
return output
def get_automaton():
input = request.forms.get('query')
spot.setup()
f = spot.formula(input)
a = f.translate('monitor')
output = a.to_str('dot')
return output
# -*- coding: utf-8 -*-
# +
import spot
spot.setup()
from spot.jupyter import display_inline
from math import inf
from fimdp.labeled import DBAWrapper, LabeledConsMDP
from fimdp.energy_solvers import BasicES
from fimdp.objectives import BUCHI
# -
# ### Test DBAWrapper
f = spot.formula("GF b & GF a")
aut = spot.translate(f, "BA", "deterministic", "complete")
aut
# +
AP = ["a", "b"]
wrapper = DBAWrapper(aut, AP)
a = AP.index("a")
b = AP.index("b")
assert wrapper.succ(2, [a]) == 1
assert wrapper.succ(2, [a, b]) == 2
assert wrapper.succ(2, [b]) == 0
assert wrapper.succ(2, []) == 1
def ltl2oa(self,ltl):
"""Constructs and returns dictionaries and lists containing the specifications of an OA obtained by translation from the ltl property.
It parses the output of ltl2ldba or ltl2dra for the ltl formula and creates a objects that store the specification of the OA.
Parameters
----------
ltl : str
The linear temporal logic (LTL) formula to be transformed to a OA.
Returns
-------
out : (q0, delta, acc, eps, shape, spot_oa)
The tuple of the initial state q0, the list of dictionaries of transitions delta,
the list of dictionaries of the accepting transitions, the list of lists of epsilon-moves,
the pair of the number of the Rabin pairs and the number of states and the spot object of the OA.
"""
# Translate the LTL formula to an OA using Rabinizer 4.
out=check_output(['ltl2ldba', '-d', '-e', ltl] if self.oa_type == 'ldba' else ['ltl2dra', '-c', ltl])
# Split the output into two parts: the header and the body
header, body = out.decode('utf-8').split('--BODY--\n')
# Parse the initial state, the atomic propositions and the number of Rabin pairs
for line in header.splitlines():
if line.startswith('Start'):
q0 = int(line[7:]) # The initial state
elif line.startswith('AP'):
char_map = {i:c for i,c in enumerate(re.sub("[^\w]", " ", line[4:]).split()[1:])} # Maps ids to atomic propositions
ap_list = [tuple(ap) for ap in self.powerset(sorted(char_map.values()))] # The list of all subsets of AP.
elif line.startswith('Acceptance'):
n_pairs = int(line.split()[1])//2 # Zero for the Buchi condition
body_lines = body.splitlines()[:-1] # Ignore the last line
# Get the number of states
n_qs = 0 # The number of states
for line in reversed(body_lines): # Loop over all states because the states might not be ordered.
if line.startswith('State'):
n_qs = max(int(line[7:]),n_qs) # Get the maximum of them
n_qs += 2 # +1 because the index origin is 0 and +1 for the trap state
n_i = max(1,n_pairs) # Because n_pairs is zero for the Buchi condition
shape = n_i, n_qs
# The transition function delta[q][label] stores the next state The OA makes a transition when the it consumes 'label' at state 'q'.
delta = [{ap:n_qs-1 for ap in ap_list} for i in range(n_qs)] # The default target of a transition is the trap state whose index is n_qs-1
acc = [{ap:[None]*n_i for ap in ap_list} for i in range(n_qs)] # The default acceptance value is None, meaning the transition does not belong to any acceptance set.
eps = [[] for i in range(n_qs)] # The epsilon moves in the OA. eps[q] is the list of states can be reached from `q` by making an epsilon-transition.
# Parse the transitions, acceptance values
q=-1 # The state to be parsed
for line in body_lines:
if line.startswith('State'):
q = int(line[7:]) # Update the state to be parsed
else:
# Parse the transition into three parts
_, _label, _dst, _, _acc_set = re.findall('(\[(.*)\])? ?(\d+) ?(\{(.*)\})?',line)[0]
dst = int(_dst) # Get the destination
if not _label: # If there is no label then the transition is an epsilon-move
eps[q].append(dst)
else:
# Get the acceptance status of the transition
acc_set = set([int(a) for a in _acc_set.split()]) # The set of acceptance states that the transition belongs to
if not n_pairs: # acc_name == 'Buchi':
t_acc = [True if 0 in acc_set else None] # If it is an Buchi set, then it is True and None otherwise
else:
t_acc = [None]*n_pairs
for i in range(n_pairs): # For each Rabin pairs
if 2*i+1 in acc_set:
t_acc[i] = True # True if it belongs to the second set of the Rabin pair
if 2*i in acc_set:
t_acc[i] = False # False if it belongs to the first set of the Rabin pair
labels = ['']
_labels = re.compile('[()]').split(_label) # The transitions might have subformulas
for _l in _labels:
labels = [l+_ll for l in labels for _ll in _l.split('|')] # Add all the combinations
for label in labels:
if label == 't': # Means all the transitions
label_acc, label_rej = set(()), set(())
else:
ls = list(filter(None,re.compile('[\s&]').split(label))) # Get the atoms
label_acc = set([char_map[int(l)] for l in ls if not l.startswith('!')]) # Transitions having these atoms
label_rej = set([char_map[int(l[1:])] for l in ls if l.startswith('!')]) # Transitions that doesn't have these
for ap in delta[q]: # Find all the matching transitions
# If matches, update the transition properties
if not(label_acc-set(ap)) and (label_rej-set(ap))==label_rej:
delta[q][ap] = dst
acc[q][ap] = t_acc
# Construct a spot object for visualization
if spot:
filename = self.random_hoa_filename()
with open(filename,'wb') as f:
f.write(check_output(['ltl2ldba', '-d', ltl] if self.oa_type == 'ldba' else ['ltl2dra', '-c', ltl]))
spot.setup()
spot_oa = spot.automaton(filename)
spot_oa.merge_edges() # For better visualization
os.remove(filename)
else:
spot_oa=None
return q0, delta, acc, eps, shape, spot_oa
