GoKawiilDeveloper's pgrust project uncovers multiple real bugs in Postgres
Michael Malis, creator of the Rust-based Postgres rewrite project pgrust, has found and reported several genuine bugs in PostgreSQL while testing his project, including issues affecting SP-GiST operations, JSONPath results, pgcrypto, the dictionary loader, and money-type overflow handling. The author of this piece notes pgrust's main goal is improved performance, not just a rewrite, and vouches for Malis's credibility based on personal familiarity with his work.
GoKawiil's interpretation of the reporting above, not reported fact.
The author suggests that even if pgrust itself never becomes a production-ready replacement, the rigorous testing behind it could durably improve Postgres by surfacing bugs that get fixed upstream, and some of its performance optimizations might eventually be adopted by Postgres or other databases like Clickhouse. This illustrates a broader point the author is making: that skepticism toward AI-assisted or ambitious rewrite projects is often justified given how many are low-quality, but can cause genuinely rigorous efforts to be unfairly dismissed without scrutiny.
- pgrust, a Rust-based Postgres project by Michael Malis, has surfaced several real Postgres bugs during testing
- The bugs found span SP-GiST, JSONPath, pgcrypto, the dictionary loader, and money-type overflow handling
- The author argues thorough testing in ambitious side projects can benefit the original software even if the project itself doesn't succeed
Source: danluu.com, 2026-10-24
Published there as: “We should be more ambitious with the software we build”
Read the original report → The summary and analysis above are GoKawiil's own, written from reporting by the source above. Facts and quotes belong to the original publisher.