Merge pull request #35 from statisticsnorway/thefuzz

migrate from fuzzywuzzy to thefuzz

Author: andersfalla

docs/conf.py

@@ -15,7 +15,6 @@
 import os
 import sys
 
-
 sys.path.insert(0, os.path.abspath("../src"))
 
 # -- Project information -----------------------------------------------------

example_notebook/notebook_example.ipynb

@@ -3,7 +3,7 @@
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "34950912-618c-4ed3-ac5d-8adcef20275b",
+   "id": "0",
    "metadata": {
     "tags": []
    },
@@ -18,7 +18,7 @@
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "c17476a1-bad8-4e21-b9fe-9ba709cbeb36",
+   "id": "1",
    "metadata": {
     "tags": []
    },
@@ -31,7 +31,7 @@
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "a07bd2aa-de4a-46cc-a028-bd1a44c49537",
+   "id": "2",
    "metadata": {},
    "outputs": [],
    "source": [
@@ -42,7 +42,7 @@
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "f89ebfb5-3dee-4a71-a876-9fd16de2a431",
+   "id": "3",
    "metadata": {
     "tags": []
    },

noxfile.py

@@ -9,7 +9,6 @@
 
 import nox
 
-
 try:
     from nox_poetry import Session
     from nox_poetry import session

poetry.lock

@@ -16,10 +16,10 @@ Changelog = "https://github.com/statisticsnorway/ssb-hermes/releases"
 [tool.poetry.dependencies]
 python = "^3.10"
 click = ">=8.0.1"
-fuzzywuzzy = "^0.18.0"
 pandas = "^2.1.4"
 numpy = "^1.26.2"
-python-levenshtein = "^0.23.0"
+thefuzz = "^0.22.1"
+ipykernel = "^7.2.0"
 
 [tool.poetry.group.dev.dependencies]
 pygments = ">=2.10.0"

pyproject.toml

@@ -2,15 +2,15 @@
 
 This python file contain the three functions for adress matching based on the three rules:
     * Rule1: If there is only one unit at a location(postnr), then we use this unit.
-    * Rule2: If there are multiple units we use fuzzywuzzy with 75% match.
+    * Rule2: If there are multiple units we use thefuzz with 75% match.
     * Rule3: If rule 1 and 2 did not work, we iterate up geographically.
 """
 
 """Importing packages"""
 from typing import Any
 
 import pandas as pd  # type: ignore
-from fuzzywuzzy import process  # type: ignore
+from thefuzz import process  # type: ignore
 
 """Importing other internal functions"""
 from ._functions import _check_for_value
@@ -46,7 +46,7 @@ def _find_match_rule2(
         tuple: item, rule
     """
     try:
-        item, match = process.extractOne(
+        item, _match = process.extractOne(
             query=query, choices=choices, score_cutoff=score_cutoff
         )
     except (ValueError, AttributeError, TypeError):

src/ssb_hermes/_find_match_rules.py

@@ -3,7 +3,7 @@
 from typing import Any
 
 import pandas as pd  # type: ignore
-from fuzzywuzzy import process  # type: ignore
+from thefuzz import process  # type: ignore
 
 
 def _add_row(

src/ssb_hermes/_functions.py

@@ -102,7 +102,7 @@ def find_match(
                 # Sjekker om det kun finnes en enhet på postnummeret i vof. Isåfall bruker jeg denne enheten
                 if len(liste_registry) == 1:
                     item, rule = _find_match_rule1(liste_registry)
-                # Om det fins flere enheter på samme postnr bruker jeg fuzzywuzzy for å matche dem.
+                # Om det fins flere enheter på samme postnr bruker jeg thefuzz for å matche dem.
                 else:
                     if _check_all_values_equal(
                         df_registry_subset, registry_type_columns