Materialize v0.79
v0.79.0
Sources and sinks
-
For PostgreSQL sources, prevent the creation of new sources when the upstream database does not have a sufficient number of replication slots available.
-
Fix a bug where subsources where created in the active schema, rather than the schema of the source, when using the
FOR SCHEMAS
option in PostgreSQL sources. -
Add
mz_aws_privatelink_connection_status_history
to the system catalog. This table contains a row describing the historical status for each AWS PrivateLink connection in the system.
SQL
-
Add
mz_compute_hydration_status
to the system catalog. This table describes the per-replica hydration status of indexes, materialized views, or subscriptions, which is useful to track when objects are “caught up” in the context of blue/green deployments. -
Add
create_sql
to object-specific tables inmz_catalog
(e.g.mz_sources
). This column provides the DDL used to create the object. -
Allow calling functions from the
mz_internal
schema, which are considered safe but unstable (e.g.is_rbac_enabled
).
Bug fixes and other improvements
-
Improve type coercion in
WITH MUTUALLY RECURSIVE
common table expressions. For example, you can now returnNUMERIC
values of arbitrary scales (e.g.NUMERIC (38,2)
for columns defined asNUMERIC
. -
Automatically enable compaction when creating the progress topic for a Kafka sink. Warning: Versions of Redpanda before v22.3 do not support using compaction for Materialize’s progress topics. You need to manually create the connection’s progress topic with compaction disabled to use sinks with these versions of Redpanda.