dan's pr

backend/address2UPRN/main.py

@@ -14,7 +14,7 @@ from utils.s3 import (
 )
 from datetime import datetime
 
-from backend.utils.addressMatch import addressMatch
+from backend.utils.addressMatch import AddressMatch
 
 logger = setup_logger()
 
@@ -35,7 +35,7 @@ def score_addresses(
     if column not in df.columns:
         raise ValueError(f"Missing column: {column}")
 
-    return df[column].apply(lambda x: addressMatch.score(user_address, x))
+    return df[column].apply(lambda x: AddressMatch.score(user_address, x))
 
 
 def get_epc_data_with_postcode(postcode, size=500, attempt=1, max_attempts=3):
@@ -127,10 +127,10 @@ def get_uprn_candidates(
 
     out = df.copy()
 
-    user_norm = addressMatch.normalise_address(user_address)
+    user_norm = AddressMatch.normalise_address(user_address)
 
     out["lexiscore"] = out[address_column].apply(
-        lambda x: addressMatch.levenshtein(user_norm, x)
+        lambda x: AddressMatch.levenshtein(user_norm, x)
     )
 
     # Normalise UPRN to string
@@ -455,7 +455,7 @@ def handler(event, context, local=False):
                 )
 
                 # Validate postcode before processing
-                if not addressMatch.is_valid_postcode(postcode):
+                if not AddressMatch.is_valid_postcode(postcode):
                     logger.warning(f"Postcode {postcode} is invalid, skipping")
                     continue
 

backend/ordnanceSurvey/main.py

@@ -1,4 +1,4 @@
-from typing import Any
+from typing import Any, Optional
 import json
 from utils.logger import setup_logger
 import logging
@@ -8,7 +8,7 @@ from utils.s3 import (
     read_csv_from_s3 as read_csv_from_s3_dict,
     parse_s3_uri,
 )
-from backend.utils.addressMatch import addressMatch
+from backend.utils.addressMatch import AddressMatch
 from backend.app.db.connection import get_db_session
 from backend.app.db.models.postcode_search import PostcodeSearchModel
 from backend.utils.ordnance_survey import (
@@ -124,7 +124,7 @@ def handler(body: dict[str, Any], context: Any, local: bool = False) -> None:
 
     s3_uri: str = body.get("s3_uri", "")
     lexiscore_threshold: float = body.get("lexiscore_threshold", 0.5)
-    lexiscore_column: str = body.get("lexiscore_column", None)
+    lexiscore_column: Optional[str] = body.get("lexiscore_column", None)
     task_id: str = body.get("task_id", "")
     sub_task_id: str = body.get("sub_task_id", "")
 
@@ -158,7 +158,7 @@ def handler(body: dict[str, Any], context: Any, local: bool = False) -> None:
     # Process each postcode group at a time
     for postcode, group in grouped:
         print(f"Processing postcode: {postcode} ({len(group)} rows)")
-        valid_group = addressMatch.is_valid_postcode(postcode)
+        valid_group = AddressMatch.is_valid_postcode(postcode)
         if not valid_group:
             logger.warning(f"Postcode {postcode} is invalid, skipping")
             for idx in group.index:
@@ -203,7 +203,7 @@ def handler(body: dict[str, Any], context: Any, local: bool = False) -> None:
 
             # Score against OS Places addresses
             scores = postcode_cache["ADDRESS"].apply(
-                lambda addr: addressMatch.score(ordnancy_survey_user_input, addr)
+                lambda addr: AddressMatch.score(ordnancy_survey_user_input, addr)
             )
             best_idx = scores.idxmax()
             best_score = scores[best_idx]
@@ -215,12 +215,10 @@ def handler(body: dict[str, Any], context: Any, local: bool = False) -> None:
             df.at[idx, "ordnance_survey_lexiscore"] = best_score
 
     # Save results locally
-    df.to_csv("ordnance_survey_results.csv", index=False)
-    print(f"Results saved to ordnance_survey_results.csv ({len(df)} rows)")
+    if local:
+        df.to_csv("ordnance_survey_results.csv", index=False)
+        print(f"Results saved to ordnance_survey_results.csv ({len(df)} rows)")
 
     # Save results to S3
     if task_id and sub_task_id:
-        try:
-            save_results_to_s3(df, task_id, sub_task_id)
-        except Exception as s3_error:
-            logger.error(f"Failed to save results to S3: {s3_error}")
+        save_results_to_s3(df, task_id, sub_task_id)

backend/utils/addressMatch.py

@@ -4,13 +4,13 @@ from difflib import SequenceMatcher
 import requests
 
 
-class addressMatch:
+class AddressMatch:
     def __init__(self):
         return None
 
     @staticmethod
     def score(a: str, b: str) -> float:
-        score: float = addressMatch.levenshtein(a, b)
+        score: float = AddressMatch.levenshtein(a, b)
 
         return score
 
@@ -143,8 +143,8 @@ class addressMatch:
 
             return None
 
-        a_norm = addressMatch.normalise_address(a)
-        b_norm = addressMatch.normalise_address(b)
+        a_norm = AddressMatch.normalise_address(a)
+        b_norm = AddressMatch.normalise_address(b)
 
         # --- hard signal: numbers ---
         nums_a = extract_numbers(a_norm)