def test_names():
auto = Automaton()
auto.add_all(NAMES)
auto.update_automaton()
auto_matches = [(m.start, m.end) for m in auto.get_matches(TEXT)]
with TemporaryDirectory() as tmpdir:
#tmpdir = ''
fnm = os.path.join(tmpdir, 'test.aca')
auto.save_to_file(fnm)
auto2 = Automaton()
auto2.load_from_file(fnm)
auto2_matches = [(m.start, m.end) for m in auto2.get_matches(TEXT)]
assert list(auto.items()) == list(auto2.items())
assert list(auto.prefixes()) == list(auto2.prefixes())
assert auto_matches == auto2_matches
auto3 = Automaton()
auto3.load_from_string(auto2.save_to_string())
auto3_matches = [(m.start, m.end) for m in auto3.get_matches(TEXT)]
assert list(auto.items()) == list(auto2.items())
assert list(auto.prefixes()) == list(auto2.prefixes())
assert auto_matches == auto3_matches
def __init__(self, service_type, label_conf_dict_path):
self.__level1_keywords__ = []
self.__level1_tag__ = []
self.__level1_automaton__ = []
label_file = "%s.rule.dat" % (service_type)
level1_keywords_map_file_path = os.path.join(label_conf_dict_path,
label_file)
with open(level1_keywords_map_file_path) as level1_f:
for line in level1_f:
if not line:
continue
line = line.decode("utf-8").strip()
if not line:
continue
if line.startswith("#"):
continue
line_arr = line.split(":")
keywords_list = line_arr[0].split(" ")
level1_list = line_arr[1].split(",")
self.__level1_keywords__.append(keywords_list)
self.__level1_tag__.append(level1_list)
automaton = Automaton()
automaton.add_all(keywords_list)
self.__level1_automaton__.append(automaton)
def test_items():
auto = Automaton()
auto.add_all(names)
ens, evs = zip(*sorted(names))
ns, vs = zip(*list(auto.items()))
ns = [''.join(n) for n in ns]
assert list(ens) == list(ns)
assert list(evs) == list(vs)
def test_names():
auto = Automaton(NAMES)
auto.add_all(NAMES)
print (auto)
matches = set(' '.join(match.elems) for match in auto.get_matches(TEXT.split()))
names = set(' '.join(name) for name in NAMES)
assert names == matches
# -*- coding: utf-8 -*-
from __future__ import unicode_literals, print_function, absolute_import
# create a new AC automaton
from aca import Automaton
automaton = Automaton()
# instead of plain strings, you can also use lists of tokens
names = [
(['Yuri', 'Artyukhin'], 'developer'),
(['Tom', 'Anderson', 'Jr'], 'designer'),
]
automaton.add_all(names)
# you can add an item like this as well
automaton[['Tom', 'Anderson']] = 'manager'
# if you are not using plain strings, make sure you tokenize the text as well
text = 'Tom Anderson Jr and Yuri Artyukhin work on my project'.split()
print('matches that maximize the number of matched words')
for match in automaton.get_matches(text):
print(match.start, match.end, match.elems, match.label)
print('all matches')
for match in automaton.get_matches(text, exclude_overlaps=False):
print(match.start, match.end, match.elems, match.label)
# Import the library and initiate the automaton
from aca import Automaton
automaton = Automaton()
# add the entities and build the automaton
automaton.add_all(['Funderbeam', 'Funderbeam Data', 'Funderbeam Markets'])
automaton.update_automaton()
# find matches
text = 'Funderbeam Data and Funderbeam Markets are two different products of Funderbeam'
for match in automaton.get_matches(text, exclude_overlaps=False):
print(match.start, match.end, match.elems)
for match in automaton.get_matches(text, exclude_overlaps=True):
print(match.start, match.end, match.elems)
# -*- coding: utf-8 -*-
from __future__ import unicode_literals, print_function, absolute_import
# create a new AC automaton
from aca import Automaton
automaton = Automaton()
# add a dictionary of words to the automaton
painkillers = ['paracetamol', 'ibuprofen', 'hydrocloride']
automaton.add_all(painkillers)
# match the dictionary on a text
text = 'paracetamol and hydrocloride are a medications to relieve pain and fever. paracetamol is less efficient than ibuprofen'
for match in automaton.get_matches(text):
print(match.start, match.end, match.elems)
def test_prefixes():
auto = Automaton()
auto.add_all(['jaanus', 'janek', 'janis'])
prefixes, values = zip(*auto.prefixes())
prefixes = [''.join(prefix) for prefix in prefixes]
assert prefixes == ['', 'j', 'ja', 'jaa', 'jaan', 'jaanu', 'jaanus', 'jan', 'jane', 'janek', 'jani', 'janis']
