dbt-labs/dbt-core · error

{}

Error message

{}

What it means

Column.get_name() (a Databricks-specific ColumnStatic Jinja method) expects a `column` keyword argument that can be deserialized into a DbtColumn struct. When the Jinja Value passed as `column` is not a dict-like object with the fields DbtColumn requires (at minimum `name`, optionally `quote`), minijinja's typed-struct deserialization fails and the library re-wraps it as a SerdeDeserializeError with message "{}". It signals that the wrong kind of value was passed to Column.get_name(), not a database or SQL problem.

Source

Thrown at crates/dbt-adapter/src/column/types.rs:103

                let columns = args.get::<Value>("columns")?;
                let columns = Column::vec_from_jinja_value(AdapterType::Databricks, columns)?;

                Ok(Value::from(self.dbx_format_add_column_list(&columns)?))
            }
            "format_remove_column_list" => {
                // TODO: ArgsIter
                let mut args = ArgParser::new(args, None);
                let columns = args.get::<Value>("columns")?;
                let columns = Column::vec_from_jinja_value(AdapterType::Databricks, columns)?;

                Ok(Value::from(self.dbx_format_remove_column_list(&columns)?))
            }
            "get_name" => {
                let mut args: ArgParser = ArgParser::new(args, None);
                let column = args.get::<Value>("column")?;
                // FIXME: why is this DbtColumn and not Column?
                let column = minijinja_value_to_typed_struct::<DbtColumn>(column).map_err(|e| {
                    minijinja::Error::new(
                        minijinja::ErrorKind::SerdeDeserializeError,
                        e.to_string(),
                    )
                })?;

                Ok(Value::from(self.dbx_get_name(&column)))
            }
            _ => Err(minijinja::Error::new(
                minijinja::ErrorKind::UnknownMethod,
                format!("Unknown method on ColumnStatic: '{name}'"),
            )),
        }
    }

    fn call(
        self: &Arc<Self>,
        _state: &minijinja::State,
        args: &[Value],

View on GitHub (pinned to 0267ce9170)

Solutions

1. Pass a dict with at least the `name` key (and `quote` if desired): Column.get_name(column={'name': col_name, 'quote': True}).
2. If you have a Column object from from_description/create, use its `name`/`quoted` attributes directly instead of get_name().
3. Inspect the wrapped SerdeDeserializeError message (the `{}` payload) to see which field failed to deserialize and fix the dict keys/types.
4. Ensure the call goes through the `column=` keyword; positional args are not read by ArgParser here.

Example fix

// before (Jinja)
{% set col = adapter.Column.from_description('id', 'INT') %}
{% set n = Column.get_name(column=col) %}
// after
{% set n = Column.get_name(column={'name': 'id', 'quote': false}) %}

// Jinja: verify the value is a dict with a name before calling
{% if column is mapping and column.name is defined %}
  {% set n = Column.get_name(column=column) %}
{% else %}
  {{ exceptions.raise_compiler_error("get_name requires a column dict with a 'name' field") }}
{% endif %}

fn is_column_dict(v: &minijinja::Value) -> bool {
    v.as_object().map(|o| o.get_attr("name").map(|n| !n.is_undefined()).unwrap_or(false)).unwrap_or(false)
}

When it happens

Trigger: Calling `Column.get_name(column=...)` from a Jinja materialization/macro with: (1) no `column` keyword arg at all, (2) a non-dict value (string, number, list, Column object instead of a node-column dict), or (3) a dict missing required DbtColumn fields such as `name`.

Common situations: Adapter materialization macros (e.g. Databricks) passing an API Column object instead of the raw node column dict; a renamed/missing field in a custom macro that builds the column dict by hand; passing results of `columns` iteration where items are already Column objects; version drift where DbtColumn gained a required field.

Understand the failure class

Background: Schema validation failed / invalid input schema: payload rejected because its shape doesn't match the expected schema — this error's family across 28 libraries.

Related errors

• Failed to downcast jinja value to Column; expected Column ob
• existing_columns must contain Column objects
• Unknown method on ColumnStatic: '{name}'
• {msg}
• render_constraints_for_create is only available for Databric

AI-assisted analysis of dbt-labs/dbt-core@0267ce9170 (2026-09-07). Data as JSON: /api/errors/3e96194134e0c827. Report an issue: GitHub.