Coverage for graphqler / compiler / resolvers / utils.py: 96%

1from Levenshtein import distance 

2from graphqler.config import MAX_LEVENSHTEIN_THRESHOLD 

3 

4 

5def find_closest_string_leveshtein(strings: list[str], target: str, threshold: float) -> str: 

6 """Finds the closest string to the target string given a threshold. If none are found, returns "" 

7 

8 Args: 

9 strings (list[str]): The list of strings to search for 

10 target (str): The target string to search for 

11 threshold (float): The treshold value 

12 

13 Returns: 

14 str: Returns a string if it's within the threshold, otherwise returns "" 

15 """ 

16 closest_distance = threshold + 1 

17 closest_string = "" 

18 for string in strings: 

19 dist = distance(string, target) 

20 if dist <= threshold and dist < closest_distance: 

21 closest_distance = dist 

22 closest_string = string 

23 return closest_string 

24 

25 

26def find_closest_string(strings: list[str], target: str) -> str: 

27 """Finds the closest string to the target string 

28 

29 Args: 

30 strings (list[str]): The list of strings (in our case, object names) 

31 target (str): The target (in our case, either the field name or the query/mutation name) 

32 

33 Returns: 

34 str: The found matching string, or "" if nothing close is found 

35 """ 

36 # Do some pre-procssing first (remove underscores, lowercase) 

37 target = target.lower().replace("_", "") 

38 lookup = {} 

39 for string in strings: 

40 lookup[string.lower().replace("_", "")] = string 

41 

42 found_similar_strings = [] 

43 for normalized_string, string in lookup.items(): 

44 if normalized_string in target: 

45 found_similar_strings.append(string) 

46 if len(found_similar_strings) == 0: 

47 return "" 

48 closest_string = find_closest_string_leveshtein(found_similar_strings, target, MAX_LEVENSHTEIN_THRESHOLD) 

49 return closest_string