Add several commits to .git-blame-ignore-revs.

Author: Nathan Bossart

Adds several preceding commits to the .git-blame-ignore-revs file, helping to clarify git blame output by ignoring commits related to mechanical changes.


Fix style in a few REPACK ereports

Author: Álvaro Herrera

This commit standardizes the naming convention for "REPACK (CONCURRENTLY)" in error hint messages and removes direct DDL mentions from translatable strings. It also improves error reporting after XLogReadRecord failures.


Use "grep -E" not "egrep".

Author: Tom Lane

Replaces the use of egrep with grep -E for POSIX compliance, as recent versions of GNU grep have started warning about egrep's deprecation. This change affects scripts and build processes rather than normal runtime code.


Pre-beta updates: run src/tools/copyright.pl.

Author: Tom Lane

Performs copyright year updates across the codebase, addressing instances where post-January 1st patches missed these updates. This is part of routine pre-beta maintenance.


Do pre-release housekeeping on catalog data.

Author: Tom Lane

Executes renumber_oids.pl to move high-numbered OIDs down, fulfilling pre-beta tasks outlined in RELEASE_CHANGES to consolidate catalog data.


Add preceding commits to .git-blame-ignore-revs.

Author: Tom Lane

Further updates the .git-blame-ignore-revs file by adding more commits to be excluded from git blame output, similar to a previous commit.


Pre-beta mechanical code beautification, step 3: run reformat-dat-files.

Author: Tom Lane

The third step in the pre-beta code beautification process involves running reformat-dat-files to ensure consistent formatting across data files.


Pre-beta mechanical code beautification, step 2: run pgperltidy.

Author: Tom Lane

This commit applies pgperltidy as the second step in mechanical code beautification, ensuring Perl scripts adhere to established style guidelines.


Pre-beta mechanical code beautification, step 1: run pgindent.

Author: Tom Lane

The initial phase of pre-beta code beautification involves updating typedefs.list and running pgindent, resulting in a significant diff due to recent pgindent behavior changes.


pgindent: improve formatting of multiline comments.

Author: Tom Lane

Enhances pgindent to enforce a standard formatting for multiline comments starting in column 1, ensuring consistent alignment and structure.


Make pg_bsd_indent add a space between comma and period.

Author: Tom LaneDiscussion: https://postgr.es/m/

This commit modifies pg_bsd_indent to correctly insert a space between a comma and a period, fixing an issue where variadic functions and struct literals with named fields were poorly formatted. The pg_bsd_indent version is bumped to signal the update.


pgindent: Fix spacing after != when member name matches typedef.

Author: Nathan Bossart

Corrects pgindent's behavior to prevent it from removing the space after != (and similar operators) when a struct member name matches a registered typedef, ensuring proper spacing in expressions.


Fix FOR PORTION OF with non-updatable view columns

Author: Peter EisentrautDiscussion: CACJufxFRqg8=gbZ-Q6ZS_UQ+...

Resolves a bug where UPDATE and DELETE operations failed to verify that the application-time column was updatable when used with FOR PORTION OF on non-updatable view columns. This ensures proper permission checks for such columns.


pg_stat_statements: Set PlannedStmt to NULL after nested utility execution

Author: Michael PaquierDiscussion: 0A9A8DAC-BC3C-4C7A-9504-2C6050405544@anarazel.de

This commit sets the PlannedStmt to NULL after nested utility execution in pg_stat_statements, particularly after an internal ROLLBACK. This makes it easier to detect bugs related to relying on a freed PlannedStmt.


Add more tests for corrupted data with pglz_decompress()

Author: Michael PaquierDiscussion: agF4xkIdRcrCIprs@paquier.xyz

Expands test coverage for corrupted data handling within pglz_decompress(), specifically addressing two previously uncovered cases related to control bits and 2-byte match tags with invalid offsets.


Fix stale COPY progress during logical replication table sync

Author: Fujii Masao

Corrects an issue where pg_stat_progress_copy could display stale progress for logical replication table synchronization, remaining active even after the data copy completed. The fix involves adding a missing EndCopyFrom() call.


HN News

Launch HN: Ardent (YC P26) – Postgres sandboxes in seconds with zero migration

Score: 51 / Comments: 20

Ardent, a YC W26 startup, introduces a new service enabling developers to create isolated PostgreSQL sandboxes quickly and without needing data migration. This aims to streamline development and testing workflows by providing disposable, full-featured database environments.


Show HN: Torrix, self hosted, LLM Observability,(no Postgres, no Redis)

Score: 25 / Comments: 1

Torrix presents a self-hosted LLM observability solution that distinguishes itself by not requiring PostgreSQL or Redis, offering an alternative for developers seeking lightweight monitoring for their AI agent applications.


Show HN: Diom – Open-source back end primitives with no runtime dependencies

Score: 4 / Comments: 0

Diom introduces open-source backend primitives designed for minimal overhead, operating without any runtime dependencies. This could be relevant for projects looking to build efficient and lean backend services.


We built our own message queue for AI agents (and put it on Postgres)

Score: 2 / Comments: 0

This article details the reasons and process behind developing a custom message queue specifically for AI agents, leveraging PostgreSQL as its underlying storage mechanism. It explores the benefits of using a relational database for message queuing in this context.


Summary

Today's PostgreSQL news includes a substantial round of pre-beta housekeeping commits, focusing on code quality, formatting, and catalog data management, alongside several critical bug fixes across various components. On Hacker News, the standout is the launch of Ardent, offering instant PostgreSQL sandboxes, while other discussions explore alternative observability tools and custom message queues built on PostgreSQL.


This digest was generated by gemini based on Hacker News, GitHub, and the PostgreSQL mailing list.Today's PostgreSQL news includes a substantial round of pre-beta housekeeping commits, focusing on code quality, formatting, and catalog data management, alongside several critical bug fixes across various components. On Hacker News, the standout is the launch of Ardent, offering instant PostgreSQL sandboxes, while other discussions explore alternative observability tools and custom message queues built on PostgreSQL.

This digest was generated by gemini based on Hacker News, GitHub, and the PostgreSQL mailing list.