Rails Daily: Migration Compatibility Overhaul and Logging Fixes

A five-part stack refactors adapter-specific migration compatibility out of string checks and into per-adapter behavior, with fixes for SQLite and Postgres edge cases. A broadcast logger fix and its Active Job follow-up, plus STI and schema-cache corrections, round out the activity.

Duration: PT2M26S

Episode overview

This episode is a short developer briefing from Rails Daily.

It explains recent repository work in plain language.

  • Show: Rails Daily
  • Published: 2026-09-22T06:02:27Z
  • Audio duration: PT2M26S

Transcript excerpt

This excerpt keeps the crawler page concise. Listen to the episode or use the RSS feed for the full update.

Good morning, it's Tuesday, September 22nd, 2026, and this is Rails Daily.

The clear signal today is a five-part overhaul of migration compatibility, moving adapter-specific logic out of version checks and into proper per-adapter behavior.

That stack comes from one contributor and builds layer by layer. It starts with pull request 58844, which adds missing regression tests for things like MySQL table options and Postgres timestamp handling, so future regressions get caught. On top of that, pull request 58845 fixes SQLite references in version 6…

Second, logging fixes with a follow-on break. Pull request 58429, now merged, fixes the broadcast logger's tagged method when sending to more than one tagging logger, which previously raised a missing-method error. But because tagged is now always defined, Active Job's logging check could no longer detect formatter…

Finally, a set of sharp correctness and performance fixes. Pull request 58783 avoids building the full descendants list on every single-table inheritance lookup, a cost that scaled with rows times subclasses. Pull request 58832 restores integer limits from the JSON schema cache, fixing a frozen-object crash during…

What's next:…

Nearby episodes from Rails Daily

  1. Weekly Recap - Active Record Correctness and Ractor Readiness
  2. Ruby 4 Compatibility and MySQL Schema Loads
  3. Test Isolation Fixes and Parser Compatibility
  4. Query Correctness Gets Serious Attention
  5. Correctness Fixes Across the Data Layer
  6. Cleaning Up Old Assumptions
  7. Quiet Data Leaks and Silent Config Bugs
  8. Ractor Safety Takes Center Stage