def visit_repetition(self, repetition):
"""
Converts a repeated regular exprsesion into an equivalent NFA.
Pushes the NFA onto a stack.
@param repetition: a Regex.Repetition object representing the repeated expression to convert.
"""
state_machine = None
repetition.child.accept(self)
child_state_machine = self.state_machines.pop()
# Create optional 0 to max-min
repetition_min = repetition.min
if repetition.max == Regex.Repetition.Infinity:
# Kleene star
inner_state_machine = copy.deepcopy(child_state_machine)
state_machine = NonDeterministicFinite()
state_machine.start_state.epsilon_edges.add(
inner_state_machine.start_state)
inner_state_machine.end_state.epsilon_edges.add(
state_machine.end_state)
inner_state_machine.end_state.epsilon_edges.add(
inner_state_machine.start_state)
if repetition.min == 0:
state_machine.start_state.epsilon_edges.add(
state_machine.end_state)
else:
repetition_min -= 1
elif repetition.max > repetition_min:
state_machines = [
copy.deepcopy(child_state_machine)
for i in xrange(repetition.max - repetition_min)
]
for state_machine in state_machines:
state_machine.start_state.epsilon_edges.add(
state_machine.end_state)
state_machine = NonDeterministicFinite.concatenate(state_machines)
# Prepend minimal repetition
if repetition_min > 0:
head_state_machines = [
copy.deepcopy(child_state_machine)
for i in xrange(repetition_min)
]
head_state_machine = NonDeterministicFinite.concatenate(
head_state_machines)
if state_machine is None:
state_machine = head_state_machine
else:
state_machine = NonDeterministicFinite.concatenate(
[head_state_machine, state_machine])
self.state_machines.append(state_machine)
def visit_literal(self, literal):
"""
Converts a regular expression literal into an NFA with a transition for each contained character.
Pushes the NFA onto a stack.
@param literal: a Regex.Literal object representing the literal to convert.
"""
state_machine = NonDeterministicFinite()
state_machine.start_state.edges[state_machine.end_state].update(
literal.characters)
self.state_machines.append(state_machine)
def visit_alternation(self, alternation):
"""
Converts an alternated (separated by "|") regular expressions into an equivalent NFA graph.
Pushes the graph onto a stack.
@param alternation: a Regex.Alternation object representing the set of alternated expressions.
"""
child_state_machines = []
for child in alternation.children:
child.accept(self)
child_state_machines.append(self.state_machines.pop())
self.state_machines.append(NonDeterministicFinite.alternate(child_state_machines))
def visit_alternation(self, alternation):
"""
Converts an alternated (separated by "|") regular expressions into an equivalent NFA graph.
Pushes the graph onto a stack.
@param alternation: a Regex.Alternation object representing the set of alternated expressions.
"""
child_state_machines = []
for child in alternation.children:
child.accept(self)
child_state_machines.append(self.state_machines.pop())
self.state_machines.append(
NonDeterministicFinite.alternate(child_state_machines))
def visit_concatenation(self, concatenation):
"""
Converts a concatenated set of regular expressions into an equivalent NFA graph.
Pushes the graph onto a stack.
@param concatenation: a Regex.Concatenation object representing the set of concatenated expressions.
"""
child_state_machines = []
for child in concatenation.children:
child.accept(self)
child_state_machines.append(self.state_machines.pop())
self.state_machines.append(NonDeterministicFinite.concatenate(child_state_machines))
def visit_repetition(self, repetition):
"""
Converts a repeated regular exprsesion into an equivalent NFA.
Pushes the NFA onto a stack.
@param repetition: a Regex.Repetition object representing the repeated expression to convert.
"""
state_machine = None
repetition.child.accept(self)
child_state_machine = self.state_machines.pop()
# Create optional 0 to max-min
repetition_min = repetition.min
if repetition.max == Regex.Repetition.Infinity:
# Kleene star
inner_state_machine = copy.deepcopy(child_state_machine)
state_machine = NonDeterministicFinite()
state_machine.start_state.epsilon_edges.add(inner_state_machine.start_state)
inner_state_machine.end_state.epsilon_edges.add(state_machine.end_state)
inner_state_machine.end_state.epsilon_edges.add(inner_state_machine.start_state)
if repetition.min == 0:
state_machine.start_state.epsilon_edges.add(state_machine.end_state)
else:
repetition_min -= 1
elif repetition.max > repetition_min:
state_machines = [copy.deepcopy(child_state_machine) for i in xrange(repetition.max - repetition_min)]
for state_machine in state_machines:
state_machine.start_state.epsilon_edges.add(state_machine.end_state)
state_machine = NonDeterministicFinite.concatenate(state_machines)
# Prepend minimal repetition
if repetition_min > 0:
head_state_machines = [copy.deepcopy(child_state_machine) for i in xrange(repetition_min)]
head_state_machine = NonDeterministicFinite.concatenate(head_state_machines)
if state_machine is None:
state_machine = head_state_machine
else:
state_machine = NonDeterministicFinite.concatenate([head_state_machine, state_machine])
self.state_machines.append(state_machine)
def visit_literal_except(self, literal_except):
"""
Converts a regular expression inverse literal ([^...]) into an equivalent NFA.
Pushes the NFA onto a stack.
@param literal_except: a Regex.LiteralExcept object representing the inverse literal to convert.
"""
state_machine = NonDeterministicFinite()
state_machine.start_state.edges[state_machine.end_state].add(
1, 0x10FFFF)
state_machine.start_state.edges[
state_machine.end_state].difference_update(
literal_except.characters)
self.state_machines.append(state_machine)
def visit_concatenation(self, concatenation):
"""
Converts a concatenated set of regular expressions into an equivalent NFA graph.
Pushes the graph onto a stack.
@param concatenation: a Regex.Concatenation object representing the set of concatenated expressions.
"""
child_state_machines = []
for child in concatenation.children:
child.accept(self)
child_state_machines.append(self.state_machines.pop())
self.state_machines.append(
NonDeterministicFinite.concatenate(child_state_machines))
