catch json ValueError and provide example schema (#281)

Closes #271 

Catches ValueError raised by partition() and transforms it to a 400
error. Also provides an example of the Unstructured schema.

CHANGELOG.md

@@ -2,7 +2,8 @@
 
 * Bump unstructured to 0.10.21
 * Fix an unhandled error when a non pdf file is sent with content-type pdf
-* Fix unhandled error when a non docx file is sent with content-type docx
+* Fix an unhandled error when a non docx file is sent with content-type docx
+* Fix an unhandled error when a non-Unstructured json schema is sent
 
 ## 0.0.51
 

prepline_general/api/general.py

@@ -457,6 +457,11 @@ def pipeline_api(
             raise HTTPException(
                 status_code=400, detail=f"{file_content_type} not currently supported"
             )
+        if "Unstructured schema" in e.args[0]:
+            raise HTTPException(
+                status_code=400,
+                detail="Json schema does not match the Unstructured schema",
+            )
         raise e
     except zipfile.BadZipFile as e:
         if "File is not a zip file" in e.args[0]:

test_general/api/test_app.py

@@ -1,4 +1,5 @@
 from pathlib import Path
+import os
 
 import io
 import pytest
@@ -760,3 +761,29 @@ def test_general_api_returns_400_bad_docx():
     )
     assert "txt is not a valid" in response.json().get("detail")
     assert response.status_code == 400
+
+
+def test_general_api_returns_400_bad_json(tmpdir):
+    """
+    Verify that we get a 400 for invalid json schemas
+    """
+    client = TestClient(app)
+    data = '{"hi": "there"}'
+
+    filepath = os.path.join(tmpdir, "unprocessable.json")
+    with open(filepath, "w") as f:
+        f.write(data)
+    response = client.post(
+        MAIN_API_ROUTE,
+        files=[
+            (
+                "files",
+                (
+                    str(filepath),
+                    open(filepath, "rb"),
+                ),
+            )
+        ],
+    )
+    assert "Unstructured schema" in response.json().get("detail")
+    assert response.status_code == 400