def test_algorithm_z(self):
text_1 = "#baabab"
text_2 = "#ababaa"
fr = FourRussiansHelpers(lcs_delete_cost_function,
lcs_insert_cost_function,
substitute_cost_function2)
m, A, step_size_bound, text_1, text_2 = fr.prepare_parameters(text_1,
text_2)
storage = fr.algorithm_y(m, A, step_size_bound)
P, Q = fr.algorithm_z(m, storage, text_1, text_2)
_, diff_between_rows, diff_between_columns = get_full_matrices(
fr, text_1, text_2,
lcs_delete_cost_function,
lcs_insert_cost_function)
for i in range(1, 4):
for j in range(1, 4):
self.assertEqual(P[i][j],
[diff_between_rows[(i - 1) * m + 1][j * m],
diff_between_rows[i * m][j * m]])
self.assertEqual(Q[i][j],
[diff_between_columns[i * m][(j - 1) * m + 1],
diff_between_columns[i * m][j * m]])
def test_edit_distance(self):
text_1 = "#baabab"
text_2 = "#ababaa"
fr = FourRussiansHelpers(lcs_delete_cost_function,
lcs_insert_cost_function,
substitute_cost_function2)
m, A, step_size_bound, text_1, text_2 = fr.prepare_parameters(text_1,
text_2)
storage = fr.algorithm_y(m, A, step_size_bound)
edit_distance = fr.get_edit_distance(m, text_1, text_2, storage)
self.assertEqual(edit_distance, 4)
def four_russians(text_1, text_2):
""" Calculates longest common subsequence of strings A and B
as a special case of four russians edit distance algorithm """
fr = FourRussiansHelpers(lcs_delete_cost_function,
lcs_insert_cost_function,
lcs_substitute_cost_function)
m, A, step_size_bound, text_1, text_2 = fr.prepare_parameters(
text_1, text_2)
storage = fr.algorithm_y(m, A, step_size_bound)
lcs, length = fr.get_lcs(m, text_1, text_2, storage)
return lcs, length
def test_restore_matrix(self):
fr = FourRussiansHelpers(lcs_delete_cost_function,
lcs_insert_cost_function,
substitute_cost_function2)
restored_matrix = fr.restore_matrix(
"#baabab", "#ababaa",
[0, 1, 2, 3, 4, 5, 6], [0, 1, 2, 3, 4, 5, 6], 6)
expected_matrix = [[0, 1, 2, 3, 4, 5, 6],
[1, 2, 1, 2, 3, 4, 5],
[2, 1, 2, 1, 2, 3, 4],
[3, 2, 2, 2, 2, 2, 3],
[4, 3, 2, 3, 2, 3, 4],
[5, 4, 3, 2, 3, 2, 3],
[6, 5, 4, 3, 2, 3, 4]]
self.assertEqual(restored_matrix, expected_matrix)
def edit_distance_four_russians(
text_1,
text_2,
delete_cost_function=lcs_delete_cost_function,
insert_cost_function=lcs_insert_cost_function,
substitute_cost_function=lcs_substitute_cost_function):
""" Algorithm proposed by William J. Masek and Michael S. Paterson,
using the method of "Four Russians """
fr = FourRussiansHelpers(delete_cost_function, insert_cost_function,
substitute_cost_function)
m, A, step_size_bound, text_1, text_2 = fr.prepare_parameters(
text_1, text_2)
storage = fr.algorithm_y(m, A, step_size_bound)
cost = fr.get_edit_distance(m, text_1, text_2, storage)
return cost
def test_get_all_strings(self):
fr = FourRussiansHelpers(lcs_delete_cost_function,
lcs_insert_cost_function,
lcs_substitute_cost_function)
self.check_get_all_strings(fr, 2, ['a', 'b'], '#',
["#aa", "#ab", "#ba", "#bb"])
self.check_get_all_strings(fr, 3, ['z'], '?',
["?zzz"])
def test_storage(self):
fr = FourRussiansHelpers(lcs_delete_cost_function,
lcs_insert_cost_function,
lcs_substitute_cost_function)
storage = {}
self.check_storage(fr, [0, -1], [1, 0], "ab", "ba",
[0, 1], [0, -1], storage)
self.check_storage(fr, [-10, -100], [1, 11], "ababa", "ba",
[1, 1], [1, -1], storage)
self.check_storage(fr, [0, -1], [1, 11], "ababa", "ba",
[1, 1], [1, -1], storage)
self.check_storage(fr, [0, -1], [1, 0], "ab", "ba",
[0, 1], [0, -1], storage)
def test_prepare_parameters(self):
fr = FourRussiansHelpers(lcs_delete_cost_function,
lcs_insert_cost_function,
lcs_substitute_cost_function)
self.check_prepare_parameters(fr, "#abab", "#abab",
2, ['a', 'b'],
"#abab", "#abab")
self.check_prepare_parameters(fr, "#ababa", "#abc",
2, ['a', 'b', 'c'],
"#ababa#", "#abc#")
self.check_prepare_parameters(fr, "#11111111", "#000",
3, ['0', '1'],
"#11111111#", "#000")
def test_algorithm_y(self):
text_1 = "#baabab"
text_2 = "#ababaa"
fr = FourRussiansHelpers(lcs_delete_cost_function,
lcs_insert_cost_function,
substitute_cost_function2)
m, A, step_size_bound, text_1, text_2 = fr.prepare_parameters(text_1,
text_2)
storage = fr.algorithm_y(m, A, step_size_bound)
_, diff_between_rows, diff_between_columns = get_full_matrices(
fr, text_1, text_2,
lcs_delete_cost_function,
lcs_insert_cost_function)
for i in range(0, 5):
for j in range(0, 5):
self.assertEqual(storage
[text_1[i + 1:i + 3]]
[text_2[j + 1:j + 3]]
[
(diff_between_rows[i + 1][j],
diff_between_rows[i + 2][j])
]
[
(diff_between_columns[i][j + 1],
diff_between_columns[i][j + 2])
],
(
[
diff_between_rows[i + 1][j + 2],
diff_between_rows[i + 2][j + 2]
],
[
diff_between_columns[i + 2][j + 1],
diff_between_columns[i + 2][j + 2]
]
))
def test_restore_lcs_part(self):
fr = FourRussiansHelpers(lcs_delete_cost_function,
lcs_insert_cost_function,
lcs_substitute_cost_function)
reversed_lcs, i, j = fr.restore_lcs_part(
"#aa", "#ab",
[0, 1, 1], [0, 1, 1], 2, 2, 2)
self.assertEqual(reversed_lcs, "a")
self.assertEqual(i, 0)
self.assertEqual(j, 0)
reversed_lcs, i, j = fr.restore_lcs_part(
"#aaccc", "#aabc#",
[0, 1, 1, 1, 1, 1], [0, 1, 1, 1, 1, 1], 5, 5, 5)
self.assertEqual(reversed_lcs[::-1], "aac")
self.assertEqual(i, 0)
self.assertEqual(j, 0)
reversed_lcs, i, j = fr.restore_lcs_part(
"#aaa", "#bbb", [0, 1, 1, 1], [0, 1, 1, 1], 3, 3, 3)
self.assertEqual(reversed_lcs, "")
self.assertEqual(i, 0)
self.assertEqual(j, 3)
def check_lcs(self, text_1, text_2, expected_lcs):
expected_length = len(expected_lcs)
fr = FourRussiansHelpers(lcs_delete_cost_function,
lcs_insert_cost_function,
lcs_substitute_cost_function)
m, A, step_size_bound, text_1, text_2 = fr.prepare_parameters(text_1,
text_2)
storage = fr.algorithm_y(m, A, step_size_bound)
P, Q = fr.algorithm_z(m, storage, text_1, text_2)
lcs = fr.restore_lcs(text_1, text_2, P, Q, m)
self.assertEqual(len(lcs), expected_length,
"Expected lcs of length {0}, actual lcs of length {1}. "
"Actual result: {2}, expected result: {3}"
.format(expected_length, len(lcs), lcs, expected_lcs))
self.assertEqual(lcs, expected_lcs)