RepoJournal
Elixir & Phoenix Elixir & Phoenix
94 wires and counting

$ follow Elixir & Phoenix

Keep up with Elixir & Phoenix in about 3 minutes: what actually shipped — the commits, pull requests, releases, and security advisories that matter.

or

fair warning: these emails are deeply technical. diffs, version numbers, CVEs, benchmark deltas. if that's not your idea of a good read, this isn't your newsletter.

Folds into your digest — weekly by default, monthly if you prefer. Unsubscribe in one click.

$ status

wire 2026-09-03
stories 15

© 2026 RepoJournal Home Showcase Explore How it works Privacy

$ the-wire · showcase

Elixir reworks test manifest for parameterized failures

By RepoJournal · Filed · About Elixir & Phoenix

A fix for parameterized test manifests changes how failures are tracked, ensuring all variants get rerun when any single one fails.

The test manifest now records parameterized test failures differently [1]. Previously, when variant 1, 2, 3, and 4 were run and only 2 failed, the manifest keyed on {module, test} would allow a later pass of a different variant to overwrite the failure, meaning the failing variant might not be rerun. Now "all parameterized tests marked as failing" are kept, so all variants get rerun, preventing sibling failures from being masked. This change conflicts with an earlier approach in [2] that aimed to solve the same issue; the newer PR supersedes it.

Phoenix has reorganized its integration test suites [3]. Tests previously lived in 7 monolithic files, which caused low concurrency at startup and idle worker tails. The suites are now split into focused async modules, better utilizing parallel test runners. Additionally, CI has been improved with "explicit database readiness checks" and more granular phasing [4].

Phoenix also fixes compiler warnings on older Elixir versions [5]. Code in `Phoenix.CodeReloader.Server` used `@requires_consolidation`, which on Elixir < 1.19 always evaluated to true, triggering warnings. The attribute is replaced with version-gated helper functions to eliminate the warnings. Elixir itself begins validating `@enforce_keys` against declared struct fields [6], which will likely make invalid usage a compile-time error in the future.

Action items

References

  1. [1] fix: Test Manifest, parameterized do not override sibling failures ↗ elixir-lang/elixir
  2. [2] Do not overlap sibling failures in parameterized tests failures manifest ↗ elixir-lang/elixir
  3. [3] Split integration test suites into focused async modules ↗ phoenixframework/phoenix
  4. [4] Improve integration test CI visibility, healthchecks, and reliability ↗ phoenixframework/phoenix
  5. [5] Fix Elixir 1.15 compiler warnings ↗ phoenixframework/phoenix
  6. [6] Improve defstruct @enforce_keys ↗ elixir-lang/elixir

Quick answers

What shipped in Elixir & Phoenix on September 3, 2026?
A fix for parameterized test manifests changes how failures are tracked, ensuring all variants get rerun when any single one fails. In total, 7 commits and 8 pull requests landed.
Who contributed to Elixir & Phoenix on September 3, 2026?
4 developers shipped this update, including PragTob, AlexGx, preciz, and rhcarvalho.
What were the notable Elixir & Phoenix updates?
fix: Test Manifest, parameterized do not override sibling failures, Do not overlap sibling failures in parameterized tests failures manifest, and Split integration test suites into focused async modules.