def detect_period (filename) :
thres = 0.4
period = None
data = read_file(filename).split('\n')
for text in data:
if distance(text[:6], 'Kỳ hạn') < thres:
if distance(text.strip(' -')[-3:], 'Năm') < thres:
period = digit_detect(replace_char(text.strip().replace(text[:6], '').replace(text[-3:], '')))[0] + ' Năm'
else:
period = digit_detect(replace_char(text.strip().replace(text[:6], '').replace(text[-5:], '')))[0] + ' Tháng'
else:
continue
return period
def match_input_weak(input_wrong, correct): valid = [] if len(valid) == 0: for input_correct in correct: if distance(input_wrong.lower(),input_correct.lower()) == 1: valid.append(input_correct) if len(valid) == 0: for input_correct in correct: if distance(input_wrong.lower(),input_correct.lower()) == 2: valid.append(input_correct) if len(valid) == 0: for input_correct in correct: if distance(input_wrong.lower(),input_correct.lower()) == 3: valid.append(input_correct) if len(valid) == 0: for input_correct in correct: if distance(input_wrong.lower(),input_correct.lower()) == 4: valid.append(input_correct) if len(valid) == 1: return valid[0] return None
def notifyUser(itemOne,itemTwo):
itemOneGiven = itemOne.givName
itemOneFI = itemOne.firstInit
itemOneSurname = itemOne.famName
itemTwoGiven = itemTwo.givName
itemTwoFI = itemTwo.firstInit
itemTwoSurname = itemTwo.famName
givenDist = distance(itemOneGiven,itemTwoGiven)
if itemOneSurname == itemTwoSurname and itemOneFI == itemTwoFI and givenDist <= 3:
return True
else:
return False
def spellCheck(itemOne,itemTwo,totalcount):
itemOneSurname = itemOne.famName
itemTwoSurname = itemTwo.famName
itemOneFirst = itemOne.firstName
itemTwoFirst = itemTwo.firstName
itemOnePub = itemOne.primPub + itemOne.secPub
itemTwoPub = itemTwo.primPub + itemTwo.secPub
itemOneMiddle = itemOne.midName
itemTwoMiddle = itemTwo.midName
if totalcount[itemOneSurname] == 1 and itemOnePub == 1 and itemOneSurname != itemTwoSurname:
if itemTwoPub >= 2:
surnameDist = distance(itemOneSurname,itemTwoSurname)
if surnameDist <= 2:
if itemTwoFirst.startswith(itemOneFirst):
if itemOneMiddle != '' and itemTwoMiddle != '' and itemTwoMiddle.startswith(itemOneMiddle):
return True
elif itemOneMiddle == '' and itemTwoMiddle == '':
return True
elif itemOneSurname == itemTwoSurname and itemOnePub == 1:
givenDist = distance(itemOneFirst,itemTwoFirst)
if givenDist <= 1 and len(itemOneFirst) > 1 and len(itemTwoFirst) > 1:
if itemOneMiddle == itemTwoMiddle:
return True
return False
def compute_similarity(a, b):
"""
Compute the similarity between two given ActiveSite instances.
Distance is calculated using the Damerau-Levenshtein distance. This algorithm quantifies the
'distance' between two strings by counting the number of operations that are needed to transform one string
into another. In this algorithm the operations allowed are inserting a letter, deleting a letter, subsituting a letter
or transposing a letter. This is an improvement on the Levenshtein distance algorithm that preceeded it as
transpositions are common errors, both in the written language, and may have a viable biological relevance.
Implementation from- https://github.com/gfairchild/pyxDamerauLevenshtein
Algorithmic understanding from- https://www.mathworks.com/matlabcentral/cody/problems/2309-calculate-the-damerau-levenshtein-distance-between-two-strings
Input: two ActiveSite instances
Output: the similarity between them (a floating point number)
"""
return distance(a, b)
def detect_currency(filename):
data = read_file(filename).split()
cur = None
for i in data :
if i in currencys:
cur = i
else:
continue
if cur != None:
return cur
else:
thres = 0.5
for i in currencys :
for j in data :
if distance(i, j) < thres:
cur = i
else:
continue
return cur
def terms_match(requested_term: str, source_term: str) -> bool:
if MAX_LEVENSHTEIN_DISTANCE > 0:
return distance(requested_term, source_term) < MAX_LEVENSHTEIN_DISTANCE
return requested_term == source_term
def match_input(input_wrong, correct):
valid = []
supplement = ''
perm = {}
# checks if a valid command is a substring of input
for input_correct in correct:
if input_correct in input_wrong:
valid.append(input_correct)
# if one is found, then check if input begins with input_wrong
# if so, user probably did an accidental concatenation
if len(valid) == 1:
if input_wrong.startswith(valid[0]):
supplement = input_wrong[len(valid[0]):]
# checks if D-L distance is 1
if len(valid) == 0:
for input_correct in correct:
if distance(input_wrong,input_correct) == 1:
valid.append(input_correct)
# checks if D-L distance is 2 or norm. D-L distance <= 3
if len(valid) == 0:
for input_correct in correct:
if distance(input_wrong,input_correct) == 2 or norm_distance(input_wrong,input_correct) <= 0.3:
valid.append(input_correct)
# checks if a permutation of a valid command is a substring of input
if len(valid) == 0:
perm = perm_dict(correct)
for input_correct in correct:
for p in perm[input_correct]:
if p in input_wrong and input_correct not in valid:
valid.append(input_correct)
# checks if a unique valid command starts with the input
if len(valid) == 0:
for input_correct in correct:
if input_correct.startswith(input_wrong):
valid.append(input_correct)
# check if commands starts with substrings of input
# probably should be a last resort measure
substr_len = 1
while len(valid) == 0 and substr_len <= len(input_wrong):
substr_input_wrong = input_wrong[:substr_len]
for input_correct in correct:
if input_correct.startswith(substr_input_wrong):
valid.append(input_correct)
substr_len += 1
# checks if D-L distance is 3
if len(valid) == 0:
for input_correct in correct:
if distance(input_wrong,input_correct) == 3:
valid.append(input_correct)
# checks if there is a permutation with D-L distance of 1
if len(valid) == 0:
for input_correct in correct:
for p in perm[input_correct]:
if distance(p,input_wrong) == 1 and input_correct not in valid:
valid.append(input_correct)
# if there are multiple valid commands found, choose the one that starts
# with same letter as input (if unique)
if len(valid) > 1:
to_remove = []
for input_wrong_temp in valid:
if input_wrong_temp[0] != input_wrong[0]:
to_remove.append(input_wrong_temp)
for input_wrong_temp in to_remove:
valid.remove(input_wrong_temp)
if len(valid) == 1:
return valid[0],supplement
return None
def single_linkage(cluster1, cluster2):
avg_cluster1 = avg_string(cluster1)
avg_cluster2 = avg_string(cluster2)
return distance(avg_cluster1, avg_cluster2)
