[Yandex Cloud documentation](../../index.md) > [Yandex DataLens](../index.md) > Calculated fields > Function reference > Native functions > DB_COLUMN

# DB_COLUMN



#### Syntax {#syntax}


```
DB_COLUMN( column_name )
```

#### Description {#description}

{% note warning %}

The availability and behavior of native functions depends on your database type and version.

It is recommended to use native functions only if there is no supported DataLens function (composition of functions) that can achieve the required result.

{% endnote %}

Returns a native database column by its name. Supported column names and their result types are defined by the connector.

The argument `column_name` must be a constant string with one of the values supported by the connector. For ClickHouse, all results have type `String` and the supported values are:
`"_table"`, `"_partition_id"`, `"_file"`, `"_path"`.

Unlike DB_CALL_* formulas, this formula emits a column reference, not a function call. On ClickHouse, virtual-column availability depends on the table engine and server version: `_partition_id` belongs to the MergeTree family, `_table` to the Merge engine/table function, and `_file` / `_path` to File/S3.

**Argument types:**
- `column_name` — `String`


**Return type**: `String`

{% note info %}

Only constant values are accepted for the arguments (`column_name`).

{% endnote %}


#### Examples {#examples}

```
DB_COLUMN("_partition_id") -- ClickHouse MergeTree: virtual partition id
```

```
DB_COLUMN("_table") -- ClickHouse Merge: source table name
```

```
DB_COLUMN("_file") -- ClickHouse File/S3: source file name
```

```
DB_COLUMN("_path") -- ClickHouse File/S3: source file path
```


#### Data source support {#data-source-support}

`ClickHouse 21.8`.