RepoJournal
Elixir & Phoenix Elixir & Phoenix
138 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 2026-10-15
stories 6
RSS Archive Reviews

© 2026 RepoJournal Home Showcase How it works Privacy

$ the-wire · showcase

Stepped binary_slice goes fast path, non-empty bitstring comprehensions inline

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

preciz
1 person shipped this
preciz @preciz 3 cited

Elixir's binary_slice/2 now takes a fast path for single-byte stepped ranges, and a proposed compiler change would inline comprehensions into non-empty bitstring literals, at the cost of some changed error messages.

Optimize Kernel.binary_slice/2 for stepped ranges elixir-lang/elixir

by preciz

Fast-paths single-byte stepped slices with binary_part/3 and precomputes skip = step - 1 before the binary comprehension, so the Erlang compiler can drop the per-iteration size subtraction and fallback match clauses. This finishes the work started in the earlier stepped-range optimization, meaning binary_slice/2 with a first..last//step range where the step selects single bytes avoids the gener...

Inline comprehensions into non-empty bitstrings elixir-lang/elixir

by preciz

Previously only into: "" was compiled inline and every other bitstring went through the Collectable protocol; now binary generators append the comprehension to the prefix and other generators start the accumulator with it. The author, preciz, flags unavoidable regressions and writes that "Some raised error messages change," so treat this as a review-stage proposal rather than a settled behavior...

Optimize Kernel.binary_slice/2 for stepped ranges (#15989) elixir-lang/elixir

by Barna Kovacs

The commit lands the stepped-range optimization for Kernel.binary_slice/2, so it is the part of this work already on main rather than proposed.

$ ls elixir/reviews/ # the step-back reads

the week in review Phoenix rewrites every CI workflow to hardened patterns the month in review Phoenix adds CI test summary formatter
Full archive → Weekly reviews →

Quick answers

What shipped in Elixir & Phoenix on October 15, 2026?
Elixir's binary_slice/2 now takes a fast path for single-byte stepped ranges, and a proposed compiler change would inline comprehensions into non-empty bitstring literals, at the cost of some changed error messages. In total, 2 commits and 4 pull requests landed.
Who contributed to Elixir & Phoenix on October 15, 2026?
1 developer shipped this update, including preciz.
What were the notable Elixir & Phoenix updates?
Optimize Kernel.binary_slice/2 for stepped ranges, Inline comprehensions into non-empty bitstrings, and Optimize Kernel.binary_slice/2 for stepped ranges (#15989).
♥

$ log Elixir & Phoenix --recent

Algorithm L replaces inside-out shuffles in Enum 3 shipped Base validation regression fixed, formatter comments stay put 16 shipped Formatter bugs that changed program meaning, and a LiveView assign that always fired 25 shipped Base validators go SWAR, formatter output stabilizes 4 shipped

$ log --all # elsewhere on the wire

Alignment & Post-training verl standardizes low-precision rollout kernels on DeepGEMM Agent Security Guardrails exposes hooks over HTTP, certifi floor raised in agent-scan

$ ls showcases/

Safety Evals AI Red Teaming Agent Security Interpretability Alignment & Post-training

$ init your-desk

RepoJournal can write one of these for any repo you care about. Sign in with GitHub, pick your repos and sources, and your desk gets a wire like this one every morning.