def WRatio(s1, s2):
p1 = utils.full_process(s1)
p2 = utils.full_process(s2)
if not utils.validate_string(p1): return 0
if not utils.validate_string(p2): return 0
# should we look at partials?
try_partial = True
unbase_scale = .95
partial_scale = .90
base = ratio(p1, p2)
len_ratio = float(max(len(p1), len(p2))) / min(len(p1), len(p2))
# if strings are similar length, don't use partials
if len_ratio < 1.5: try_partial = False
# if one string is much much shorter than the other
if len_ratio > 8: partial_scale = .6
if try_partial:
partial = partial_ratio(p1, p2) * partial_scale
ptsor = partial_token_sort_ratio(p1, p2) * unbase_scale * partial_scale
ptser = partial_token_set_ratio(p1, p2) * unbase_scale * partial_scale
return int(max(base, partial, ptsor, ptser))
else:
tsor = token_sort_ratio(p1, p2) * unbase_scale
tser = token_set_ratio(p1, p2) * unbase_scale
return int(max(base, tsor, tser))
def WRatio(s1, s2):
if not utils.validate_string(s1): return 0
if not utils.validate_string(s2): return 0
p1 = utils.full_process(s1)
p2 = utils.full_process(s2)
# should we look at partials?
try_partial = True
unbase_scale = .95
partial_scale = .90
base = ratio(p1, p2)
len_ratio = float(max(len(p1),len(p2)))/min(len(p1),len(p2))
# if strings are similar length, don't use partials
if len_ratio < 1.5: try_partial = False
# if one string is much much shorter than the other
if len_ratio > 8: partial_scale = .6
if try_partial:
partial = partial_ratio(p1, p2) * partial_scale
ptsor = partial_token_sort_ratio(p1, p2) * unbase_scale * partial_scale
ptser = partial_token_set_ratio(p1, p2) * unbase_scale * partial_scale
return int(max(base, partial, ptsor, ptser))
else:
tsor = token_sort_ratio(p1, p2) * unbase_scale
tser = token_set_ratio(p1, p2) * unbase_scale
return int(max(base, tsor, tser))
def QRatio(s1, s2):
if not utils.validate_string(s1): return 0
if not utils.validate_string(s2): return 0
p1 = utils.full_process(s1)
p2 = utils.full_process(s2)
return ratio(p1, p2)
def QRatio(s1, s2):
if not utils.validate_string(s1): return 0
if not utils.validate_string(s2): return 0
p1 = utils.full_process(s1)
p2 = utils.full_process(s2)
return ratio(p1, p2)
def QRatio(s1, s2, force_ascii=True):
p1 = utils.full_process(s1, force_ascii=force_ascii)
p2 = utils.full_process(s2, force_ascii=force_ascii)
if not utils.validate_string(p1):
return 0
if not utils.validate_string(p2):
return 0
return ratio(p1, p2)
def QRatio(s1, s2, force_ascii=True):
p1 = utils.full_process(s1, force_ascii=force_ascii)
p2 = utils.full_process(s2, force_ascii=force_ascii)
if not utils.validate_string(p1):
return 0
if not utils.validate_string(p2):
return 0
return ratio(p1, p2)
def _token_set(s1, s2, partial=True, force_ascii=True):
"""Find all alphanumeric tokens in each string...
- treat them as a set
- construct two strings of the form:
<sorted_intersection><sorted_remainder>
- take ratios of those two strings
- controls for unordered partial matches"""
if s1 is None:
raise TypeError("s1 is None")
if s2 is None:
raise TypeError("s2 is None")
p1 = utils.full_process(s1, force_ascii=force_ascii)
p2 = utils.full_process(s2, force_ascii=force_ascii)
if not utils.validate_string(p1):
return 0
if not utils.validate_string(p2):
return 0
# pull tokens
tokens1 = set(utils.full_process(p1).split())
tokens2 = set(utils.full_process(p2).split())
intersection = tokens1.intersection(tokens2)
diff1to2 = tokens1.difference(tokens2)
diff2to1 = tokens2.difference(tokens1)
sorted_sect = " ".join(sorted(intersection))
sorted_1to2 = " ".join(sorted(diff1to2))
sorted_2to1 = " ".join(sorted(diff2to1))
combined_1to2 = sorted_sect + " " + sorted_1to2
combined_2to1 = sorted_sect + " " + sorted_2to1
# strip
sorted_sect = sorted_sect.strip()
combined_1to2 = combined_1to2.strip()
combined_2to1 = combined_2to1.strip()
if partial:
ratio_func = partial_ratio
else:
ratio_func = ratio
pairwise = [
ratio_func(sorted_sect, combined_1to2),
ratio_func(sorted_sect, combined_2to1),
ratio_func(combined_1to2, combined_2to1),
]
return max(pairwise)
def _token_set(s1, s2, partial=True, force_ascii=True):
"""Find all alphanumeric tokens in each string...
- treat them as a set
- construct two strings of the form:
<sorted_intersection><sorted_remainder>
- take ratios of those two strings
- controls for unordered partial matches"""
if s1 is None:
raise TypeError("s1 is None")
if s2 is None:
raise TypeError("s2 is None")
p1 = utils.full_process(s1, force_ascii=force_ascii)
p2 = utils.full_process(s2, force_ascii=force_ascii)
if not utils.validate_string(p1):
return 0
if not utils.validate_string(p2):
return 0
# pull tokens
tokens1 = set(utils.full_process(p1).split())
tokens2 = set(utils.full_process(p2).split())
intersection = tokens1.intersection(tokens2)
diff1to2 = tokens1.difference(tokens2)
diff2to1 = tokens2.difference(tokens1)
sorted_sect = " ".join(sorted(intersection))
sorted_1to2 = " ".join(sorted(diff1to2))
sorted_2to1 = " ".join(sorted(diff2to1))
combined_1to2 = sorted_sect + " " + sorted_1to2
combined_2to1 = sorted_sect + " " + sorted_2to1
# strip
sorted_sect = sorted_sect.strip()
combined_1to2 = combined_1to2.strip()
combined_2to1 = combined_2to1.strip()
if partial:
ratio_func = partial_ratio
else:
ratio_func = ratio
pairwise = [
ratio_func(sorted_sect, combined_1to2),
ratio_func(sorted_sect, combined_2to1),
ratio_func(combined_1to2, combined_2to1)
]
return max(pairwise)
def WRatio(s1, s2, force_ascii=True):
"""Return a measure of the sequences' similarity between 0 and 100,
using different algorithms.
"""
p1 = utils.full_process(s1, force_ascii=force_ascii)
p2 = utils.full_process(s2, force_ascii=force_ascii)
if not utils.validate_string(p1):
return 0
if not utils.validate_string(p2):
return 0
# should we look at partials?
try_partial = True
unbase_scale = .95
partial_scale = .90
base = ratio(p1, p2)
len_ratio = float(max(len(p1), len(p2))) / min(len(p1), len(p2))
# if strings are similar length, don't use partials
if len_ratio < 1.5:
try_partial = False
# if one string is much much shorter than the other
if len_ratio > 8:
partial_scale = .6
if try_partial:
partial = partial_ratio(p1, p2) * partial_scale
ptsor = partial_token_sort_ratio(p1, p2, force_ascii=force_ascii) \
* unbase_scale * partial_scale
ptser = partial_token_set_ratio(p1, p2, force_ascii=force_ascii) \
* unbase_scale * partial_scale
return int(max(base, partial, ptsor, ptser))
else:
tsor = token_sort_ratio(p1, p2, force_ascii=force_ascii) * unbase_scale
tser = token_set_ratio(p1, p2, force_ascii=force_ascii) * unbase_scale
return int(max(base, tsor, tser))
def WRatio(s1, s2, force_ascii=True):
"""Return a measure of the sequences' similarity between 0 and 100,
using different algorithms.
"""
p1 = utils.full_process(s1, force_ascii=force_ascii)
p2 = utils.full_process(s2, force_ascii=force_ascii)
if not utils.validate_string(p1):
return 0
if not utils.validate_string(p2):
return 0
# should we look at partials?
try_partial = True
unbase_scale = 0.95
partial_scale = 0.90
base = ratio(p1, p2)
len_ratio = float(max(len(p1), len(p2))) / min(len(p1), len(p2))
# if strings are similar length, don't use partials
if len_ratio < 1.5:
try_partial = False
# if one string is much much shorter than the other
if len_ratio > 8:
partial_scale = 0.6
if try_partial:
partial = partial_ratio(p1, p2) * partial_scale
ptsor = partial_token_sort_ratio(p1, p2, force_ascii=force_ascii) * unbase_scale * partial_scale
ptser = partial_token_set_ratio(p1, p2, force_ascii=force_ascii) * unbase_scale * partial_scale
return int(max(base, partial, ptsor, ptser))
else:
tsor = token_sort_ratio(p1, p2, force_ascii=force_ascii) * unbase_scale
tser = token_set_ratio(p1, p2, force_ascii=force_ascii) * unbase_scale
return int(max(base, tsor, tser))
def _token_print_set(s1, s2, partial=True, force_ascii=True, full_process=True):
"""Find all alphanumeric tokens in each string...
- treat them as a set
- construct two strings of the form:
<sorted_intersection><sorted_remainder>
- take ratios of those two strings
- controls for unordered partial matches"""
p1 = utils.full_process(s1, force_ascii=force_ascii) if full_process else s1
p2 = utils.full_process(s2, force_ascii=force_ascii) if full_process else s2
if not utils.validate_string(p1):
return 0
if not utils.validate_string(p2):
return 0
# pull tokens
tokens1 = set(p1.split())
tokens2 = set(p2.split())
intersection = tokens1.intersection(tokens2)
diff1to2 = tokens1.difference(tokens2)
diff2to1 = tokens2.difference(tokens1)
print(intersection)
print(diff1to2)
print(diff2to1)
sorted_sect = " ".join(sorted(intersection))
sorted_1to2 = " ".join(sorted(diff1to2))
sorted_2to1 = " ".join(sorted(diff2to1))
combined_1to2 = sorted_sect + " " + sorted_1to2
combined_2to1 = sorted_sect + " " + sorted_2to1
# strip
sorted_sect = sorted_sect.strip()
combined_1to2 = combined_1to2.strip()
combined_2to1 = combined_2to1.strip()
def test_validate_strings(self):
tester = None
self.assertTrue(not utils.validate_string(tester))
tester = ""
self.assertTrue(not utils.validate_string(tester))
tester = 0.00123
self.assertTrue(not utils.validate_string(tester))
tester = 999999999999999999999999999999999L
self.assertTrue(not utils.validate_string(tester))
tester = 'a'
self.assertTrue(utils.validate_string(tester))
tester = "This is a perfectly valid string"
self.assertTrue(utils.validate_string(tester))
tester = "This \n\ris \n\r\ra \n\r\t\033[49m\n \033[31mperfectly\033[39m \r\nvalid string"
self.assertTrue(utils.validate_string(tester))
def test_validate_strings(self):
tester = None
self.assertTrue(not utils.validate_string(tester))
tester = ""
self.assertTrue(not utils.validate_string(tester))
tester = 0.00123
self.assertTrue(not utils.validate_string(tester))
tester = 999999999999999999999999999999999L
self.assertTrue(not utils.validate_string(tester))
tester = 'a'
self.assertTrue(utils.validate_string(tester))
tester = "This is a perfectly valid string"
self.assertTrue(utils.validate_string(tester))
tester = "This \n\ris \n\r\ra \n\r\t\033[49m\n \033[31mperfectly\033[39m \r\nvalid string"
self.assertTrue(utils.validate_string(tester))
