def parse(self, tokens):
""" Parse a list of tokens and return the possible tree based on the
``grammar of this ``Parser``
Args:
tokens: a string
"""
if len(tokens) == 0:
raise ValueError("Tokens can't be empy")
# Tokenize if needed
if not isinstance(tokens, list):
tokens = tokenize(tokens)
root_tree = Tree(self._grammar.start(), [])
frontier = [()]
return self._parse(tokens, root_tree, frontier)
def test_removes_dot_at_end_of_sentence(self):
res = ["fall", "leaves", "fall"]
self.string += "."
self.assertEqual(res,
tokenize(self.string),
msg="It should remove the dot")
def test_returns_list_of_words(self):
res = ["fall", "leaves", "fall"]
self.assertEqual(res,
tokenize(self.string),
msg="It should return list of words")
def test_returns_a_list(self):
self.assertIsInstance(tokenize(self.string),
list,
msg="It should return a ``list``")
def test_returns_empty_list(self):
self.assertIsInstance(tokenize(""), list)
self.assertEqual(tokenize(""), [])
VP -> V NP | V | V NP NP
N -> 'fall' | 'spring' | 'leaves' | 'dog' | 'cat'
V -> 'spring' | 'leaves' | 'fall' | 'left'
D -> 'the'
C -> 'and'
Adj -> 'fall' | 'spring' | 'purple' | 'left'
"""
sentences = [
"Fall leaves fall.",
"Fall leaves fall and spring leaves spring.",
"The fall leaves left.",
"The purple dog left",
"The dog and cat left"
]
grammar = Grammar.parse_grammar(grammar_as_string)
parser = TopDownParser(grammar)
tokens = tokenize(sentences[4])
for sentence in sentences:
tokens = tokenize(sentence)
parse = parser.parse(tokens)
results = [p for p in parse]
print("==========================")
print(sentence)
print("--------------------------")
for index, parse in enumerate(results):
print("Parse #%d:\n%s" % (index, parse))
print("--------------------------")
print("Count: %d" % len(results))