# DuckDB

Supported since 5.0.2.

Through the DuckDB (opens new window) connector, Ontop is able to construct VKGs on DuckDB database files.

# Limitations & Exceptions

  • DuckDB database files prior to version 1.0 are not guaranteed to be compatible with later versions.
  • Setting a default schema is not supported when connecting to DuckDB.
  • Nested data types are only supported starting from version 0.7 of DuckDB.
  • Accessing struct fields using the "dot operator" is not supported.
  • The Ontop option ontop.exposeSystemTables is not supported for DuckDB, as the JBDC does not allow us to access system tables.
  • DuckDB has a very specific concurrency model (opens new window). In particular, it does not support concurrent multi-process read-write queries.
  • DuckDB supports integrity constraints, but they tend to costly for large tables (opens new window). Consider instead specifying constraints using lenses for large datasets.

# Database Connection

The following shows the content of a sample .properties file that can be used to connect Ontop to DuckDB:

jdbc.url = jdbc:duckdb:path/to/database.db
jdbc.driver = org.duckdb.DuckDBDriver

# Nested Type Support

Ontop implements explicit compatibility with the DuckDB array type T[]. When used with the flatten lens, it is able to automatically infer the type of the result column.

The flatten lens cannot be used on arrays stored as JSON-encoded columns in the database. For such use cases, the column first has to be converted to an array type.

NOTE

Nested data types are only supported starting from version 0.7 of DuckDB.

# Struct Access

Individual struct objects can be accessed by SQL expressions in the dialect's default way:

SELECT my_struct['my_attribute'] FROM ...

WARNING

Accessing struct fields using the "dot operator" is not supported.

Last Updated: 11/26/2024, 1:08:23 PM