def test_edit_distance_text_empty2(self):
s1 = ""
s2 = "AA\nCC\nDD"
d, aligned, final = edit_distance_text(s1, s2) # pylint: disable=W0632
self.assertEqual(len(aligned), 0)
self.assertEqual(aligned, [])
self.assertEqual(final, [(0, None), (None, 0), (None, 1), (None, 2)])
self.assertEqual(d, 10.)
def test_edit_distance_text(self):
s1 = "AA\nBB\nCC\nDD"
s2 = "AA\nCC\nDD"
d, aligned, final = edit_distance_text(s1, s2) # pylint: disable=W0632
self.assertEqual(len(aligned), 3)
self.assertEqual(aligned, [(0, 0, 0.0, [(0, 0), (1, 1)]),
(2, 1, 0.0, [(0, 0), (1, 1)]),
(3, 2, 0.0, [(0, 0), (1, 1)])])
self.assertEqual(final, [(0, 0), (1, None), (2, 1), (3, 2)])
self.assertEqual(d, 3.)
d, aligned, final = edit_distance_text(s2, s1) # pylint: disable=W0632
self.assertEqual(len(aligned), 3)
self.assertEqual(aligned, [(0, 0, 0.0, [(0, 0), (1, 1)]),
(1, 2, 0.0, [(0, 0), (1, 1)]),
(2, 3, 0.0, [(0, 0), (1, 1)])])
self.assertEqual(final, [(0, 0), (None, 1), (1, 2), (2, 3)])
self.assertEqual(d, 3.)
def test_edit_distance_text_space(self):
s1 = "AA\n\nCC"
s2 = "AA\n\nCC"
d, aligned, final = edit_distance_text(s1, s2) # pylint: disable=W0632
self.assertEqual(d, 0)
self.assertEqual(len(aligned), 3)
self.assertEqual(aligned, [(0, 0, 0.0, [(0, 0), (1, 1)]),
(1, 1, 0.0, []),
(2, 2, 0.0, [(0, 0), (1, 1)])])
self.assertEqual(final, [(0, 0), (1, 1), (2, 2)])
def test_edit_distance_html2(self):
s1 = "AA\nBB\nCC\nZZZZZA\nDD"
s2 = "AA\nCC\nDD\nZZZZZB\nEE"
_, aligned, final = edit_distance_text(s1, s2) # pylint: disable=W0632
ht = diff2html(s1, s2, aligned, final, two_columns=True)
self.assertIn('<tr style="1px solid black;">', ht)
self.assertIn('<td style="background-color:#ABEBC6;">', ht)
self.assertIn('<td style="background-color:#E5E7E9;">', ht)
self.assertIn(
'AA</code></td><td style="background-color:#FFFFFF;">'
'<code style="background-color:#FFFFFF;">AA', ht)
def test_edit_distance_no_align2(self):
s1 = "AA\nBB\nCC\nreturn d\nDD"
s2 = "AA\nBB\nCC\nfor i in jjjjjjjjj\nDD"
d, aligned, final, mats = edit_distance_text(s1,
s2,
return_matrices=True)
dd = edit_distance_string("return d", "for i in jjjjjjjjj")
self.assertEqual(len(final), 6)
self.assertEqual(len(mats), 2)
self.assertEqual(len(aligned), 4)
self.assertGreater(d, 0)
self.assertGreater(dd[0], 0)
def test_edit_distance_text_big(self):
f1 = dedent('''
def edit_distance_string(s1, s2):
"""
Computes the edit distance between strings *s1* and *s2*.
:param s1: first string
:param s2: second string
:return: dist, list of tuples of aligned characters
"""
n1 = len(s1) + 1
n2 = len(s2) + 1
dist = numpy.full((n1, n2), n1 * n2, dtype=numpy.float64)
pred = numpy.full(dist.shape, 0, dtype=numpy.int32)
for j in range(1, n2):
dist[0, j] = j
pred[0, j] = 2
for i in range(0, n1):
dist[i, 0] = i
pred[i, 0] = 1
pred[0, 0] = -1
for j in range(1, n2):
for i in range(1, n1):
c = dist[i, j]
p = 0
if dist[i - 1, j] + 1 < c:
c = dist[i - 1, j] + 1
p = 1
if dist[i, j - 1] + 1 < c:
c = dist[i, j - 1] + 1
p = 2
d = 0 if s1[i - 1] == s2[j - 1] else 1
if dist[i - 1, j - 1] + d < c:
c = dist[i - 1, j - 1] + d
p = 3
if p == 0:
raise RuntimeError(
"Unexpected value for p=%d at position=%r." % (p, (i, j)))
dist[i, j] = c
pred[i, j] = p
d = dist[len(s1), len(s2)]
return d
''')
f2 = dedent('''
def edit_distance_string(s1, s2):
"""
Computes the edit distance between strings *s1* and *s2*.
:param s1: first string
:param s2: second string
:return: dist, list of tuples of aligned characters
"""
n1 = len(s1) + 1
n2 = len(s2) + 1
dist = numpy.full((n1, n2), n1 * n2, dtype=numpy.float64)
pred = numpy.full(dist.shape, 0, dtype=numpy.int32)
for i in range(0, n1):
dist[i, 0] = i
pred[i, 0] = 1
for j in range(1, n2):
dist[0, j] = j
pred[0, j] = 2
pred[0, 0] = -1
for i in range(1, n1):
for j in range(1, n2):
c = dist[i, j]
p = 0
if dist[i - 1, j] + 1 < c:
c = dist[i - 1, j] + 1
p = 1
if dist[i, j - 1] + 1 < c:
c = dist[i, j - 1] + 1
p = 2
d = 0 if s1[i - 1] == s2[j - 1] else 1
if dist[i - 1, j - 1] + d < c:
c = dist[i - 1, j - 1] + d
p = 3
if p == 0:
raise RuntimeError(
"Unexpected value for p=%d at position=%r." % (p, (i, j)))
dist[i, j] = c
pred[i, j] = p
d = dist[len(s1), len(s2)]
equals = []
i, j = len(s1), len(s2)
p = pred[i, j]
while p != -1:
if p == 3:
equals.append((i - 1, j - 1))
i -= 1
j -= 1
elif p == 2:
j -= 1
elif p == 1:
i -= 1
else:
raise RuntimeError(
"Unexpected value for p=%d at position=%r." % (p, (i, j)))
p = pred[i, j]
return d, list(reversed(equals))
''')
res = self.profile(lambda: edit_distance_text(f1, f2, verbose=False),
return_results=True,
rootrem=self.abs_path_join(__file__, '..', '..',
'..'))
d, aligned, final = res[0]
if __name__ == '__main__':
print(res[-1])
self.assertGreater(d, 0)
self.assertGreater(len(aligned), 10)
self.assertIn((1, 1), final)