Struct datafusion_iceberg_sql::context::IcebergContext

pub struct IcebergContext { /* private fields */ }

Implementations

impl IcebergContext

pub async fn new( tables: &[(String, String, String)], catalogs: Arc<dyn CatalogList>, branch: Option<&str> ) -> Result<IcebergContext, DataFusionError>

Trait Implementations

impl ContextProvider for IcebergContext

fn get_table_source( &self, name: TableReference<'_> ) -> Result<Arc<dyn TableSource>, DataFusionError>

Getter for a datasource

fn get_function_meta(&self, _name: &str) -> Option<Arc<ScalarUDF>>

Getter for a UDF description

fn get_variable_type(&self, _variable_names: &[String]) -> Option<DataType>

Getter for system/user-defined variable type

fn get_aggregate_meta(&self, _name: &str) -> Option<Arc<AggregateUDF>>

Getter for a UDAF description

fn get_window_meta(&self, _name: &str) -> Option<Arc<WindowUDF>>

Getter for a UDWF

fn options(&self) -> &ConfigOptions

Get configuration options

fn get_table_provider( &self, name: TableReference<'_> ) -> Result<Arc<dyn TableSource>, DataFusionError>

👎Deprecated since 32.0.0: please use get_table_source instead

fn get_table_function_source( &self, _name: &str, _args: Vec<Expr> ) -> Result<Arc<dyn TableSource>, DataFusionError>

Getter for a table function

fn create_cte_work_table( &self, _name: &str, _schema: Arc<Schema> ) -> Result<Arc<dyn TableSource>, DataFusionError>

This provides a worktable (an intermediate table that is used to store the results of a CTE during execution) We don’t directly implement this in the logical plan’s [’SqlToRel] because the sql code needs access to a table that contains execution-related types that can't be a direct dependency of the sql crate (namely, the CteWorktable). The [ContextProvider`] provides a way to “hide” this dependency.