ClickHouse/docs/en/sql-reference/table-functions/input.md
2023-06-23 07:16:22 -06:00

45 lines
1.9 KiB
Markdown

---
slug: /en/sql-reference/table-functions/input
sidebar_position: 95
sidebar_label: input
---
# input
`input(structure)` - table function that allows effectively convert and insert data sent to the
server with given structure to the table with another structure.
`structure` - structure of data sent to the server in following format `'column1_name column1_type, column2_name column2_type, ...'`.
For example, `'id UInt32, name String'`.
This function can be used only in `INSERT SELECT` query and only once but otherwise behaves like ordinary table function
(for example, it can be used in subquery, etc.).
Data can be sent in any way like for ordinary `INSERT` query and passed in any available [format](../../interfaces/formats.md#formats)
that must be specified in the end of query (unlike ordinary `INSERT SELECT`).
The main feature of this function is that when server receives data from client it simultaneously converts it
according to the list of expressions in the `SELECT` clause and inserts into the target table. Temporary table
with all transferred data is not created.
**Examples**
- Let the `test` table has the following structure `(a String, b String)`
and data in `data.csv` has a different structure `(col1 String, col2 Date, col3 Int32)`. Query for insert
data from the `data.csv` into the `test` table with simultaneous conversion looks like this:
<!-- -->
``` bash
$ cat data.csv | clickhouse-client --query="INSERT INTO test SELECT lower(col1), col3 * col3 FROM input('col1 String, col2 Date, col3 Int32') FORMAT CSV";
```
- If `data.csv` contains data of the same structure `test_structure` as the table `test` then these two queries are equal:
<!-- -->
``` bash
$ cat data.csv | clickhouse-client --query="INSERT INTO test FORMAT CSV"
$ cat data.csv | clickhouse-client --query="INSERT INTO test SELECT * FROM input('test_structure') FORMAT CSV"
```