Amazon Trino ODBC Driver

Overview

The ODBC driver for the Amazon Trino serverless time series database provides an SQL-relational interface for developers and BI tool users.

Full source code is provided. Users should be able build the library for intended platform.

For build instructions please refer to developer guide.

For details on ODBC driver installation and usage please refer to rest of this document.

See CONTRIBUTING for more information.

This project is licensed under the Apache-2.0 License.

See the product documentation for more detailed information about this driver, such as setup and configuration.

To setup and use the Trino ODBC driver, follow these directions.

See Azure Active Directory set-up guide for setting up Azure AD account and connecting to Trino. See Okta set-up guide for setting up Okta account and connecting to Trino.

See PowerBI oveview for using PowerBI with Trino ODBC driver and connector.

See performance testing setup for setting up and running performance tests.

DRIVER={Amazon Trino ODBC Driver};DSN={Trino DSN};<option>=<value>;

For more information about connecting to an Amazon Trino database using this ODBC driver, see the connection string documentation for more details.

Platform	Amazon Trino ODBC Driver File
Windows	trino.odbc.dll
macOS	libtrino-odbc.dylib
Linux	libtrino-odbc.so

Trino does not support fully qualified table names. Trino ODBC driver reports databases as catalogs instead of schemas by default, as a result, tools like Tableau may not work as expected. For example, Tableau does not include database name in the auto-generated queries to fetch data if databases are reported as catalogs. Users are recommended to use Trino JDBC driver with Tableau. To make the Trino ODBC driver work with Tableau, the user can set environment variable DATABASE_AS_SCHEMA to TRUE to make the driver report databases as schemas instead. For details, see odbc-support-and-limitations.
The Trino ODBC driver is a unicode driver and will convert non-ANSI characters from data sources to ANSI; the environment variable ANSI_STRING_ONLY can be set to TRUE to indicate a data source only uses ANSI characters and skip conversion. For details, see unicode-support.