Skip to content

Commit a8867e8

Browse files
ebyhrebyhr
committed
Add Trino 477 release notes
1 parent 424e8b5 commit a8867e8

File tree

2 files changed

+203
-0
lines changed

2 files changed

+203
-0
lines changed

‎docs/src/main/sphinx/release.md‎

Lines changed: 1 addition & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -6,6 +6,7 @@
66
```{toctree}
77
:maxdepth: 1
88
9+
release/release-477
910
release/release-476
1011
release/release-475
1112
release/release-474

‎docs/src/main/sphinx/release/release-477.md‎

Lines changed: 202 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,202 @@
1+
# Release 477 (dd MMM 2025)
2+
3+
## General
4+
5+
* Improve reliability of spilling for aggregation by retaining revocable memory, preventing out‑of‑memory errors. ({issue}`25892`)
6+
* Fix correctness issues when unspilling hash aggregation by using correct hash values, avoiding aggregation miscalculations. ({issue}`25892`)
7+
* Reduce memory consumed by grouped Top N rank and row number operations ({issue}`25946`)
8+
* Remove the HTTP server event listener plugin from the tar.gz archive and the Docker container. ({issue}`25967`)
9+
* Fix failure when aggregation exists in other expressions in `GROUP BY AUTO`. ({issue}`25987`)
10+
* Improve spilling reliability for join queries. ({issue}`25976`)
11+
* Add support for `ALTER MATERIALIZED VIEW ... SET AUTHORIZATION`. ({issue}`25910`)
12+
* Fix incorrect results with spatial joins. ({issue}`26021`)
13+
* Reduce memory usage of aggregations by incrementally releasing memory as output rows are produced ({issue}`25879`)
14+
* Add the `coordinatorId` to the `/v1/info` endpoint. ({issue}`23910`)
15+
* Use UUIDv7 for `runId` in OpenLineage event listener. ({issue}`25534`)
16+
* Partition large pages to avoid OOM while serializing data in the spooling protocol. ({issue}`25999`)
17+
* All catalogs are now required to be deployed to all nodes. ({issue}`26063`)
18+
* Add Lakehouse connector. ({issue}`25347`)
19+
* Add user identifying fields to OpenLineage `trino_query_context` facet in OpenLineage event listener. ({issue}`26074`)
20+
* Add `query_id` field to `trino_metadata` facet in OpenLineage event listener. ({issue}`26074`)
21+
* Add `query.max-write-physical-size` configuration property and `query_max_write_physical_size` session property to allow configuring limits on the amount of data written by a query. ({issue}`25955`)
22+
* Add `system.metadata.tables_authorization`, `system.metadata.schemas_authorization`, `system.metadata.functions_authorization` tables that expose the information about the authorization for given entities. ({issue}`25907`)
23+
* Add Announce node inventory to replace Airlift Discovery. ({issue}`26119`)
24+
* Add DNS node inventory for use in K8s like environments that provide a DNS name for all workers. ({issue}`26119`)
25+
* Add `discovery.type` to set node inventory system. The value can be `announce`, `dns` or `airlift-discovery`. ({issue}`26119`)
26+
* Add support for creating, dropping, fast-forwarding and listing branches. ({issue}`25751`)
27+
* Fix spill files leaking in aggregation queries. ({issue}`26141`)
28+
* Reduce number of worker OOMs when running join queries. ({issue}`26142`)
29+
* Add support for default column values. ({issue}`25679`)
30+
* Add support for configuring max request size with the `kafka-event-listener.max-request-size` config property in Kafka Event Listener. ({issue}`26129`)
31+
* Add support for configuring batch size with the `kafka-event-listener.batch-size` config property in Kafka Event Listener. ({issue}`26129`)
32+
* Add support for branching with `INSERT`, `DELETE`, `UPDATE`, and `MERGE` statements. ({issue}`26136`)
33+
* Add `debug_adaptive_planner` session property which allows gathering extra diagnostics information regarding adaptive planner operation. ({issue}`26274`)
34+
* Reduce memory required for distinct and ordered grouped aggregations. ({issue}`26276`)
35+
* Fix memory tracking for ordered grouped aggregations. ({issue}`26276`)
36+
* Add support for view refresh operation. ({issue}`25906`)
37+
* Add support for creating a branch from specified branches. ({issue}`26300`)
38+
* Add support for configuring the HTTP method with `http-event-listener.connect-http-method` config property. ({issue}`26181`)
39+
* Add physical data scan tracking to resource groups. ({issue}`25003`)
40+
* Fix access control check when access is granted through groups in `SET SESSION AUTHORIZATION`. ({issue}`26344`)
41+
42+
## Security
43+
44+
## Web UI
45+
46+
* Improve UI rendering when Trino cluster is air-gapped ({issue}`26031`)
47+
* Add query details page to the Preview Web UI. ({issue}`25554`)
48+
49+
## JDBC driver
50+
51+
* Send detailed client information in the source ({issue}`25889`)
52+
53+
## Docker image
54+
55+
## CLI
56+
57+
* Send detailed client information in the source. ({issue}`25889`)
58+
* Allow configuring `--max-buffered-rows` and `--max-queued-rows`. ({issue}`26015`)
59+
* Use actual stage id instead incremental number in progress output. ({issue}`26139`)
60+
* Use {kbd}`Alt+↑` or {kbd}`Alt+↓` to move through the history. ({issue}`26138`)
61+
62+
## BigQuery connector
63+
64+
## Blackhole connector
65+
66+
## Cassandra connector
67+
68+
## ClickHouse connector
69+
70+
## Delta Lake connector
71+
72+
* Prevent workers from going into full GC or crashing when decoding unusually large parquet footers. ({issue}`25973`)
73+
* Add support for using GCS without credentials. ({issue}`25810`)
74+
* Fix failure when reading tables with `null` on `variant` type. ({issue}`26016`)
75+
* Fix incorrect results when reading from parquet files produced by old versions of pyarrow. ({issue}`26058`)
76+
* Release native filesystem resources/prevent leaks. ({issue}`26085`)
77+
* Add ability to detect resource leakage in the runtime. ({issue}`26087`)
78+
* Fix delta lake connector not closing resource streams properly. ({issue}`26092`)
79+
* Prevent creating multiple Alluxio client which can cause excessive resource usage. ({issue}`26121`)
80+
* Fix writing malformed checkpoint files when deletion vector is enabled. ({issue}`26145`)
81+
* Fix failure when reading `null` values on `json` type columns. ({issue}`26184`)
82+
* Fix skipping of row groups when the trino type is different from logical types in case of parquet files. ({issue}`26203`)
83+
* Add `azure.multipart-write-enabled` that enables multipart uploads for large files. ({issue}`26225`)
84+
* Rename `s3.socket-read-timeout` config property to `s3.socket-timeout`. ({issue}`26263`)
85+
* Add support for `GRANT`, `DENY`, `REVOKE` with branches. ({issue}`25152`)
86+
* Fix incorrect results when the table uses deletion vectors and special characters are used in the partition path. ({issue}`26299`)
87+
* Reduce query failures from S3 throttling. ({issue}`26407`)
88+
89+
## Druid connector
90+
91+
## DuckDB connector
92+
93+
## Elasticsearch connector
94+
95+
## Exasol connector
96+
97+
## Faker connector
98+
99+
## Google Sheets connector
100+
101+
## Hive connector
102+
103+
* Prevent workers from going into full GC or crashing when decoding unusually large parquet footers. ({issue}`25973`)
104+
* Add support for using GCS without credentials. ({issue}`25810`)
105+
* Fix incorrect results when reading from parquet files produced by old versions of pyarrow. ({issue}`26058`)
106+
* Add support for reading tables using the ESRI JSON format ({issue}`25241`)
107+
* Add ability to detect resource leakage in the runtime. ({issue}`26087`)
108+
* Prevent creating multiple Alluxio client which can cause excessive resource usage. ({issue}`26121`)
109+
* Fix reading specifying format for date partition projection. ({issue}`25642`)
110+
* Fix skipping of row groups when the trino type is different from logical types in case of parquet files. ({issue}`26203`)
111+
* Add `azure.multipart-write-enabled` that enables multipart uploads for large files. ({issue}`26225`)
112+
* Rename `s3.socket-read-timeout` config property to `s3.socket-timeout`. ({issue}`26263`)
113+
* Reduce query failures from S3 throttling. ({issue}`26407`)
114+
* Add support for `extended_boolean_literal` in text-file formats. ({issue}`21156`)
115+
116+
## Hudi connector
117+
118+
* Prevent workers from going into full GC or crashing when decoding unusually large parquet footers. ({issue}`25973`)
119+
* Add support for `parquet_max_read_block_row_count` session property. ({issue}`25981`)
120+
* Add support for using GCS without credentials. ({issue}`25810`)
121+
* Fix incorrect results when reading from parquet files produced by old versions of pyarrow. ({issue}`26058`)
122+
* Add ability to detect resource leakage in the runtime. ({issue}`26087`)
123+
* Prevent creating multiple Alluxio client which can cause excessive resource usage. ({issue}`26121`)
124+
* Fix skipping of row groups when the trino type is different from logical types in case of parquet files. ({issue}`26203`)
125+
* Add `azure.multipart-write-enabled` that enables multipart uploads for large files. ({issue}`26225`)
126+
* Rename `s3.socket-read-timeout` config property to `s3.socket-timeout`. ({issue}`26263`)
127+
* Reduce query failures from S3 throttling. ({issue}`26407`)
128+
129+
## Iceberg connector
130+
131+
* Prevent workers from going into full GC or crashing when decoding unusually large parquet footers. ({issue}`25973`)
132+
* Add support for using GCS without credentials. ({issue}`25810`)
133+
* Fix latency regression and potential query failures for `REFRESH MATERIALIZED VIEW` command from 475 release. ({issue}`26051`)
134+
* Fix incorrect results when reading from parquet files produced by old versions of pyarrow. ({issue}`26058`)
135+
* Add ability to detect resource leakage in the runtime. ({issue}`26087`)
136+
* Prevent creating multiple Alluxio client which can cause excessive resource usage. ({issue}`26121`)
137+
* Remove `iceberg.rest-catalog.sigv4-enabled` config property and add `SIGV4` to `iceberg.rest-catalog.security`. ({issue}`26218`)
138+
* Fix failure when executing `optimize_manifests` procedure on tables having `NULL` on the top level partition. ({issue}`26185`)
139+
* Fix skipping of row groups when the trino type is different from logical types in case of parquet files. ({issue}`26203`)
140+
* Add `azure.multipart-write-enabled` that enables multipart uploads for large files. ({issue}`26225`)
141+
* Rename `s3.socket-read-timeout` config property to `s3.socket-timeout`. ({issue}`26263`)
142+
* Improve performance of expire_snapshots procedure. ({issue}`26230`)
143+
* Reduce memory usage of remove_orphan_files procedure. ({issue}`25847`)
144+
* Improve performance of remove_orphan_files procedure. ({issue}`26326`)
145+
* Improve performance of queries on `$files` metadata table. ({issue}`25677`)
146+
* Add `compression_codec` table property and remove `compression_codec` session property. ({issue}`25755`)
147+
* Reduce query failures from S3 throttling. ({issue}`26407`)
148+
149+
## Ignite connector
150+
151+
## JMX connector
152+
153+
## Kafka connector
154+
155+
## Loki connector
156+
157+
## MariaDB connector
158+
159+
## Memory connector
160+
161+
* Add support for default column values. ({issue}`25679`)
162+
* Add support for view refresh operation. ({issue}`25906`)
163+
164+
## MongoDB connector
165+
166+
## MySQL connector
167+
168+
## OpenSearch connector
169+
170+
## Oracle connector
171+
172+
## Pinot connector
173+
174+
## PostgreSQL connector
175+
176+
* Add support for `geometry` types installed in schemas other than `public`. ({issue}`25972`)
177+
178+
## Prometheus connector
179+
180+
## Redis connector
181+
182+
## Redshift connector
183+
184+
## SingleStore connector
185+
186+
## Snowflake connector
187+
188+
## SQL Server connector
189+
190+
## TPC-H connector
191+
192+
## TPC-DS connector
193+
194+
## Vertica connector
195+
196+
## SPI
197+
198+
* Remove `ConnectorSession` from `Type.getObjectValue`. ({issue}`25945`)
199+
* Remove unused `NodeManager` `getEnvironment` method. ({issue}`26096`)
200+
* Deprecate `NodeManager` `getCurrentNode` in favor of `ConnectorContext` `getCurrentNode`. ({issue}`26096`)
201+
* Remove `@Experimental` annotation. ({issue}`26200`)
202+
* Remove deprecated `ConnectorPageSource.getNextPage` method. ({issue}`26222`)

0 commit comments

Comments
 (0)