# SQL examples

This page provides hands-on SQL examples for common tasks in Foundry, including reading and writing tabular data and querying ontology object types. Examples can be run from [SQL Studio](../sql-studio/index.md), the embedded [SQL console](../sql-console/index.md), or external SQL clients connected via [Arrow Flight SQL](../arrow-flight-sql/index.md). For full syntax reference, see the [SQL dialect](../sql-dialect/index.md) documentation.

## Tabular data

The following examples query and modify [datasets](../../data-integration/datasets/index.md) and [Iceberg tables](../../data-integration/iceberg-tables/index.md) in [data mode](../sql-studio/index.md#data-mode), executed by the [Furnace](../furnace/index.md) engine.

### Create and read a dataset

```sql
-- Create a dataset
CREATE OR REPLACE TABLE `/path/dataset-name` USING parquet AS
SELECT *
FROM `/path/dataset-input`;

-- Query a dataset
SELECT *
FROM `/path/dataset-name`
WHERE column = 'value';
```

### Create and modify an Iceberg table

:::callout{theme="warning"}
You must have [Iceberg tables](../../iceberg/overview/index.md) configured for your project to create Iceberg tables. See [Configuring Iceberg settings in Control Panel](../../iceberg/iceberg-settings/index.md) for details.
:::

```sql
-- Create an empty Iceberg table
CREATE TABLE `/path/table-name` (
    id   INT,
    name STRING
)
USING iceberg;

-- Insert rows
INSERT INTO `/path/table-name`
VALUES (1, 'apple'), (2, 'pear');

-- Update rows
UPDATE `/path/table-name`
SET name = 'clementine'
WHERE id = 2;

-- Delete rows
DELETE FROM `/path/table-name`
WHERE id = 1;
```

## Ontology object types

The following examples query ontology object types in [object mode](../sql-studio/index.md#object-mode), executed by the [Ontology SQL](../ontology-sql/index.md) engine. Object types are referenced by their resource identifier (RID).

### Query an object type

```sql
SELECT employeeId, firstName, department
FROM `ri.ontology.main.object-type.<employee-rid>`
WHERE department = 'Engineering';
```

### Query a many-to-many link

```sql
-- Find all cars driven by a specific person
SELECT c.*
FROM `ri.ontology.main.relation.<relation-rid>` AS linkTable
INNER JOIN `car` AS c
  ON c.`carId` = linkTable.`person_vehicles`
WHERE linkTable.`car_drivers` = 'person-123';
```

For details on link table column conventions and other object-querying patterns, see the [Ontology SQL](../ontology-sql/index.md) documentation.

## Troubleshooting

If you see an error such as `Output folder is not enabled for Iceberg tables. Please use a different format.`, you may be trying to write an Iceberg table to an unsupported location. This indicates that [Iceberg tables](../../iceberg/overview/index.md) are not enabled for your environment or for the specific project location you are writing to. Modify your code to write to a different data format (for example, `USING parquet`), or contact Palantir Support to enable Foundry Iceberg tables.