Releases · risingwavelabs/risingwave

Support for RisingWave v2.1 will end upon the release of v2.3. If you're using v2.1, please plan your upgrade to a newer version (v2.2 or later) soon to ensure continued support and updates. For more details, see our version support policy.

Main changes

SQL features

Query syntax:
- Supports cross-database queries in both streaming and ad-hoc modes. #20261
- Support grouping by window_start and window_end at the same time in emit-on-window-close queries. #19998
- Support ASOF JOIN for batch queries to match the nearest record with possible variations in case of ties. #19790
SQL commands:
- Public preview: Supports specifying resource_group for CREATE DATABASE command. #20751
- Allows enabling or disabling the creation of embedded Python, JavaScript, and WebAssembly UDFs. #20893
- Supports async and batched functions for embedded JavaScript UDFs. #20403
- Rejects schema-qualified CURSOR names in DECLARE statements to align with Postgres specifications. #20221
- CREATE TABLE command supports partially specifying the schema instead of using * to ingest all columns. #20203
- DROP SCHEMA command supports the CASCADE option to drop the specified schema and all its dependencies. #19702
- Supports switching databases during an active session using either SET DATABASE TO or USE. #19786
- Supports SET PARALLELISM clause for ALTER SOURCE command. #20606
- Support IF NOT EXISTS option for CREATE FUNCTION/CREATE AGGREGATE. #20079
- Supports GRANT or REVOKE privileges on view. #20670
SQL functions & operators:
- Support IGNORE NULLS for first_value and last_value window functions. #19847
System catalog:
- Public preview: Adds system catalog rw_resource_groups. #20920
- Public preview: Adds system catalog rw_streaming_jobs. #20207

Connectors

Breaking change: The legacy S3 source will be fully deprecated in v2.4.0. The properties and enumerator will be retained to ensure that the metadata can start normally. #20658
Public preview: Support ingesting data from PostgreSQL 17 using CDC. #20668
Supports creating connections for Iceberg. #20189
Supports AWS S3 Tables integration for the Iceberg source, sink, and table engine. #21018
Supports creating Iceberg sources and sinks with GCS. #19999
Supports generated columns for tables on CDC sources. #20380
Supports reading Parquet files from GCS and Azure blob. #19974, #20046
Removes schema from DML statements for JDBC sinks if not specified one. #20632
Adds partition_by to enable partitioning using column names or transformations for Iceberg sinks. #20495
Adds jdbc.auto.commit for JDBC sinks to automatically commit transactions. #20597
Adds snowflake catalog for Iceberg source. #20421
Adds aws.msk.signer_timeout_sec to set timeout limit for loading AWS credentials of AWS MSK. #20610
jdbc.query.timeout for JDBC sinks is reduced from 600s to 60s. #20641

Installation and deployment

For the production cluster, a license key with Paid tier and CPU limit of 4 is set by default to allow the free trial of paid features on a small scale. #20296
Add a function rw_license() to retrieve the license information. #20629