rank address similiarity

backend/address2UPRN/main.py

@@ -15,7 +15,7 @@ from utils.s3 import (
 from datetime import datetime
 
 from backend.utils.addressMatch import AddressMatch
-from backend.address2UPRN.scoring import all_uprns_match, rank_by_address_similarity
+from backend.address2UPRN.scoring import all_uprns_match, rank_address_similarity
 from datatypes.epc.domain.historic_epc_matching import (
     match_addresses_for_postcode,
 )
@@ -79,7 +79,7 @@ def get_uprn_with_epc_df(
     if epc_df.empty:
         return None
 
-    scored_df = rank_by_address_similarity(
+    scored_df = rank_address_similarity(
         epc_df,
         user_address=user_inputed_address,
     )
@@ -171,7 +171,7 @@ def resolve_uprns_for_postcode_group(
     for _, row in group_df.iterrows():
         user_address = str(row[address_col]).strip()
 
-        scored_df = rank_by_address_similarity(
+        scored_df = rank_address_similarity(
             epc_df,
             user_address=user_address,
         )

backend/address2UPRN/scoring.py

@@ -19,8 +19,8 @@ def all_uprns_match(
     return len(uprns) == 1 and uprns[0] == str(target_uprn)
 
 
-def rank_by_address_similarity(
+def rank_address_similarity(
-    df: pd.DataFrame,
+    address_list_df: pd.DataFrame,
     user_address: str,
     address_column: str = "address",
     uprn_column: str = "uprn",
@@ -32,13 +32,13 @@ def rank_by_address_similarity(
     DOES NOT choose or return a UPRN.
     """
 
-    if address_column not in df.columns:
+    if address_column not in address_list_df.columns:
         raise ValueError(f"Missing column: {address_column}")
 
-    if uprn_column not in df.columns:
+    if uprn_column not in address_list_df.columns:
         raise ValueError(f"Missing column: {uprn_column}")
 
-    out = df.copy()
+    out = address_list_df.copy()
 
     user_norm = AddressMatch.normalise_address(user_address)
 

datatypes/epc/domain/historic_epc_matching.py

@@ -4,7 +4,7 @@ from typing import Optional
 import pandas as pd
 from botocore.exceptions import ClientError
 
-from backend.address2UPRN.scoring import rank_by_address_similarity
+from backend.address2UPRN.scoring import rank_address_similarity
 from backend.utils.addressMatch import AddressMatch
 from datatypes.epc.domain.historic_epc import HistoricEpc
 from utils.pandas_utils import pandas_cell_to_str
@@ -85,7 +85,7 @@ def match_addresses_for_postcode(
             ) from e
         raise
 
-    scored = rank_by_address_similarity(
+    scored = rank_address_similarity(
         df,
         user_address=user_address,
         address_column=address_column,

etl/hubspot/hubspotClient.py

@@ -1,6 +1,7 @@
 import os
 import time
 from enum import Enum
+from http import HTTPStatus
 from typing import Optional, cast, Callable, Any
 
 from hubspot.client import Client  # type: ignore[reportMissingTypeStubs]
@@ -86,19 +87,27 @@ class HubspotClient:
 
     def _call_with_retry(self, fn: Callable[[], Any], max_retries: int = 2) -> Any:
         """
-        Call fn(), retrying up to max_retries times on 429 rate-limit errors.
+        Call fn(), retrying up to max_retries times on 429 rate-limit errors
+        or transient 5xx server errors.
         Waits the minimal amount: the remaining interval window reported by HubSpot headers.
         Falls back to the full interval (10s) if headers are absent.
 
         Note: each HubSpot sub-module (deals, companies, etc.) ships its own ApiException
-        class with no shared base beyond Exception, so we detect 429s via duck-typing.
+        class with no shared base beyond Exception, so we detect retryable statuses via duck-typing.
         """
+        retryable_statuses = {
+            HTTPStatus.TOO_MANY_REQUESTS,
+            HTTPStatus.INTERNAL_SERVER_ERROR,
+            HTTPStatus.BAD_GATEWAY,
+            HTTPStatus.SERVICE_UNAVAILABLE,
+            HTTPStatus.GATEWAY_TIMEOUT,
+        }
         for attempt in range(max_retries + 1):
             try:
                 return fn()
             except Exception as e:
                 status = getattr(e, "status", None)
-                if status != 429 or attempt == max_retries:
+                if status not in retryable_statuses or attempt == max_retries:
                     raise
                 headers = getattr(e, "headers", None) or {}
                 interval_ms = int(
@@ -106,7 +115,7 @@ class HubspotClient:
                 )
                 wait_s = interval_ms / 1000.0
                 self.logger.warning(
-                    f"HubSpot 429 (attempt {attempt + 1}/{max_retries}), "
+                    f"HubSpot {status} (attempt {attempt + 1}/{max_retries}), "
                     f"waiting {wait_s:.1f}s before retry."
                 )
                 time.sleep(wait_s)