def __preprocess_sentence(sentence):
"""
Preprocesses a sentence for the grammar parser. Words are concatenated with their adjoining punctuation marks. This method provides heuristics to recognize those entities prior to processing.
:param str sentence: The sentence to preprocess.
:rtype list(str): The list of proper nouns in the pre-processed sentence and fillers to replace removed, recognized entities without altering the grammatical structure.
"""
split_sentence = re.split(r"\s+", sentence)
list_indicies = utils.comma_delimited_list_indicies(split_sentence)
if list_indicies:
sentences = utils.enumerate_sentence_with_list(split_sentence, list_indicies)
return None
def test_enumerate_sentence_with_list(self):
# first_indicies = utils.comma_delimited_list_indicies( self.sentence_with_list )
second_indicies = utils.comma_delimited_list_indicies(self.sentence_with_two_lists)
# print utils.enumerate_sentence_with_list( self.sentence_with_list, first_indicies )
sens = utils.enumerate_sentence_with_list(self.sentence_with_two_lists, second_indicies)
eq_(
sens[0],
[
"This",
"is",
"a",
"sentence",
"listing",
"vestiges",
"and",
"boolean,",
"but",
"it",
"also",
"lists",
"the",
"hats",
"and",
"gloves.",
],
)
eq_(
sens[1],
[
"This",
"is",
"a",
"sentence",
"listing",
"aardvarks",
"and",
"boolean,",
"but",
"it",
"also",
"lists",
"the",
"shoes",
"and",
"gloves.",
],
)
eq_(
sens[2],
[
"This",
"is",
"a",
"sentence",
"listing",
"igloos",
"and",
"boolean,",
"but",
"it",
"also",
"lists",
"the",
"carribean",
"and",
"gloves.",
],
)
def test_comma_delimited_list_indicies(self):
first_indicies = utils.comma_delimited_list_indicies(self.sentence_with_list)
second_indicies = utils.comma_delimited_list_indicies(self.sentence_with_two_lists)
eq_(first_indicies[0], set([5, 6, 7]))
eq_(second_indicies[0], set([5, 6, 7]))
eq_(second_indicies[1], set([16, 17, 15]))
