Fix minor issues found in PR review

Author: cbrown350

backend/onyx/connectors/paperless_ngx/connector.py

@@ -38,6 +38,8 @@
 DOCUMENT_TYPES_ENDPOINT = "/api/document_types/"
 PROFILE_ENDPOINT = "/api/profile/"
 
+PAPERLESS_NGX_DATE_FORMAT = "%Y-%m-%dT%H:%M:%S+00:00"
+
 # TODO: handle custom fields?
 # CUSTOM_FIELDS_ENDPOINT = "/api/custom_fields/"
 
@@ -124,24 +126,26 @@ def __init__(
                         break
                     except ValueError:
                         continue
-                if not hasattr(self, "ui_min_start_date"):
+                if not self.master_start_date:
                     raise ValueError("Could not parse date with any format")
             except ValueError as e:
                 raise ConnectorValidationError(
                     f"Could not parse ui_min_start_date: {ui_min_start_date}. Error: {e}"
                 )
 
         # match ui_name in Datefield from ui_date_field str from UI
+        self.master_date_field: Optional[DateField] = DateField.MODIFIED_DATE
         if ui_date_field:
+            found = False
             for date_field in DateField:
                 if date_field.ui_name == ui_date_field:
                     self.master_date_field = date_field
+                    found = True
                     break
-            raise ConnectorValidationError(
-                f"Could not parse ui_date_field: {ui_date_field}."
-            )
-        else:
-            self.master_date_field = DateField.MODIFIED_DATE
+            if not found:
+                raise ConnectorValidationError(
+                    f"Could not parse ui_date_field: {ui_date_field}."
+                )
 
         logger.info("Initialized PaperlessNgxConnector")
 
@@ -499,7 +503,9 @@ def _parse_document(self, doc_data: Dict[str, Any]) -> Optional[Document]:
             logger.warning(
                 f"Could not parse timestamp for document {doc_id}: {modified_timestamp_str}"
             )
-            updated_at = datetime.now(timezone.utc)
+            raise ValueError(
+                f"Invalid timestamp found in document data {doc_id}: {modified_timestamp_str}"
+            )
 
         doc_owner = next(
             (user for user in self.all_users if user["id"] == doc_data.get("owner")),
@@ -625,8 +631,8 @@ def poll_source(
         end_dt = datetime.fromtimestamp(end, timezone.utc)
 
         # Format dates for API query in ISO format
-        start_iso = start_dt.strftime("%Y-%m-%dT%H:%M:%S+00:00")
-        end_iso = end_dt.strftime("%Y-%m-%dT%H:%M:%S+00:00")
+        start_iso = start_dt.strftime(PAPERLESS_NGX_DATE_FORMAT)
+        end_iso = end_dt.strftime(PAPERLESS_NGX_DATE_FORMAT)
 
         logger.info(
             f"Polling Paperless-ngx for documents modified between {start_iso} and {end_iso}."
@@ -680,12 +686,12 @@ def retrieve_all_slim_documents(
         # Apply date filtering if provided
         if start is not None:
             start_dt = datetime.fromtimestamp(start, timezone.utc)
-            start_iso = start_dt.strftime("%Y-%m-%d")
+            start_iso = start_dt.strftime(PAPERLESS_NGX_DATE_FORMAT)
             logger.info(f"Filtering slim documents from {start_iso}")
 
         if end is not None:
             end_dt = datetime.fromtimestamp(end, timezone.utc)
-            end_iso = end_dt.strftime("%Y-%m-%d")
+            end_iso = end_dt.strftime(PAPERLESS_NGX_DATE_FORMAT)
             logger.info(f"Filtering slim documents until {end_iso}")
 
         # Get only the IDs by making the request and extracting minimal information

backend/tests/daily/connectors/paperless_ngx/run_manual_test_paperless_server.sh

@@ -11,7 +11,7 @@ USERNAME="admin"
 PASSWORD="admin"
 
 # change to your local IP address if necessary
-HOST=$(ifconfig | grep -Eo 'inet (addr:)?([0-9]*\.){3}[0-9]*' | grep -Eo '([0-9]*\.){3}[0-9]*' | grep -v '127.0.0.1' | head -n 1) 
+HOST=$(ip addr | grep -Eo 'inet (addr:)?([0-9]*\.){3}[0-9]*' | grep -Eo '([0-9]*\.){3}[0-9]*' | grep -v '127.0.0.1' | head -n 1) 
 if [ -z "$HOST" ]; then
     HOST="localhost"
 fi
@@ -87,10 +87,10 @@ while ! curl -s "http://$HOST:$PORT" > /dev/null; do
 done
 echo -ne "\nPaperless-ngx is up and running!\n\n"
 
-# get the CSRF token
+# get the CSRF token in a cookies file
 curl -s -u $USERNAME:$PASSWORD "http://$HOST:$PORT/api/schema/view/" -H "Content-Type: application/json" -c paperless-test-cookies.txt -o /dev/null 
 
-# get the auth token
+# get the auth token and use the CSRF token from the cookies file
 AUTH_TOKEN=$(curl -s -u $USERNAME:$PASSWORD -X POST "http://$HOST:$PORT/api/profile/generate_auth_token/" -H "Content-Type: application/json" -c paperless-test-cookies.txt | tr -d '"')
 rm paperless-test-cookies.txt
 

backend/tests/daily/connectors/paperless_ngx/test_paperless_ngx_connector.py

@@ -124,6 +124,10 @@ def test_load_credentials(
 
         assert connector.api_url == "http://test.com"
         assert connector.auth_token == "test_token"
+        mock_get.assert_called_with(
+            f"http://test.com{PROFILE_ENDPOINT}",
+            headers={"Authorization": "Token test_token", "Accept": "application/json"},
+        )
 
 
 def test_load_from_state(