clean up http status code usage

Author: jgbradley1

backend/graphrag_app/api/data.py

@@ -47,7 +47,7 @@
     "",
     summary="Get list of data containers.",
     response_model=StorageNameList,
-    responses={200: {"model": StorageNameList}},
+    responses={status.HTTP_200_OK: {"model": StorageNameList}},
 )
 async def get_all_data_containers():
     """
@@ -169,20 +169,20 @@ async def upload_files(
 
         if len(processing_errors) > 0:
             raise HTTPException(
-                status_code=400,
+                status_code=status.HTTP_422_UNPROCESSABLE_ENTITY,
                 detail=f"Error uploading files: {processing_errors}.",
             )
-        return BaseResponse(status="File upload successful.")
+        return BaseResponse(status="Success.")
     except Exception as e:
         logger = load_pipeline_logger()
         logger.error(
             message="Error uploading files.",
             cause=e,
             stack=traceback.format_exc(),
-            details={"files": [f.filename for f in files]},
+            details={"files": processing_errors},
         )
         raise HTTPException(
-            status_code=500,
+            status_code=status.HTTP_500_INTERNAL_SERVER_ERROR,
             detail=f"Error uploading files to container '{container_name}'.",
         )
 
@@ -191,7 +191,7 @@ async def upload_files(
     "/{container_name}",
     summary="Delete a data storage container",
     response_model=BaseResponse,
-    responses={200: {"model": BaseResponse}},
+    responses={status.HTTP_200_OK: {"model": BaseResponse}},
 )
 async def delete_files(
     container_name: str, sanitized_container_name: str = Depends(sanitize_name)

backend/graphrag_app/api/index.py

@@ -72,7 +72,7 @@ async def schedule_index_job(
         sanitized_storage_container_name
     ).exists():
         raise HTTPException(
-            status_code=500,
+            status_code=status.HTTP_412_PRECONDITION_FAILED,
             detail=f"Storage container '{storage_container_name}' does not exist",
         )
 
@@ -102,7 +102,7 @@ async def schedule_index_job(
             PipelineJobState(existing_job.status) == PipelineJobState.RUNNING
         ):
             raise HTTPException(
-                status_code=202,  # request has been accepted for processing but is not complete.
+                status_code=status.HTTP_425_TOO_EARLY,  # request has been accepted for processing but is not complete.
                 detail=f"Index '{index_container_name}' already exists and has not finished building.",
             )
         # if indexing job is in a failed state, delete the associated K8s job and pod to allow for a new job to be scheduled
@@ -258,7 +258,8 @@ async def delete_index(
             details={"container": container_name},
         )
         raise HTTPException(
-            status_code=500, detail=f"Error deleting '{container_name}'."
+            status_code=status.HTTP_500_INTERNAL_SERVER_ERROR,
+            detail=f"Error deleting '{container_name}'.",
         )
 
     return BaseResponse(status="Success")
@@ -277,7 +278,7 @@ async def get_index_status(
     if pipelinejob.item_exist(sanitized_container_name):
         pipeline_job = pipelinejob.load_item(sanitized_container_name)
         return IndexStatusResponse(
-            status_code=200,
+            status_code=status.HTTP_200_OK,
             index_name=pipeline_job.human_readable_index_name,
             storage_name=pipeline_job.human_readable_storage_name,
             status=pipeline_job.status.value,
@@ -286,5 +287,6 @@ async def get_index_status(
         )
     else:
         raise HTTPException(
-            status_code=404, detail=f"'{container_name}' does not exist."
+            status_code=status.HTTP_404_NOT_FOUND,
+            detail=f"'{container_name}' does not exist.",
         )

backend/graphrag_app/api/query.py

@@ -52,7 +52,7 @@ async def global_query(request: GraphRequest):
 
     if not _is_index_complete(sanitized_index_name):
         raise HTTPException(
-            status_code=500,
+            status_code=status.HTTP_425_TOO_EARLY,
             detail=f"{index_name} not ready for querying.",
         )
 
@@ -131,7 +131,7 @@ async def local_query(request: GraphRequest):
 
     if not _is_index_complete(sanitized_index_name):
         raise HTTPException(
-            status_code=500,
+            status_code=status.HTTP_425_TOO_EARLY,
             detail=f"{index_name} not ready for querying.",
         )
 

backend/graphrag_app/main.py

@@ -33,7 +33,7 @@
 
 
 async def catch_all_exceptions_middleware(request: Request, call_next):
-    """a function to globally catch all exceptions and return a 500 response with the exception message"""
+    """A global function to catch all exceptions and produce a standard error message"""
     try:
         return await call_next(request)
     except Exception as e:
@@ -44,7 +44,10 @@ async def catch_all_exceptions_middleware(request: Request, call_next):
             cause=e,
             stack=stack,
         )
-        return Response("Unexpected internal server error.", status_code=500)
+        return Response(
+            "Unexpected internal server error.",
+            status_code=status.HTTP_500_INTERNAL_SERVER_ERROR,
+        )
 
 
 # NOTE: this function is not currently used, but it is a placeholder for future use once RBAC issues have been resolved

backend/graphrag_app/utils/common.py

@@ -14,15 +14,18 @@
 from azure.cosmos import ContainerProxy, exceptions
 from azure.identity import DefaultAzureCredential
 from azure.storage.blob.aio import ContainerClient
-from fastapi import Header, HTTPException
+from fastapi import Header, HTTPException, status
 
 from graphrag_app.logger.load_logger import load_pipeline_logger
 from graphrag_app.utils.azure_clients import AzureClientManager
 
+FILE_UPLOAD_CACHE = "cache/uploaded_files.csv"
+
 
 def get_df(
     table_path: str,
 ) -> pd.DataFrame:
+    """Read a parquet file from Azure Storage and return it as a pandas DataFrame."""
     df = pd.read_parquet(
         table_path,
         storage_options=pandas_storage_options(),
@@ -126,7 +129,10 @@ def get_cosmos_container_store_client() -> ContainerProxy:
             cause=e,
             stack=traceback.format_exc(),
         )
-        raise HTTPException(status_code=500, detail="Error fetching cosmosdb client.")
+        raise HTTPException(
+            status_code=status.HTTP_500_INTERNAL_SERVER_ERROR,
+            detail="Error fetching cosmosdb client.",
+        )
 
 
 async def get_blob_container_client(name: str) -> ContainerClient:
@@ -144,7 +150,10 @@ async def get_blob_container_client(name: str) -> ContainerClient:
             cause=e,
             stack=traceback.format_exc(),
         )
-        raise HTTPException(status_code=500, detail="Error fetching storage client.")
+        raise HTTPException(
+            status_code=status.HTTP_500_INTERNAL_SERVER_ERROR,
+            detail="Error fetching storage client.",
+        )
 
 
 def sanitize_name(container_name: str) -> str:
@@ -191,22 +200,23 @@ def desanitize_name(sanitized_container_name: str) -> str | None:
             return None
     except Exception:
         raise HTTPException(
-            status_code=500, detail="Error retrieving original container name."
+            status_code=status.HTTP_500_INTERNAL_SERVER_ERROR,
+            detail="Error retrieving original container name.",
         )
 
 
 async def subscription_key_check(
     Ocp_Apim_Subscription_Key: Annotated[str, Header()],
 ):
     """
-    Verifies if user has passed the Ocp_Apim_Subscription_Key (APIM subscription key) in the request header.
-    If it is not present, an HTTPException with a 400 status code is raised.
-    Note: this check is unnecessary (APIM validates subscription keys automatically), but this will add the key
+    Verify if user has passed the Ocp_Apim_Subscription_Key (APIM subscription key) in the request header.
+    Note: this check is unnecessary (APIM validates subscription keys automatically), but it effectively adds the key
     as a required parameter in the swagger docs page, enabling users to send requests using the swagger docs "Try it out" feature.
     """
     if not Ocp_Apim_Subscription_Key:
         raise HTTPException(
-            status_code=400, detail="Ocp-Apim-Subscription-Key required"
+            status_code=status.HTTP_400_BAD_REQUEST,
+            detail="Ocp-Apim-Subscription-Key required",
         )
     return Ocp_Apim_Subscription_Key
 
@@ -216,7 +226,7 @@ async def create_cache(container_client: ContainerClient) -> None:
     Create a file cache (csv) to track uploaded files.
     """
     try:
-        cache_blob_client = container_client.get_blob_client("uploaded_files_cache.csv")
+        cache_blob_client = container_client.get_blob_client(FILE_UPLOAD_CACHE)
         if not await cache_blob_client.exists():
             # create the empty file cache csv
             headers = [["Filename", "Hash"]]
@@ -231,20 +241,20 @@ async def create_cache(container_client: ContainerClient) -> None:
                 os.remove("uploaded_files_cache.csv")
     except Exception:
         raise HTTPException(
-            status_code=500,
+            status_code=status.HTTP_500_INTERNAL_SERVER_ERROR,
             detail="Error creating file cache in Azure Blob Storage.",
         )
 
 
 async def check_cache(file_stream: BinaryIO, container_client: ContainerClient) -> bool:
     """
-    Check a cache file to determine if a file has previously been uploaded.
+    Check a file cache (csv) to determine if a file has previously been uploaded.
 
     Note: This function creates/checks a CSV file in azure storage to act as a cache of previously uploaded files.
     """
     try:
         # load the file cache
-        cache_blob_client = container_client.get_blob_client("uploaded_files_cache.csv")
+        cache_blob_client = container_client.get_blob_client(FILE_UPLOAD_CACHE)
         cache_download_stream = await cache_blob_client.download_blob()
         cache_bytes = await cache_download_stream.readall()
         cache_content = StringIO(cache_bytes.decode("utf-8"))
@@ -258,7 +268,7 @@ async def check_cache(file_stream: BinaryIO, container_client: ContainerClient)
         return False
     except Exception:
         raise HTTPException(
-            status_code=500,
+            status_code=status.HTTP_500_INTERNAL_SERVER_ERROR,
             detail="Error checking file cache in Azure Blob Storage.",
         )
 
@@ -267,11 +277,11 @@ async def update_cache(
     filename: str, file_stream: BinaryIO, container_client: ContainerClient
 ) -> None:
     """
-    Update the file cache (csv) with a new file by adding a new row.
+    Update a file cache (csv) with a new file by adding a new row.
     """
     try:
         # Load the file cache
-        cache_blob_client = container_client.get_blob_client("uploaded_files_cache.csv")
+        cache_blob_client = container_client.get_blob_client(FILE_UPLOAD_CACHE)
         cache_download_stream = await cache_blob_client.download_blob()
         cache_bytes = await cache_download_stream.readall()
         cache_content = StringIO(cache_bytes.decode("utf-8"))
@@ -297,6 +307,6 @@ async def update_cache(
         )
     except Exception:
         raise HTTPException(
-            status_code=500,
+            status_code=status.HTTP_500_INTERNAL_SERVER_ERROR,
             detail="Error updating file cache in Azure Blob Storage.",
         )