RepoJournal
Elixir & Phoenix Elixir & Phoenix
156 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 2026-11-09
stories 20

© 2026 RepoJournal Home Showcase How it works Privacy

$ the-wire · showcase

Spec-only Elixir fixes: map_get, Process.spawn, send_after

By RepoJournal · Filed · About Elixir & Phoenix · Composed from the cited sources · methodology

Elixir spent the day correcting what its typespecs and docs claim about core Process, File, List, and pin operators, while Ecto touched up Ecto.HMAC documentation; no runtime behavior changes for code already written against the real semantics.

Treat map_get/2 with non-atom keys as failing on values elixir-lang/elixir

by José Valim

Elixir now treats map_get/2 with non-atom keys as failing on values rather than only on the map argument, closing the gap where a spec allowed keys the function would reject. That is the one change in this batch that can surface new type-checking failures: code passing a non-atom to map_get/2 in a pinned or guard position should be corrected, not silenced.

Correct Process.spawn monitor options in typespecs and documentation (#16083) elixir-lang/elixir

by Barna Kovacs

The monitor options in Process.spawn's typespecs and documentation were wrong and are now corrected, so dialyzer and editors stop endorsing option shapes the function does not accept. If your code was written to satisfy the old spec, re-read the docs before trusting it.

Correct the absolute-time argument type for Process.send_after/4 (#16084) elixir-lang/elixir

by Barna Kovacs

Process.send_after/4's absolute-time argument type is fixed in the spec, which matters for anyone passing a monotonic or absolute timestamp instead of a relative millisecond timeout. Specs that previously type-checked against the wrong type may now complain.

Include closed-device errors in File.stat/2 documentation and spec (#16078) elixir-lang/elixir

by Barna Kovacs

File.stat/2 documentation and its spec now include the closed-device errors the function can actually return. Code that matched only :enoent and :eacces against File.stat/2 was matching against an incomplete list of outcomes; the spec now names them.

Small documentation improvements for Ecto.HMAC elixir-ecto/ecto

by LostKobrakai

The long tail is documentation: List.pop_at/3 now spells out its return value and negative indexing, List.pop_at/3 and File.stat/2 clarify behavior already in place, and Ecto.HMAC got small doc improvements in two branches of the same change.

Quick answers

What shipped in Elixir & Phoenix on November 9, 2026?
Elixir spent the day correcting what its typespecs and docs claim about core Process, File, List, and pin operators, while Ecto touched up Ecto.HMAC documentation; no runtime behavior changes for code already written against the real semantics. In total, 13 commits and 7 pull requests landed.
Who contributed to Elixir & Phoenix on November 9, 2026?
3 developers shipped this update, including Barna Kovacs, José Valim, and LostKobrakai.
What were the notable Elixir & Phoenix updates?
Treat map_get/2 with non-atom keys as failing on values, Correct Process.spawn monitor options in typespecs and documentation (#16083), and Correct the absolute-time argument type for Process.send_after/4 (#16084).