Skip to content

v2.3.0

Latest
Compare
Choose a tag to compare
@risingwave-ci risingwave-ci released this 14 Apr 05:42
· 342 commits to main since this release
e8f053e

Release v2.3.0

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

Cluster configuration changes

  • Adds runtime parameter sink_rate_limit . #19660

Assets