move fuzzy searching into separate file
This commit is contained in:
parent
e99ef5203f
commit
72cd844672
|
@ -1,6 +1,52 @@
|
|||
import datetime
|
||||
|
||||
|
||||
class fuzzy_result:
|
||||
def __init__(self, string, score):
|
||||
self.string = string
|
||||
self.score = score
|
||||
|
||||
# consider a score worse if the edit distance is larger
|
||||
def __lt__(self, other):
|
||||
return self.score > other.score
|
||||
|
||||
def __gt__(self, other):
|
||||
return self.score < other.score
|
||||
|
||||
def __le__(self, other):
|
||||
return self.score >= other.score
|
||||
|
||||
def __ge__(self, other):
|
||||
return self.score <= other.score
|
||||
|
||||
def __eq__(self, other):
|
||||
return self.score == other.score
|
||||
|
||||
def __ne__(self, other):
|
||||
return self.score != other.score
|
||||
|
||||
|
||||
# compute levenshtein edit distance, adapted from rosetta code
|
||||
def fuzzy_match(s1, s2):
|
||||
if len(s1) == 0:
|
||||
return len(s2)
|
||||
if len(s2) == 0:
|
||||
return len(s1)
|
||||
edits = [i for i in range(len(s2) + 1)]
|
||||
for i in range(len(s1)):
|
||||
corner = i
|
||||
edits[0] = i + 1
|
||||
for j in range(len(s2)):
|
||||
upper = edits[j + 1]
|
||||
if s1[i] == s2[j]:
|
||||
edits[j + 1] = corner
|
||||
else:
|
||||
m = min(corner, upper, edits[j])
|
||||
edits[j + 1] = m + 1
|
||||
corner = upper
|
||||
return edits[-1]
|
||||
|
||||
|
||||
def get_current_datetime() -> datetime.datetime:
|
||||
# We place this in a separate function so that we can mock it out
|
||||
# in our unit tests.
|
||||
|
|
|
@ -5,6 +5,7 @@ from zope import component
|
|||
from .utils import authz_restrict_to_syscom, is_truthy, \
|
||||
create_streaming_response, development_only
|
||||
from ceo_common.interfaces import ILDAPService
|
||||
from ceo_common.utils import fuzzy_result, fuzzy_match
|
||||
from ceod.transactions.groups import (
|
||||
AddGroupTransaction,
|
||||
AddMemberToGroupTransaction,
|
||||
|
@ -37,58 +38,14 @@ def get_group(group_name):
|
|||
|
||||
@bp.route('/search/<query>/<count>')
|
||||
def search_group(query, count):
|
||||
# compute levenshtein edit distance, adapted from rosetta code
|
||||
def _fuzzy_match(s1, s2):
|
||||
if len(s1) == 0:
|
||||
return len(s2)
|
||||
if len(s2) == 0:
|
||||
return len(s1)
|
||||
edits = [i for i in range(len(s2) + 1)]
|
||||
for i in range(len(s1)):
|
||||
corner = i
|
||||
edits[0] = i + 1
|
||||
for j in range(len(s2)):
|
||||
upper = edits[j + 1]
|
||||
if s1[i] == s2[j]:
|
||||
edits[j + 1] = corner
|
||||
else:
|
||||
m = min(corner, upper, edits[j])
|
||||
edits[j + 1] = m + 1
|
||||
corner = upper
|
||||
return edits[-1]
|
||||
|
||||
class _fuzzy_result:
|
||||
def __init__(self, string, score):
|
||||
self.string = string
|
||||
self.score = score
|
||||
|
||||
# consider a score worse if the edit distance is larger
|
||||
def __lt__(self, other):
|
||||
return self.score > other.score
|
||||
|
||||
def __gt__(self, other):
|
||||
return self.score < other.score
|
||||
|
||||
def __le__(self, other):
|
||||
return self.score >= other.score
|
||||
|
||||
def __ge__(self, other):
|
||||
return self.score <= other.score
|
||||
|
||||
def __eq__(self, other):
|
||||
return self.score == other.score
|
||||
|
||||
def __ne__(self, other):
|
||||
return self.score != other.score
|
||||
|
||||
query = str(query)
|
||||
count = int(count)
|
||||
ldap_srv = component.getUtility(ILDAPService)
|
||||
clubs = ldap_srv.get_clubs()
|
||||
scores = [_fuzzy_result("", 99999) for _ in range(count)]
|
||||
scores = [fuzzy_result("", 99999) for _ in range(count)]
|
||||
for club in clubs:
|
||||
score = _fuzzy_match(query, str(club.cn))
|
||||
result = _fuzzy_result(str(club.cn), score)
|
||||
score = fuzzy_match(query, str(club.cn))
|
||||
result = fuzzy_result(str(club.cn), score)
|
||||
heappushpop(scores, result)
|
||||
|
||||
result = [score.string for score in nlargest(count, scores)]
|
||||
|
|
Loading…
Reference in New Issue