fix: Fix error handling test so that it tests correct behaviour (#939)

The test appeared to be checking the error status under various error
conditions but those conditions were never being reached due to other
values causing errors before they were raised.

If the intended error was raised, the exit status was also not the one
expected.

Author: tpoliaw

tests/unit_tests/test_cli.py

@@ -406,14 +406,22 @@ def test_env_reload_server_side_error(runner: CliRunner):
 
 
 @pytest.mark.parametrize(
-    "exception, expected_exit_code",
+    "exception, error_message",
     [
-        (ValidationError.from_exception_data(title="Base model", line_errors=[]), 1),
-        (BlueskyRemoteControlError("Server error"), 1),
-        (ValueError("Error parsing parameters"), 1),
+        (
+            ValidationError.from_exception_data(title="Base model", line_errors=[]),
+            "('failed to validate the task parameters, ,"
+            + " error: 0 validation errors for '\n 'Base model\\n')\n",
+        ),
+        (
+            BlueskyRemoteControlError("Server error"),
+            "'server error with this message: Server error'\n",
+        ),
+        (ValueError("Error parsing parameters"), "'task could not run'\n"),
     ],
+    ids=["validation_error", "remote_control", "value_error"],
 )
-def test_error_handling(exception, expected_exit_code, runner: CliRunner):
+def test_error_handling(exception, error_message, runner: CliRunner):
     # Patching the create_task method to raise different exceptions
     with patch(
         "blueapi.client.rest.BlueapiRestClient.create_task", side_effect=exception
@@ -422,15 +430,15 @@ def test_error_handling(exception, expected_exit_code, runner: CliRunner):
             main,
             [
                 "-c",
-                "tests/example_yaml/valid_stomp_config.yaml",
+                "tests/unit_tests/example_yaml/valid_stomp_config.yaml",
                 "controller",
                 "run",
                 "sleep",
-                "'{\"time\": 5}'",
+                '{"time": 5}',
             ],
-            input="\n",
         )
-        assert result.exit_code == expected_exit_code
+        # error message is printed to stderr but test runner combines output
+        assert result.stdout == error_message
 
 
 def test_device_output_formatting():