unsanistiesed address list instead of raw address lit

applications/SAL/handler.py

@@ -1,13 +1,13 @@
 from typing import Any
 import boto3
-from orchestration.landlord_description_overrides_orchestrator import (
-    LandlordDescriptionOverridesOrchestrator,
+from orchestration.sal_orchestrator import (
+    SALOrchestrator,
 )
 from infrastructure.csv_s3_client import CsvS3Client
-from repositories.user_address.user_address_csv_s3_repository import (
-    UserAddressCsvS3Repository,
+from repositories.unsanitised_address.unsanitised_address_list_csv_s3_repository import (
+    UnsanitisedAddressListCsvS3Repository,
 )
-from domain.addresses.user_address import LandlordAssetList
+from domain.addresses.unsanitised_address import AddressList
 
 
 def handler(
@@ -20,24 +20,20 @@ def handler(
 
     # boto3.client is overloaded per-service in the installed stubs; cast
     # to Any so the strict-mode checker treats it as opaque.
-    boto3_client: Any = (
-        boto3.client
-    )  # pyright: ignore[reportUnknownMemberType, reportUnknownVariableType]
+    boto3_client: Any = boto3.client  # noqa
     boto_s3: Any = boto3_client("s3")
 
     csv_client = CsvS3Client(boto_s3, bucket)
-    user_address_repo = UserAddressCsvS3Repository(csv_client, bucket)
+    unsanitised_address_repo = UnsanitisedAddressListCsvS3Repository(csv_client, bucket)
 
-    orchestrator = LandlordDescriptionOverridesOrchestrator(
-        user_address_repo=user_address_repo,
+    sal = SALOrchestrator(
+        unsanitised_address_repo=unsanitised_address_repo,
     )
 
-    list_of_user_address: list[LandlordAssetList] = orchestrator.get_user_address(
-        input_s3_uri=s3_uri
-    )
+    addressList: AddressList = sal.get_unsanitised_addresses(input_s3_uri=s3_uri)
 
-    col_to_desc_map = orchestrator.get_col_to_description_mappings(
-        list_of_user_address=list_of_user_address
+    col_to_desc_map = sal.get_col_to_description_mappings(
+        list_of_unsanitised_address=addressList
     )
 
     # Read csv of user input

applications/postcode_splitter/handler.py

@@ -12,8 +12,8 @@ from infrastructure.address2uprn_queue_client import Address2UprnQueueClient
 from infrastructure.csv_s3_client import CsvS3Client
 from orchestration.postcode_splitter_orchestrator import PostcodeSplitterOrchestrator
 from orchestration.task_orchestrator import TaskOrchestrator
-from repositories.user_address.user_address_csv_s3_repository import (
-    UserAddressCsvS3Repository,
+from repositories.unsanitised_address.unsanitised_address_list_csv_s3_repository import (
+    UnsanitisedAddressListCsvS3Repository,
 )
 from utilities.aws_lambda.subtask_handler import subtask_handler
 
@@ -29,17 +29,19 @@ def handler(
 
     # boto3.client is overloaded per-service in the installed stubs; cast
     # to Any so the strict-mode checker treats it as opaque.
-    boto3_client: Any = boto3.client  # pyright: ignore[reportUnknownMemberType, reportUnknownVariableType]
+    boto3_client: Any = (
+        boto3.client
+    )  # pyright: ignore[reportUnknownMemberType, reportUnknownVariableType]
     boto_s3: Any = boto3_client("s3")
     boto_sqs: Any = boto3_client("sqs")
 
     csv_client = CsvS3Client(boto_s3, bucket)
-    user_address_repo = UserAddressCsvS3Repository(csv_client, bucket)
+    unsanitised_address_repo = UnsanitisedAddressListCsvS3Repository(csv_client, bucket)
     queue_client = Address2UprnQueueClient(boto_sqs, queue_url)
 
     splitter = PostcodeSplitterOrchestrator(
         task_orchestrator=task_orchestrator,
-        user_address_repo=user_address_repo,
+        unsanitised_address_repo=unsanitised_address_repo,
         queue_client=queue_client,
     )
 

domain/addresses/postcode_batching.py

@@ -2,21 +2,21 @@ from __future__ import annotations
 
 from collections.abc import Iterable, Iterator
 
-from domain.addresses.user_address import LandlordAssetList
+from domain.addresses.unsanitised_address import AddressList, UnsanitisedAddress
 from domain.postcode import Postcode
 
 
 def iter_postcode_grouped_batches(
-    addresses: Iterable[LandlordAssetList],
+    addresses: Iterable[UnsanitisedAddress],
     *,
     max_batch_size: int = 500,
-) -> Iterator[list[LandlordAssetList]]:
+) -> Iterator[AddressList]:
     if max_batch_size < 1:
         raise ValueError("max_batch_size must be >= 1")
 
     groups = _group_by_postcode_in_order(addresses)
 
-    buffer: list[LandlordAssetList] = []
+    buffer: AddressList = AddressList([])
     for group in groups.values():
         group_len = len(group)
 
@@ -26,14 +26,14 @@ def iter_postcode_grouped_batches(
         if group_len >= max_batch_size:
             if buffer:
                 yield buffer
-                buffer = []
+                buffer = AddressList([])
             yield group
             continue
 
         # Adding this group would overflow: flush buffer before appending.
         if len(buffer) + group_len > max_batch_size:
             yield buffer
-            buffer = []
+            buffer = AddressList([])
 
         buffer.extend(group)
 
@@ -43,9 +43,9 @@ def iter_postcode_grouped_batches(
 
 
 def _group_by_postcode_in_order(
-    addresses: Iterable[LandlordAssetList],
-) -> dict[Postcode, list[LandlordAssetList]]:
-    groups: dict[Postcode, list[LandlordAssetList]] = {}
+    addresses: Iterable[UnsanitisedAddress],
+) -> dict[Postcode, AddressList]:
+    groups: dict[Postcode, AddressList] = {}
     for address in addresses:
-        groups.setdefault(address.postcode, []).append(address)
+        groups.setdefault(address.postcode, AddressList([])).append(address)
     return groups

domain/addresses/unsanitised_address.py

@@ -0,0 +1,24 @@
+from __future__ import annotations
+
+from dataclasses import dataclass, field
+from typing import Optional, NewType
+
+from domain.postcode import Postcode
+
+
+def _empty_source_row() -> dict[str, str]:
+    return {}
+
+
+@dataclass(frozen=True)
+class UnsanitisedAddress:
+    address: str
+    postcode: Postcode
+    org_reference: Optional[str] = None
+    additional_info: dict[str, str] = field(
+        default_factory=_empty_source_row, compare=False
+    )
+
+
+# A batch of raw, pre-standardisation addresses as supplied by a landlord.
+AddressList = NewType("AddressList", list[UnsanitisedAddress])

domain/addresses/user_address.py

@@ -1,20 +0,0 @@
-from __future__ import annotations
-
-from dataclasses import dataclass, field
-from typing import Optional
-
-from domain.postcode import Postcode
-
-
-def _empty_source_row() -> dict[str, str]:
-    return {}
-
-
-@dataclass(frozen=True)
-class LandlordAssetList:
-    user_address: str
-    postcode: Postcode
-    internal_reference: Optional[str] = None
-    landlord_additional_info: dict[str, str] = field(
-        default_factory=_empty_source_row, compare=False
-    )

orchestration/landlord_description_overrides_orchestrator.py

@@ -1,23 +0,0 @@
-from repositories.user_address.user_address_repository import UserAddressRepository
-from domain.addresses.user_address import LandlordAssetList
-
-
-class LandlordDescriptionOverridesOrchestrator:
-    def __init__(self, user_address_repo: UserAddressRepository) -> None:
-        self._user_address_repo = user_address_repo
-
-    def get_user_address(
-        self,
-        input_s3_uri: str,
-    ) -> list[LandlordAssetList]:
-        return self._user_address_repo.load_batch(input_s3_uri)
-
-    def get_col_to_description_mappings(
-        self, list_of_user_address: list[LandlordAssetList]
-    ) -> dict[str, set[str]]:
-        mappings: dict[str, set[str]] = {}
-        for user_address in list_of_user_address:
-            for key, value in user_address.landlord_additional_info.items():
-                # Lower-case so case-only typos collapse to one variant.
-                mappings.setdefault(key, set()).add(value.lower())
-        return mappings

orchestration/postcode_splitter_orchestrator.py

@@ -5,19 +5,21 @@ from uuid import UUID
 from infrastructure.address2uprn_queue_client import Address2UprnQueueClient
 from orchestration.task_orchestrator import TaskOrchestrator
 from domain.addresses.postcode_batching import iter_postcode_grouped_batches
-from repositories.user_address.user_address_repository import UserAddressRepository
+from repositories.unsanitised_address.unsanitised_address_list_repository import (
+    UnsanitisedAddressListRepository,
+)
 
 
 class PostcodeSplitterOrchestrator:
     def __init__(
         self,
         task_orchestrator: TaskOrchestrator,
-        user_address_repo: UserAddressRepository,
+        unsanitised_address_repo: UnsanitisedAddressListRepository,
         queue_client: Address2UprnQueueClient,
         max_batch_size: int = 500,
     ) -> None:
         self._task_orchestrator = task_orchestrator
-        self._user_address_repo = user_address_repo
+        self._unsanitised_address_repo = unsanitised_address_repo
         self._queue_client = queue_client
         self._max_batch_size = max_batch_size
 
@@ -28,7 +30,7 @@ class PostcodeSplitterOrchestrator:
         parent_subtask_id: UUID,
         input_s3_uri: str,
     ) -> list[UUID]:
-        addresses = self._user_address_repo.load_batch(input_s3_uri)
+        addresses = self._unsanitised_address_repo.load_batch(input_s3_uri)
         path_prefix = (
             f"ara_postcode_splitter_batches/{parent_task_id}/{parent_subtask_id}"
         )
@@ -37,7 +39,7 @@ class PostcodeSplitterOrchestrator:
         for batch in iter_postcode_grouped_batches(
             addresses, max_batch_size=self._max_batch_size
         ):
-            batch_uri = self._user_address_repo.save_batch(batch, path_prefix)
+            batch_uri = self._unsanitised_address_repo.save_batch(batch, path_prefix)
             child = self._task_orchestrator.create_child_subtask(
                 parent_task_id,
                 inputs={

orchestration/sal_orchestrator.py

@@ -0,0 +1,25 @@
+from repositories.unsanitised_address.unsanitised_address_list_repository import (
+    UnsanitisedAddressListRepository,
+)
+from domain.addresses.unsanitised_address import AddressList
+
+
+class SALOrchestrator:
+    def __init__(self, unsanitised_address_repo: UnsanitisedAddressListRepository) -> None:
+        self._unsanitised_address_repo = unsanitised_address_repo
+
+    def get_unsanitised_addresses(
+        self,
+        input_s3_uri: str,
+    ) -> AddressList:
+        return self._unsanitised_address_repo.load_batch(input_s3_uri)
+
+    def get_col_to_description_mappings(
+        self, list_of_unsanitised_address: AddressList
+    ) -> dict[str, set[str]]:
+        mappings: dict[str, set[str]] = {}
+        for unsanitised_address in list_of_unsanitised_address:
+            for key, value in unsanitised_address.additional_info.items():
+                # Lower-case so case-only typos collapse to one variant.
+                mappings.setdefault(key, set()).add(value.lower())
+        return mappings

repositories/unsanitised_address/unsanitised_address_list_csv_s3_repository.py

@@ -4,10 +4,12 @@ import uuid
 from datetime import datetime, timezone
 from typing import Optional
 
-from domain.addresses.user_address import LandlordAssetList
+from domain.addresses.unsanitised_address import AddressList, UnsanitisedAddress
 from domain.postcode import Postcode
 from infrastructure.csv_s3_client import CsvS3Client
-from repositories.user_address.user_address_repository import UserAddressRepository
+from repositories.unsanitised_address.unsanitised_address_list_repository import (
+    UnsanitisedAddressListRepository,
+)
 
 _ADDRESS_COLUMNS: tuple[str, str, str] = ("Address 1", "Address 2", "Address 3")
 _POSTCODE_COLUMN: str = "postcode"
@@ -15,43 +17,43 @@ _INTERNAL_REFERENCE_COLUMN: str = "Internal Reference"
 _POSTCODE_CLEAN_COLUMN: str = "postcode_clean"
 
 
-class UserAddressCsvS3Repository(UserAddressRepository):
+class UnsanitisedAddressListCsvS3Repository(UnsanitisedAddressListRepository):
     def __init__(self, csv_client: CsvS3Client, bucket: str) -> None:
         self._csv_client = csv_client
         self._bucket = bucket
 
-    def load_batch(self, s3_uri: str) -> list[LandlordAssetList]:
+    def load_batch(self, s3_uri: str) -> AddressList:
         rows = self._csv_client.read_rows(s3_uri)
         if rows and _POSTCODE_COLUMN not in rows[0]:
             raise ValueError(
                 f"Input CSV {s3_uri} has no {_POSTCODE_COLUMN!r} column; "
                 f"columns present: {sorted(rows[0])}"
             )
-        addresses: list[LandlordAssetList] = []
+        addresses: AddressList = AddressList([])
         for row in rows:
             parts = [
                 row[col].strip()
                 for col in _ADDRESS_COLUMNS
                 if col in row and row[col].strip()
             ]
-            user_address = ", ".join(parts)
+            unsanitised_address = ", ".join(parts)
             postcode = row.get(_POSTCODE_COLUMN, "")
             raw_ref = row.get(_INTERNAL_REFERENCE_COLUMN, "").strip()
             internal_reference: Optional[str] = raw_ref or None
             addresses.append(
-                LandlordAssetList(
-                    user_address=user_address,
+                UnsanitisedAddress(
+                    address=unsanitised_address,
                     postcode=Postcode(postcode),
-                    internal_reference=internal_reference,
-                    landlord_additional_info=row,
+                    org_reference=internal_reference,
+                    additional_info=row,
                 )
             )
         return addresses
 
-    def save_batch(self, addresses: list[LandlordAssetList], path_prefix: str) -> str:
+    def save_batch(self, addresses: AddressList, path_prefix: str) -> str:
         rows: list[dict[str, str]] = [
             {
-                **addr.landlord_additional_info,
+                **addr.additional_info,
                 _POSTCODE_CLEAN_COLUMN: str(addr.postcode),
             }
             for addr in addresses

repositories/unsanitised_address/unsanitised_address_list_repository.py

@@ -0,0 +1,13 @@
+from __future__ import annotations
+
+from abc import ABC, abstractmethod
+
+from domain.addresses.unsanitised_address import AddressList
+
+
+class UnsanitisedAddressListRepository(ABC):
+    @abstractmethod
+    def load_batch(self, s3_uri: str) -> AddressList: ...
+
+    @abstractmethod
+    def save_batch(self, addresses: AddressList, path_prefix: str) -> str: ...

repositories/user_address/user_address_repository.py

@@ -1,15 +0,0 @@
-from __future__ import annotations
-
-from abc import ABC, abstractmethod
-
-from domain.addresses.user_address import LandlordAssetList
-
-
-class UserAddressRepository(ABC):
-    @abstractmethod
-    def load_batch(self, s3_uri: str) -> list[LandlordAssetList]: ...
-
-    @abstractmethod
-    def save_batch(
-        self, addresses: list[LandlordAssetList], path_prefix: str
-    ) -> str: ...

tests/domain/addresses/test_postcode_batching.py

@@ -1,17 +1,17 @@
 import pytest
 
 from domain.addresses.postcode_batching import iter_postcode_grouped_batches
-from domain.addresses.user_address import LandlordAssetList
+from domain.addresses.unsanitised_address import AddressList, UnsanitisedAddress
 from domain.postcode import Postcode
 
 
-def _addrs(postcode: str, n: int) -> list[LandlordAssetList]:
-    return [
-        LandlordAssetList(
-            user_address=f"{i} {postcode} Street", postcode=Postcode(postcode)
-        )
-        for i in range(n)
-    ]
+def _addrs(postcode: str, n: int) -> AddressList:
+    return AddressList(
+        [
+            UnsanitisedAddress(address=f"{i} {postcode} Street", postcode=Postcode(postcode))
+            for i in range(n)
+        ]
+    )
 
 
 def test_empty_input_yields_no_batches() -> None:

tests/domain/addresses/test_unsanitised_address.py

@@ -0,0 +1,96 @@
+import dataclasses
+
+import pytest
+
+from domain.addresses.unsanitised_address import UnsanitisedAddress
+from domain.postcode import Postcode
+
+
+def test_unsanitised_address_holds_postcode_value_object() -> None:
+    # act
+    addr = UnsanitisedAddress(address="1 The Street", postcode=Postcode("sw1a 1aa"))
+    # assert
+    assert addr.postcode == Postcode("SW1A1AA")
+
+
+def test_unsanitised_address_preserves_unsanitised_address_verbatim() -> None:
+    # The free-text unsanitised_address string is intentionally NOT normalised --
+    # only the postcode is canonicalised, and that happens inside Postcode.
+    # act
+    addr = UnsanitisedAddress(address="  1 The   Street  ", postcode=Postcode("SW1A1AA"))
+    # assert
+    assert addr.address == "  1 The   Street  "
+
+
+def test_unsanitised_address_internal_reference_defaults_to_none() -> None:
+    # act
+    addr = UnsanitisedAddress(address="1 The Street", postcode=Postcode("SW1A1AA"))
+    # assert
+    assert addr.org_reference is None
+
+
+def test_unsanitised_address_internal_reference_accepted() -> None:
+    # act
+    addr = UnsanitisedAddress(
+        address="1 The Street",
+        postcode=Postcode("SW1A1AA"),
+        org_reference="cust-42",
+    )
+    # assert
+    assert addr.org_reference == "cust-42"
+
+
+def test_unsanitised_address_is_frozen() -> None:
+    # arrange
+    addr = UnsanitisedAddress(address="1 The Street", postcode=Postcode("SW1A1AA"))
+    # act / assert
+    with pytest.raises(dataclasses.FrozenInstanceError):
+        addr.postcode = Postcode("OTHER")  # type: ignore[misc]
+
+
+def test_unsanitised_address_equality_uses_canonical_postcode() -> None:
+    # Postcode sanitises eagerly, so addresses built from different surface
+    # forms of the same postcode compare equal.
+    # arrange
+    a = UnsanitisedAddress(address="1 The Street", postcode=Postcode("sw1a 1aa"))
+    b = UnsanitisedAddress(address="1 The Street", postcode=Postcode("SW1A1AA"))
+    # act / assert
+    assert a == b
+
+
+def test_unsanitised_address_source_row_defaults_to_empty_dict() -> None:
+    # act
+    addr = UnsanitisedAddress(address="1 The Street", postcode=Postcode("SW1A1AA"))
+    # assert
+    assert addr.additional_info == {}
+
+
+def test_unsanitised_address_carries_source_row() -> None:
+    # arrange
+    row = {"Address 1": "1 The Street", "postcode": "SW1A 1AA", "SAP Score": "72"}
+    # act
+    addr = UnsanitisedAddress(
+        address="1 The Street",
+        postcode=Postcode("SW1A 1AA"),
+        additional_info=row,
+    )
+    # assert
+    assert addr.additional_info == row
+
+
+def test_unsanitised_address_equality_ignores_source_row() -> None:
+    # source_row is excluded from equality (and hashing): identity stays
+    # defined by the parsed fields.
+    # arrange
+    a = UnsanitisedAddress(
+        address="1 The Street",
+        postcode=Postcode("SW1A1AA"),
+        additional_info={"x": "1"},
+    )
+    b = UnsanitisedAddress(
+        address="1 The Street",
+        postcode=Postcode("SW1A1AA"),
+        additional_info={"y": "2"},
+    )
+    # act / assert
+    assert a == b

tests/domain/addresses/test_user_address.py

@@ -1,98 +0,0 @@
-import dataclasses
-
-import pytest
-
-from domain.addresses.user_address import LandlordAssetList
-from domain.postcode import Postcode
-
-
-def test_user_address_holds_postcode_value_object() -> None:
-    # act
-    addr = LandlordAssetList(user_address="1 The Street", postcode=Postcode("sw1a 1aa"))
-    # assert
-    assert addr.postcode == Postcode("SW1A1AA")
-
-
-def test_user_address_preserves_user_address_verbatim() -> None:
-    # The free-text user_address string is intentionally NOT normalised --
-    # only the postcode is canonicalised, and that happens inside Postcode.
-    # act
-    addr = LandlordAssetList(
-        user_address="  1 The   Street  ", postcode=Postcode("SW1A1AA")
-    )
-    # assert
-    assert addr.user_address == "  1 The   Street  "
-
-
-def test_user_address_internal_reference_defaults_to_none() -> None:
-    # act
-    addr = LandlordAssetList(user_address="1 The Street", postcode=Postcode("SW1A1AA"))
-    # assert
-    assert addr.internal_reference is None
-
-
-def test_user_address_internal_reference_accepted() -> None:
-    # act
-    addr = LandlordAssetList(
-        user_address="1 The Street",
-        postcode=Postcode("SW1A1AA"),
-        internal_reference="cust-42",
-    )
-    # assert
-    assert addr.internal_reference == "cust-42"
-
-
-def test_user_address_is_frozen() -> None:
-    # arrange
-    addr = LandlordAssetList(user_address="1 The Street", postcode=Postcode("SW1A1AA"))
-    # act / assert
-    with pytest.raises(dataclasses.FrozenInstanceError):
-        addr.postcode = Postcode("OTHER")  # type: ignore[misc]
-
-
-def test_user_address_equality_uses_canonical_postcode() -> None:
-    # Postcode sanitises eagerly, so addresses built from different surface
-    # forms of the same postcode compare equal.
-    # arrange
-    a = LandlordAssetList(user_address="1 The Street", postcode=Postcode("sw1a 1aa"))
-    b = LandlordAssetList(user_address="1 The Street", postcode=Postcode("SW1A1AA"))
-    # act / assert
-    assert a == b
-
-
-def test_user_address_source_row_defaults_to_empty_dict() -> None:
-    # act
-    addr = LandlordAssetList(user_address="1 The Street", postcode=Postcode("SW1A1AA"))
-    # assert
-    assert addr.landlord_additional_info == {}
-
-
-def test_user_address_carries_source_row() -> None:
-    # arrange
-    row = {"Address 1": "1 The Street", "postcode": "SW1A 1AA", "SAP Score": "72"}
-    # act
-    addr = LandlordAssetList(
-        user_address="1 The Street",
-        postcode=Postcode("SW1A 1AA"),
-        landlord_additional_info=row,
-    )
-    # assert
-    assert addr.landlord_additional_info == row
-
-
-def test_user_address_equality_ignores_source_row() -> None:
-    # source_row is excluded from equality (and hashing): identity stays
-    # defined by the parsed fields.
-    # arrange
-    a = LandlordAssetList(
-        user_address="1 The Street",
-        postcode=Postcode("SW1A1AA"),
-        landlord_additional_info={"x": "1"},
-    )
-    b = LandlordAssetList(
-        user_address="1 The Street",
-        postcode=Postcode("SW1A1AA"),
-        landlord_additional_info={"y": "2"},
-    )
-    # act / assert
-    assert a == b

tests/orchestration/test_landlord_description_overrides_orchestrator.py

@@ -1,44 +1,46 @@
 from __future__ import annotations
 
-from domain.addresses.user_address import LandlordAssetList
+from domain.addresses.unsanitised_address import AddressList, UnsanitisedAddress
 from domain.postcode import Postcode
-from orchestration.landlord_description_overrides_orchestrator import (
-    LandlordDescriptionOverridesOrchestrator,
+from orchestration.sal_orchestrator import (
+    SALOrchestrator,
+)
+from repositories.unsanitised_address.unsanitised_address_list_repository import (
+    UnsanitisedAddressListRepository,
 )
-from repositories.user_address.user_address_repository import UserAddressRepository
 
 
-class _StubUserAddressRepository(UserAddressRepository):
+class _StubUnsanitisedAddressRepository(UnsanitisedAddressListRepository):
     """``get_col_to_description_mappings`` never touches the repo."""
 
-    def load_batch(self, s3_uri: str) -> list[LandlordAssetList]:
+    def load_batch(self, s3_uri: str) -> AddressList:
         raise NotImplementedError()
 
-    def save_batch(self, addresses: list[LandlordAssetList], path_prefix: str) -> str:
+    def save_batch(self, addresses: AddressList, path_prefix: str) -> str:
         raise NotImplementedError()
 
 
-def _make_user_address(landlord_additional_info: dict[str, str]) -> LandlordAssetList:
-    return LandlordAssetList(
-        user_address="1 High St",
+def _make_unsanitised_address(landlord_additional_info: dict[str, str]) -> UnsanitisedAddress:
+    return UnsanitisedAddress(
+        address="1 High St",
         postcode=Postcode("AA1 1AA"),
-        landlord_additional_info=landlord_additional_info,
+        additional_info=landlord_additional_info,
     )
 
 
-def _orchestrator() -> LandlordDescriptionOverridesOrchestrator:
-    return LandlordDescriptionOverridesOrchestrator(
-        user_address_repo=_StubUserAddressRepository()
-    )
+def _orchestrator() -> SALOrchestrator:
+    return SALOrchestrator(unsanitised_address_repo=_StubUnsanitisedAddressRepository())
 
 
 def test_collects_every_value_per_shared_key() -> None:
     # arrange: every address carries the same keys, all values distinct.
-    addresses = [
-        _make_user_address({"description": "cosy", "condition": "new"}),
-        _make_user_address({"description": "spacious", "condition": "worn"}),
-        _make_user_address({"description": "bright", "condition": "fair"}),
-    ]
+    addresses = AddressList(
+        [
+            _make_unsanitised_address({"description": "cosy", "condition": "new"}),
+            _make_unsanitised_address({"description": "spacious", "condition": "worn"}),
+            _make_unsanitised_address({"description": "bright", "condition": "fair"}),
+        ]
+    )
 
     # act
     mappings = _orchestrator().get_col_to_description_mappings(addresses)
@@ -52,11 +54,13 @@ def test_collects_every_value_per_shared_key() -> None:
 
 def test_repeated_values_collapse_to_one_variant() -> None:
     # arrange: two addresses share the same wall description.
-    addresses = [
-        _make_user_address({"description": "cosy"}),
-        _make_user_address({"description": "cosy"}),
-        _make_user_address({"description": "bright"}),
-    ]
+    addresses = AddressList(
+        [
+            _make_unsanitised_address({"description": "cosy"}),
+            _make_unsanitised_address({"description": "cosy"}),
+            _make_unsanitised_address({"description": "bright"}),
+        ]
+    )
 
     # act
     mappings = _orchestrator().get_col_to_description_mappings(addresses)
@@ -67,11 +71,13 @@ def test_repeated_values_collapse_to_one_variant() -> None:
 
 def test_case_only_variants_collapse_to_one() -> None:
     # arrange: the same description typed with inconsistent casing.
-    addresses = [
-        _make_user_address({"description": "Cosy"}),
-        _make_user_address({"description": "cosy"}),
-        _make_user_address({"description": "COSY"}),
-    ]
+    addresses = AddressList(
+        [
+            _make_unsanitised_address({"description": "Cosy"}),
+            _make_unsanitised_address({"description": "cosy"}),
+            _make_unsanitised_address({"description": "COSY"}),
+        ]
+    )
 
     # act
     mappings = _orchestrator().get_col_to_description_mappings(addresses)
@@ -82,7 +88,7 @@ def test_case_only_variants_collapse_to_one() -> None:
 
 def test_empty_address_list_yields_empty_mapping() -> None:
     # arrange / act
-    mappings = _orchestrator().get_col_to_description_mappings([])
+    mappings = _orchestrator().get_col_to_description_mappings(AddressList([]))
 
     # assert
     assert mappings == {}
@@ -90,7 +96,7 @@ def test_empty_address_list_yields_empty_mapping() -> None:
 
 def test_single_address_yields_single_value_per_key() -> None:
     # arrange
-    addresses = [_make_user_address({"description": "cosy"})]
+    addresses = AddressList([_make_unsanitised_address({"description": "cosy"})])
 
     # act
     mappings = _orchestrator().get_col_to_description_mappings(addresses)

tests/orchestration/test_postcode_splitter_orchestrator.py

@@ -18,8 +18,8 @@ from orchestration.postcode_splitter_orchestrator import PostcodeSplitterOrchest
 from orchestration.task_orchestrator import TaskOrchestrator
 from repositories.tasks.subtask_postgres_repository import SubTaskPostgresRepository
 from repositories.tasks.task_postgres_repository import TaskPostgresRepository
-from repositories.user_address.user_address_csv_s3_repository import (
-    UserAddressCsvS3Repository,
+from repositories.unsanitised_address.unsanitised_address_list_csv_s3_repository import (
+    UnsanitisedAddressListCsvS3Repository,
 )
 
 BUCKET = "splitter-bucket"
@@ -27,7 +27,9 @@ REGION = "us-east-1"
 
 
 def _make_boto_client(service_name: str) -> Any:
-    factory: Any = boto3.client  # pyright: ignore[reportUnknownMemberType, reportUnknownVariableType]
+    factory: Any = (
+        boto3.client
+    )  # pyright: ignore[reportUnknownMemberType, reportUnknownVariableType]
     return factory(service_name, region_name=REGION)
 
 
@@ -62,7 +64,7 @@ class Harness:
     csv_client: CsvS3Client
     boto_sqs: Any
     queue_url: str
-    repo: UserAddressCsvS3Repository
+    repo: UnsanitisedAddressListCsvS3Repository
 
 
 @pytest.fixture
@@ -76,7 +78,7 @@ def harness(db_engine: Engine) -> Iterator[Harness]:
         queue_url = cast(str, queue["QueueUrl"])
 
         csv_client = CsvS3Client(boto_s3, BUCKET)
-        repo = UserAddressCsvS3Repository(csv_client, BUCKET)
+        repo = UnsanitisedAddressListCsvS3Repository(csv_client, BUCKET)
         queue_client = Address2UprnQueueClient(boto_sqs, queue_url)
 
         # DB: ephemeral PostgreSQL TaskOrchestrator
@@ -89,7 +91,7 @@ def harness(db_engine: Engine) -> Iterator[Harness]:
 
             splitter = PostcodeSplitterOrchestrator(
                 task_orchestrator=task_orchestrator,
-                user_address_repo=repo,
+                unsanitised_address_repo=repo,
                 queue_client=queue_client,
                 max_batch_size=3,
             )
@@ -169,10 +171,8 @@ def test_split_and_dispatch_creates_three_children_for_fixture(
     harness: Harness,
 ) -> None:
     # arrange
-    parent_task, parent_subtask = (
-        harness.task_orchestrator.create_task_with_subtask(
-            task_source="manual:postcode-splitter-int"
-        )
+    parent_task, parent_subtask = harness.task_orchestrator.create_task_with_subtask(
+        task_source="manual:postcode-splitter-int"
     )
     input_uri = _upload_fixture_csv(harness.csv_client)
 
@@ -197,10 +197,8 @@ def test_split_and_dispatch_persists_child_inputs_with_task_id_and_s3_uri(
     harness: Harness,
 ) -> None:
     # arrange
-    parent_task, parent_subtask = (
-        harness.task_orchestrator.create_task_with_subtask(
-            task_source="manual:postcode-splitter-int"
-        )
+    parent_task, parent_subtask = harness.task_orchestrator.create_task_with_subtask(
+        task_source="manual:postcode-splitter-int"
     )
     input_uri = _upload_fixture_csv(harness.csv_client)
 
@@ -230,10 +228,8 @@ def test_split_and_dispatch_publishes_one_message_per_child_with_matching_ids(
     harness: Harness,
 ) -> None:
     # arrange
-    parent_task, parent_subtask = (
-        harness.task_orchestrator.create_task_with_subtask(
-            task_source="manual:postcode-splitter-int"
-        )
+    parent_task, parent_subtask = harness.task_orchestrator.create_task_with_subtask(
+        task_source="manual:postcode-splitter-int"
     )
     input_uri = _upload_fixture_csv(harness.csv_client)
 
@@ -267,10 +263,8 @@ def test_split_and_dispatch_returns_child_ids_in_dispatch_order(
     harness: Harness,
 ) -> None:
     # arrange
-    parent_task, parent_subtask = (
-        harness.task_orchestrator.create_task_with_subtask(
-            task_source="manual:postcode-splitter-int"
-        )
+    parent_task, parent_subtask = harness.task_orchestrator.create_task_with_subtask(
+        task_source="manual:postcode-splitter-int"
     )
     input_uri = _upload_fixture_csv(harness.csv_client)
 

tests/repositories/unsanitised_address/test_unsanitised_address_list_csv_s3_repository.py

@@ -3,11 +3,11 @@ from collections.abc import Iterator
 import pytest
 from moto import mock_aws
 
-from domain.addresses.user_address import LandlordAssetList
+from domain.addresses.unsanitised_address import AddressList, UnsanitisedAddress
 from domain.postcode import Postcode
 from infrastructure.csv_s3_client import CsvS3Client
-from repositories.user_address.user_address_csv_s3_repository import (
-    UserAddressCsvS3Repository,
+from repositories.unsanitised_address.unsanitised_address_list_csv_s3_repository import (
+    UnsanitisedAddressListCsvS3Repository,
 )
 from tests.infrastructure import make_boto_client
 
@@ -15,22 +15,22 @@ BUCKET = "user-address-bucket"
 
 
 @pytest.fixture
-def repo() -> Iterator[UserAddressCsvS3Repository]:
+def repo() -> Iterator[UnsanitisedAddressListCsvS3Repository]:
     with mock_aws():
         boto_client = make_boto_client("s3")
         boto_client.create_bucket(Bucket=BUCKET)
         csv_client = CsvS3Client(boto_client, BUCKET)
-        yield UserAddressCsvS3Repository(csv_client, BUCKET)
+        yield UnsanitisedAddressListCsvS3Repository(csv_client, BUCKET)
 
 
 def _upload_csv(
-    repo: UserAddressCsvS3Repository, rows: list[dict[str, str]], key: str
+    repo: UnsanitisedAddressListCsvS3Repository, rows: list[dict[str, str]], key: str
 ) -> str:
     return repo._csv_client.save_rows(rows, key)  # pyright: ignore[reportPrivateUsage]
 
 
 def test_load_batch_parses_address_postcode_and_reference(
-    repo: UserAddressCsvS3Repository,
+    repo: UnsanitisedAddressListCsvS3Repository,
 ) -> None:
     # arrange
     rows = [
@@ -50,13 +50,13 @@ def test_load_batch_parses_address_postcode_and_reference(
     # assert
     assert len(addresses) == 1
     address = addresses[0]
-    assert address.user_address == "1 High Street, Flat 2, Townville"
+    assert address.address == "1 High Street, Flat 2, Townville"
     assert address.postcode == Postcode("SW1A1AA")
-    assert address.internal_reference == "REF-001"
+    assert address.org_reference == "REF-001"
 
 
 def test_load_batch_uses_only_address_1_when_others_missing(
-    repo: UserAddressCsvS3Repository,
+    repo: UnsanitisedAddressListCsvS3Repository,
 ) -> None:
     # arrange
     rows = [
@@ -75,13 +75,13 @@ def test_load_batch_uses_only_address_1_when_others_missing(
 
     # assert
     assert len(addresses) == 1
-    assert addresses[0].user_address == "10 Cardiff Road"
+    assert addresses[0].address == "10 Cardiff Road"
     assert addresses[0].postcode == Postcode("CF101AA")
-    assert addresses[0].internal_reference == "REF-002"
+    assert addresses[0].org_reference == "REF-002"
 
 
 def test_load_batch_handles_missing_internal_reference(
-    repo: UserAddressCsvS3Repository,
+    repo: UnsanitisedAddressListCsvS3Repository,
 ) -> None:
     # arrange
     rows = [
@@ -100,16 +100,16 @@ def test_load_batch_handles_missing_internal_reference(
 
     # assert
     assert len(addresses) == 1
-    assert addresses[0].user_address == "5 Park Lane"
+    assert addresses[0].address == "5 Park Lane"
     assert addresses[0].postcode == Postcode("M11AA")
-    assert addresses[0].internal_reference is None
+    assert addresses[0].org_reference is None
 
 
 def test_load_batch_captures_full_source_row(
-    repo: UserAddressCsvS3Repository,
+    repo: UnsanitisedAddressListCsvS3Repository,
 ) -> None:
     # A raw EPC-export-shaped row: the splitter must preserve every column,
-    # not just the ones it parses into UserAddress fields.
+    # not just the ones it parses into UnsanitisedAddress fields.
     # arrange
     row = {
         "Asset Reference": "511",
@@ -124,11 +124,11 @@ def test_load_batch_captures_full_source_row(
     addresses = repo.load_batch(uri)
 
     # assert
-    assert addresses[0].landlord_additional_info == row
+    assert addresses[0].additional_info == row
 
 
 def test_load_batch_raises_when_postcode_column_absent(
-    repo: UserAddressCsvS3Repository,
+    repo: UnsanitisedAddressListCsvS3Repository,
 ) -> None:
     # arrange
     rows = [{"Address 1": "1 High Street", "Property Type": "Flat"}]
@@ -140,7 +140,7 @@ def test_load_batch_raises_when_postcode_column_absent(
 
 
 def test_save_batch_passes_through_all_columns_and_appends_postcode_clean(
-    repo: UserAddressCsvS3Repository,
+    repo: UnsanitisedAddressListCsvS3Repository,
 ) -> None:
     # arrange
     row = {
@@ -169,19 +169,21 @@ def test_save_batch_passes_through_all_columns_and_appends_postcode_clean(
 
 
 def test_save_batch_returns_uri_under_path_prefix(
-    repo: UserAddressCsvS3Repository,
+    repo: UnsanitisedAddressListCsvS3Repository,
 ) -> None:
     # arrange
-    addresses = [
-        LandlordAssetList(
-            user_address="1 High Street",
-            postcode=Postcode("SW1A 1AA"),
-            landlord_additional_info={
-                "Address 1": "1 High Street",
-                "postcode": "SW1A 1AA",
-            },
-        ),
-    ]
+    addresses = AddressList(
+        [
+            UnsanitisedAddress(
+                address="1 High Street",
+                postcode=Postcode("SW1A 1AA"),
+                additional_info={
+                    "Address 1": "1 High Street",
+                    "postcode": "SW1A 1AA",
+                },
+            ),
+        ]
+    )
 
     # act
     uri = repo.save_batch(addresses, "tasks/abc/batches")
@@ -192,7 +194,7 @@ def test_save_batch_returns_uri_under_path_prefix(
 
 
 def test_save_then_reload_round_trip_preserves_columns(
-    repo: UserAddressCsvS3Repository,
+    repo: UnsanitisedAddressListCsvS3Repository,
 ) -> None:
     # arrange
     rows = [
@@ -225,19 +227,21 @@ def test_save_then_reload_round_trip_preserves_columns(
 
 
 def test_save_batch_uses_unique_filename_per_call(
-    repo: UserAddressCsvS3Repository,
+    repo: UnsanitisedAddressListCsvS3Repository,
 ) -> None:
     # arrange
-    addresses = [
-        LandlordAssetList(
-            user_address="1 High Street",
-            postcode=Postcode("SW1A 1AA"),
-            landlord_additional_info={
-                "Address 1": "1 High Street",
-                "postcode": "SW1A 1AA",
-            },
-        ),
-    ]
+    addresses = AddressList(
+        [
+            UnsanitisedAddress(
+                address="1 High Street",
+                postcode=Postcode("SW1A 1AA"),
+                additional_info={
+                    "Address 1": "1 High Street",
+                    "postcode": "SW1A 1AA",
+                },
+            ),
+        ]
+    )
 
     # act
     uri_1 = repo.save_batch(addresses, "tasks/uniqueness")