Handle unknown hi_res_model_name

CHANGELOG.md

@@ -1,3 +1,7 @@
+## 0.0.61-dev0
+
+* Handle invalid hi_res_model_name kwarg
+
 ## 0.0.60
 
 * Enable self-hosted authorization using UNSTRUCTURED_API_KEY env variable

prepline_general/api/app.py

@@ -11,7 +11,7 @@
 app = FastAPI(
     title="Unstructured Pipeline API",
     description="""""",
-    version="0.0.60",
+    version="0.0.61",
     docs_url="/general/docs",
     openapi_url="/general/openapi.json",
 )

prepline_general/api/general.py

@@ -47,6 +47,7 @@
     elements_from_json,
 )
 from unstructured_inference.models.chipper import MODEL_TYPES as CHIPPER_MODEL_TYPES
+from unstructured_inference.models.base import UnknownModelException
 
 
 app = FastAPI()
@@ -470,6 +471,12 @@ def pipeline_api(
             detail="File is not a valid docx",
         )
 
+    except UnknownModelException:
+        raise HTTPException(
+            status_code=400,
+            detail=f"Unknown model type: {hi_res_model_name}",
+        )
+
     # Clean up returned elements
     # Note(austin): pydantic should control this sort of thing for us
     for i, element in enumerate(elements):
@@ -675,7 +682,7 @@ def return_content_type(filename):
 
 
 @router.post("/general/v0/general")
-@router.post("/general/v0.0.60/general")
+@router.post("/general/v0.0.61/general")
 def pipeline_1(
     request: Request,
     gz_uncompressed_content_type: Optional[str] = Form(default=None),

preprocessing-pipeline-family.yaml

@@ -1,2 +1,2 @@
 name: general
-version: 0.0.60
+version: 0.0.61

test_general/api/test_app.py

@@ -900,3 +900,19 @@ def test_chipper_not_available_errors(monkeypatch, mocker, exception, status_cod
 
     assert resp.status_code == status_code
     assert resp.json().get("detail") == message
+
+
+def test_invalid_hi_res_model_name_returns_400():
+    """Verify that we get a 400 if we pass in a bad model_name"""
+    client = TestClient(app)
+    test_file = Path("sample-docs") / "layout-parser-paper-fast.pdf"
+    response = client.post(
+        MAIN_API_ROUTE,
+        files=[("files", (str(test_file), open(test_file, "rb")))],
+        data={
+            "strategy": "hi_res",
+            "hi_res_model_name": "invalid_model",
+        },
+    )
+    assert response.status_code == 400
+    assert "Unknown model type" in response.text