Principle:Eventual Inc Daft Table Creation
| Knowledge Sources | |
|---|---|
| Domains | Data_Engineering, Catalog_Management |
| Last Updated | 2026-02-08 00:00 GMT |
Overview
Technique for creating persistent named tables in a catalog.
Description
Table creation registers a new table in the current catalog from a DataFrame or schema, with optional properties. The table persists beyond the session and is accessible by other sessions. If no namespace is specified in the identifier, the session's current namespace is used for resolution. A current catalog must be set for table creation to succeed.
Usage
Use table creation when you need to create a persistent table in a catalog for long-term storage and shared access. This is appropriate for ETL pipelines that produce output tables, data warehouse table management, and creating tables that other users or processes need to access.
Theoretical Basis
DDL operation creating table metadata in a catalog with schema definition and optional properties:
create_table(identifier, source, **properties):
catalog = session.current_catalog()
if catalog is None:
raise Error("No current catalog")
# Resolve namespace for unqualified names
if identifier has no namespace:
identifier = current_namespace + identifier
# Delegate to catalog implementation
catalog.create_table(identifier, source, properties)
The underlying catalog implementation determines how the table is physically stored (e.g., Iceberg, Delta Lake, in-memory).