Fix no-implicit-optional change in mypy/pep484

python/typing#275

Author: chr1st1ank

narrow_down/_minhash.py

@@ -106,7 +106,7 @@ def __init__(
         self.rows_per_band = lsh_config.rows_per_band
         self._hashfunc = hash_.murmur3_32bit
 
-    def _hash(self, arr: npt.NDArray, exact_part: str = None) -> int:
+    def _hash(self, arr: npt.NDArray, exact_part: Optional[str] = None) -> int:
         """Merge multiple hashes together to one hash."""
         if arr.dtype != np.uint32:
             # Other data types like the standard int64 have a different binary representation
@@ -176,7 +176,7 @@ async def remove_by_id(self, document_id: int, check_if_exists: bool = False) ->
         await self._storage.remove_document(document_id=document_id)
 
     async def query(
-        self, fingerprint: Fingerprint, *, exact_part: str = None
+        self, fingerprint: Fingerprint, *, exact_part: Optional[str] = None
     ) -> Collection[StoredDocument]:
         """Find all similar documents."""
         tasks = []
@@ -192,7 +192,7 @@ async def query(
         return await self._query_documents(list(candidates))
 
     async def query_top_n(
-        self, n, fingerprint: Fingerprint, *, exact_part: str = None
+        self, n, fingerprint: Fingerprint, *, exact_part: Optional[str] = None
     ) -> Collection[StoredDocument]:
         """Find n most similar documents."""
         tasks = []

narrow_down/_rust.pyi

@@ -17,7 +17,7 @@ class RustMemoryStore:
     def from_file(cls, file_path: str) -> "RustMemoryStore": ...
     def insert_setting(self, key: str, value: str): ...
     def query_setting(self, key: str) -> Optional[str]: ...
-    def insert_document(self, document: bytes, document_id: int = None) -> int: ...
+    def insert_document(self, document: bytes, document_id: Optional[int] = None) -> int: ...
     def query_document(self, document_id: int) -> bytes: ...
     def remove_document(self, document_id: int): ...
     def add_document_to_bucket(self, bucket_id: int, document_hash: int, document_id: int): ...

narrow_down/scylladb.py

@@ -49,7 +49,7 @@ def __init__(
         self,
         cluster_or_session: Union[cassandra.cluster.Cluster, cassandra.cluster.Session],
         keyspace: str,
-        table_prefix: str = None,
+        table_prefix: Optional[str] = None,
     ) -> None:
         """Create a new empty or connect to an existing SQLite database.
 
@@ -198,7 +198,7 @@ async def query_setting(self, key: str) -> Optional[str]:
                     return None
                 raise  # Don't swallow unknown errors
 
-    async def insert_document(self, document: bytes, document_id: int = None) -> int:
+    async def insert_document(self, document: bytes, document_id: Optional[int] = None) -> int:
         """Add the data of a document to the storage and return its ID."""
         with self._session() as session:
             if document_id:

narrow_down/similarity_store.py

@@ -1,7 +1,7 @@
 """High-level API for indexing and retrieval of documents."""
 import re
 import warnings
-from typing import Callable, Collection, Iterable, List, Union
+from typing import Callable, Collection, Iterable, List, Optional, Union
 
 from narrow_down import _minhash, _tokenize
 from narrow_down._minhash import MinhashLshConfig
@@ -47,9 +47,9 @@ def __init__(self):  # noqa: D107  # Not meant to be called, therefore omitting
     async def create(
         cls,
         *,
-        storage: StorageBackend = None,
+        storage: Optional[StorageBackend] = None,
         storage_level: StorageLevel = StorageLevel.Minimal,
-        tokenize: Union[str, Callable[[str], Collection[str]]] = None,
+        tokenize: Optional[Union[str, Callable[[str], Collection[str]]]] = None,
         max_false_negative_proba: float = 0.05,
         max_false_positive_proba: float = 0.05,
         similarity_threshold: float = 0.75,
@@ -108,7 +108,9 @@ async def create(
 
     @classmethod
     async def load_from_storage(
-        cls, storage: StorageBackend, tokenize: Union[str, Callable[[str], Collection[str]]] = None
+        cls,
+        storage: StorageBackend,
+        tokenize: Optional[Union[str, Callable[[str], Collection[str]]]] = None,
     ) -> "SimilarityStore":
         """Load a SimilarityStore object from already initialized storage.
 
@@ -213,13 +215,18 @@ async def _initialize_storage(self):
         await self._storage.initialize()
         await self._storage.insert_setting("similarity_threshold", str(self._similarity_threshold))
         await self._storage.insert_setting("storage_level", str(self._storage_level.value))
-        await self._storage.insert_setting("tokenize", self._tokenize)
+        await self._storage.insert_setting("tokenize", self._tokenize)  # type: ignore
         await self._storage.insert_setting("lsh_config", self._lsh_config.to_json())
         self._minhasher = _minhash.MinHasher(n_hashes=self._lsh_config.n_hashes)
         self._lsh = _minhash.LSH(self._lsh_config, storage=self._storage)
 
     async def insert(
-        self, document: str, *, document_id: int = None, exact_part: str = None, data: str = None
+        self,
+        document: str,
+        *,
+        document_id: Optional[int] = None,
+        exact_part: Optional[str] = None,
+        data: Optional[str] = None,
     ) -> int:
         """Index a new document.
 
@@ -282,7 +289,7 @@ def _filter_candidates(self, candidates, tokens, exact_part) -> List[StoredDocum
         return candidates
 
     async def query(
-        self, document: str, *, exact_part: str = None, validate: bool = None
+        self, document: str, *, exact_part: Optional[str] = None, validate: Optional[bool] = None
     ) -> Collection[StoredDocument]:
         """Query all similar documents.
 
@@ -305,7 +312,12 @@ async def query(
         return candidates
 
     async def query_top_n(
-        self, n: int, document: str, *, exact_part: str = None, validate: bool = None
+        self,
+        n: int,
+        document: str,
+        *,
+        exact_part: Optional[str] = None,
+        validate: Optional[bool] = None,
     ) -> Collection[StoredDocument]:
         """Query the top n similar documents.
 

narrow_down/sqlite.py

@@ -85,7 +85,7 @@ def _query_setting_sync(self, key: str) -> Optional[str]:
                 return None
             raise
 
-    async def insert_document(self, document: bytes, document_id: int = None) -> int:
+    async def insert_document(self, document: bytes, document_id: Optional[int] = None) -> int:
         """Add the data of a document to the storage and return its ID."""
         with self._connection as conn:
             if document_id:

narrow_down/storage.py

@@ -136,7 +136,7 @@ async def query_setting(self, key: str) -> Optional[str]:
         raise NotImplementedError
 
     @abstractmethod
-    async def insert_document(self, document: bytes, document_id: int = None) -> int:
+    async def insert_document(self, document: bytes, document_id: Optional[int] = None) -> int:
         """Add the data of a document to the storage and return its ID."""
         raise NotImplementedError()
 
@@ -228,7 +228,7 @@ async def query_setting(self, key: str) -> Optional[str]:
         """Query a setting with the given key."""
         return self.rms.query_setting(key)
 
-    async def insert_document(self, document: bytes, document_id: int = None) -> int:
+    async def insert_document(self, document: bytes, document_id: Optional[int] = None) -> int:
         """Add the data of a document to the storage and return its ID."""
         return self.rms.insert_document(document, document_id)
 

pyproject.toml

@@ -249,6 +249,12 @@ disable = [
     "abstract-class-instantiated",
 ]
 
+[tool.mypy]
+follow_imports = "silent"
+python_version = "3.8"
+
+check_untyped_defs = true
+
 [[tool.mypy.overrides]]
 module = [
     "pytest",

tasks.py

@@ -194,7 +194,7 @@ def lint(c):
 def mypy(c):
     # type: (Context) -> None
     """Run mypy."""
-    _run(c, f"mypy --follow-imports silent --python-version 3.8 {PYTHON_TARGETS_STR}")
+    _run(c, f"mypy --config-file pyproject.toml {SOURCE_DIR}")
 
 
 @task()