def _get_expression(self):
"""Gets the expression and symbols of current state.
Returns:
expression: String.
symbols: List of symbols.
"""
symbols = postprocessor.production_rules_sequence_to_symbols(
prod_rules_sequence=self.production_rules_sequence)
return ' '.join([str(symbol) for symbol in symbols]), symbols
def test_production_rules_sequence_to_symbols_not_terminal(self):
prod_rule_strings = [
'S -> S "+" T',
'S -> T',
]
prod_rules_sequence = [
self.prod_rules_dict[prod_rule_string]
for prod_rule_string in prod_rule_strings
]
# Parsing tree:
# S
# |
# S "+" T
# |
# T
# Expression (non-terminal):
# T + T
t, = nltk.grammar.nonterminals('T')
self.assertEqual(
postprocessor.production_rules_sequence_to_symbols(
prod_rules_sequence), [t, '+', t])
def test_production_rules_sequence_to_symbols_terminal(self):
prod_rule_strings = [
'S -> S "+" T',
'S -> T',
'T -> "x"',
'T -> "x"',
# The generation of symbols by grammar production rules sequence will
# stop if all the symbols are terminal. For the grammar rules in this
# unittest, the last two dummy rules are actually not used.
constants.DUMMY_PRODUCTION_RULE,
constants.DUMMY_PRODUCTION_RULE,
]
prod_rules_sequence = [
self.prod_rules_dict[prod_rule_string]
for prod_rule_string in prod_rule_strings
]
# Expression:
# x + x
# Parsing tree:
# S
# |
# S "+" T
# | |
# T "x"
# |
# "x"
# Production rules sequence (preorder tree traversal)
# 'S -> S "+" T'
# 'S -> T'
# 'T -> "x"
# 'T -> "x"
self.assertEqual(
postprocessor.production_rules_sequence_to_symbols(
prod_rules_sequence), ['x', '+', 'x'])
def test_production_rules_sequence_to_symbols_empty(self):
# Empty production rule sequence should have empty symbols.
self.assertEqual(
postprocessor.production_rules_sequence_to_symbols([]), [])