Refactor libs/oracledb to meet mypy standard

Author: paxiaatucsdedu

libs/oracledb/langchain_oracledb/document_loaders/oracleadb_loader.py

@@ -95,7 +95,10 @@ def _run_query(self) -> List[Dict[str, Any]]:
                 cursor.execute(self.query, self.parameter)
             else:
                 cursor.execute(self.query)
-            columns = [col[0] for col in cursor.description]
+
+            columns = (
+                [col[0] for col in cursor.description] if cursor.description else []
+            )
             data = cursor.fetchall()
             data = [
                 {

libs/oracledb/langchain_oracledb/document_loaders/oracleai.py

@@ -130,7 +130,7 @@ def read_file(
             with open(file_path, "rb") as f:
                 data = f.read()
 
-            if data is None:
+            if not data:
                 return Document(page_content="", metadata=metadata)
 
             mdata = cursor.var(oracledb.DB_TYPE_CLOB)
@@ -151,7 +151,7 @@ def read_file(
             )
             cursor.close()
 
-            if mdata is None:
+            if mdata.getvalue() is None:
                 metadata = {}
             else:
                 doc_data = str(mdata.getvalue())
@@ -166,7 +166,7 @@ def read_file(
             metadata["_oid"] = doc_id
             metadata["_file"] = file_path
 
-            if text is None:
+            if text.getvalue() is None:
                 return Document(page_content="", metadata=metadata)
             else:
                 return Document(page_content=str(text.getvalue()), metadata=metadata)

libs/oracledb/langchain_oracledb/embeddings/oracleai.py

@@ -102,9 +102,6 @@ def embed_documents(self, texts: List[str]) -> List[List[float]]:
             List of embeddings, one for each input text.
         """
 
-        if texts is None:
-            return None
-
         embeddings: List[List[float]] = []
         try:
             # returns strings or bytes instead of a locator

libs/oracledb/langchain_oracledb/utilities/oracleai.py

@@ -92,7 +92,7 @@ def get_summary(self, docs: Any) -> List[str]:
                     summ=summary,
                 )
 
-                if summary is None:
+                if summary.getvalue() is None:
                     results.append("")
                 else:
                     results.append(str(summary.getvalue()))
@@ -114,7 +114,7 @@ def get_summary(self, docs: Any) -> List[str]:
                     summ=summary,
                 )
 
-                if summary is None:
+                if summary.getvalue() is None:
                     results.append("")
                 else:
                     results.append(str(summary.getvalue()))

libs/oracledb/langchain_oracledb/vectorstores/oraclevs.py

@@ -1330,7 +1330,7 @@ async def _aembed_documents(self, texts: List[str]) -> List[List[float]]:
         if isinstance(self.embedding_function, Embeddings):
             return await self.embedding_function.aembed_documents(texts)
         elif inspect.isawaitable(self.embedding_function):
-            return [await self.embedding_function(text) for text in texts]
+            return [await self.embedding_function(text) for text in texts]  # type: ignore[unreachable]
         elif callable(self.embedding_function):
             return [self.embedding_function(text) for text in texts]
         else:
@@ -1348,7 +1348,7 @@ async def _aembed_query(self, text: str) -> List[float]:
         if isinstance(self.embedding_function, Embeddings):
             return await self.embedding_function.aembed_query(text)
         elif inspect.isawaitable(self.embedding_function):
-            return await self.embedding_function(text)
+            return await self.embedding_function(text)  # type: ignore[unreachable]
         else:
             return self.embedding_function(text)
 

libs/oracledb/pyproject.toml

@@ -72,20 +72,17 @@ ignore = [
 docstring-code-format = true
 
 [tool.mypy]
-ignore_missing_imports = "True"
-
-# Disable specific error codes that are causing issues
-disallow_untyped_defs = "False"
-disable_error_code = ["attr-defined", "assignment", "var-annotated", "override", "union-attr", "arg-type"]
-
-# TODO: LangChain Google settings
-# plugins = ["pydantic.mypy"]
-# strict = true
-# disallow_untyped_defs = true
-
-# # TODO: activate for 'strict' checking
-# disallow_any_generics = false
-# warn_return_any = false
+plugins = ["pydantic.mypy"]
+check_untyped_defs = true
+error_summary = false
+pretty = true
+show_column_numbers = true
+show_error_codes = true
+show_error_context = true
+warn_redundant_casts = true
+warn_unreachable = true
+warn_unused_configs = true
+warn_unused_ignores = true
 
 [tool.coverage.run]
 omit = ["tests/*"]

libs/oracledb/tests/integration_tests/vectorstores/test_oraclevs.py

@@ -1918,7 +1918,7 @@ def test_db_filter_test() -> None:
 
         exception_occurred = False
         try:
-            db_filter_exc: dict = {  # type: ignore[typeddict-unknown-key]
+            db_filter_exc: dict = {
                 "_xor": [  # Incorrect operation _xor
                     {"order": {"$lte": 4}},
                     {"id": "st"},
@@ -2055,7 +2055,7 @@ async def test_db_filter_test_async() -> None:
 
         exception_occurred = False
         try:
-            db_filter_exc: dict = {  # type: ignore[typeddict-unknown-key]
+            db_filter_exc: dict = {
                 "_xor": [  # Incorrect operation _xor
                     {"order": {"$lte": 4}},
                     {"id": "st"},
@@ -2213,11 +2213,11 @@ def test_from_texts_lobs() -> None:
     metadatas = [
         {
             "id": "cncpt_15.5.3.2.2_P4",
-            "link": "https://docs.oracle.com/en/database/oracle/oracle-database/23/cncpt/logical-storage-structures.html#GUID-5387D7B2-C0CA-4C1E-811B-C7EB9B636442",  # type: ignore[E501]
+            "link": "https://docs.oracle.com/en/database/oracle/oracle-database/23/cncpt/logical-storage-structures.html#GUID-5387D7B2-C0CA-4C1E-811B-C7EB9B636442",
         },
         {
             "id": "cncpt_15.5.5_P1",
-            "link": "https://docs.oracle.com/en/database/oracle/oracle-database/23/cncpt/logical-storage-structures.html#GUID-D02B2220-E6F5-40D9-AFB5-BC69BCEF6CD4",  # type: ignore[E501]
+            "link": "https://docs.oracle.com/en/database/oracle/oracle-database/23/cncpt/logical-storage-structures.html#GUID-D02B2220-E6F5-40D9-AFB5-BC69BCEF6CD4",
         },
     ]
 
@@ -2613,7 +2613,7 @@ def test_filters() -> None:
     except Exception:
         sys.exit(1)
 
-    def model1(_) -> list[float]:  # type: ignore[no-untyped-def]
+    def model1(_) -> list[float]:
         return [0.1, 0.2, 0.3]
 
     # model1 = lambda x: [0.1, 0.2, 0.3]
@@ -2804,7 +2804,7 @@ async def test_filters_async() -> None:
     except Exception:
         sys.exit(1)
 
-    def model1(_) -> list[float]:  # type: ignore[no-untyped-def]
+    def model1(_) -> list[float]:
         return [0.1, 0.2, 0.3]
 
     # model1 = lambda x: [0.1, 0.2, 0.3]