---
title: "Materialize v0.69"
date: 2023-09-20
released: true
patch: 6
---

## v0.69.0

#### Sources and sinks

[//]: # "NOTE(morsapaes) This feature was released in v0.59 behind a feature
flag. The flag was raised in v0.69 — so mentioning it here."

* Support validating the parameters provided in a `CREATE CONNECTION` statement
  against the target external system. For most connection types,
  Materialize **automatically validates** connections on creation.

  For connection types that require additional setup steps after creation
  (AWS PrivateLink, SSH tunnel), you can **manually validate** connections
  using the new [`VALIDATE CONNECTION`](https://materialize.com/docs/sql/validate-connection/)
  syntax:

   ```mzsql
   VALIDATE CONNECTION ssh_connection;
   ```

#### SQL

* Add support for new SQL functions:

  | Function                                                           | Description                                                 |
  | ------------------------------------------------------------------ | ----------------------------------------------------------- |
  | [`mz_is_superuser`](/sql/functions/#access-privilege-inquiry-functions) |  Reports whether the current role is a _superuser_ with administration privileges in Materialize. |
  | [`regexp_replace`](/sql/functions/#string-functions) | Replaces the first occurrence of the specified regular expression in a string with the specified replacement string.    |
  | [`regexp_split_to_array`](/sql/functions/#string-functions)      | Splits a string by the specified regular expression into an array. |
  | [`regexp_split_to_table`](/sql/functions/#table-functions)       | Splits a string by the specified regular expression.               |

<br>

* Add the `IN CLUSTER` option to the `SHOW { SOURCES | SINKS }` commands to
  restrict the objects listed to a specific cluster.

  ```mzsql
  SHOW SOURCES;
  ```
  ```nofmt
              name    | type     | size  | cluster
  --------------------+----------+-------+---------
   my_kafka_source    | kafka    |       | c1
   my_postgres_source | postgres |       | c2
  ```

  ```mzsql
  SHOW SOURCES IN CLUSTER c2;
  ```
  ```nofmt
  name       | type  | size     | cluster
  -----------+-------+----------+--------
  my_postgres_source | postgres | c2
  ```

* Make the syntax for [`GROUP SIZE` query hints](/transform-data/optimization/#query-hints)
  more intuitive by deprecating the `EXPECTED GROUP SIZE` hint and introducing
  three new hints: `AGGREGATE INPUT GROUP SIZE`, `DISTINCT ON INPUT GROUP SIZE`
  and `LIMIT INPUT GROUP SIZE`; which more clearly map to the target operation
  to optimize.

  The old `EXPECTED GROUP SIZE` hint is still supported for backwards
  compatibility, but its use is discouraged.

* Add `savings` to the [`mz_internal.mz_expected_group_size_advice`](/sql/system-catalog/mz_introspection/#mz_expected_group_size_advice)
  system catalog table. This column provides a conservative estimate of the memory
  savings that can be expected by using [`GROUP SIZE` query hints](/transform-data/optimization/#query-hints).

#### Bug fixes and other improvements

* Support SQL parameters in `SUBSCRIBE` and `DECLARE` statements.