Allow workspace brand metadata aliases by oferchen · Pull Request #1411 · oferchen/rsync

oferchen · 2025-10-28T17:43:53Z

Summary

resolve the workspace default brand through Brand::from_str so case-insensitive labels and program-style aliases remain valid
add coverage for the accepted alias set to guard future regressions

Testing

cargo test -p rsync-core branding::tests::resolve_default_brand_accepts_aliases

https://chatgpt.com/codex/tasks/task_e_6900fdc33f34832393a41d43faf502ab

Document the scope of #1593 (async at the transport-level) versus the parallel #1411 (async runtime adoption). Defer the Linux pipe-FD path to #1859, recommend status quo on macOS / BSD / Windows / older Linux, and articulate why the rsync wire-protocol pipeline single-threading caps the win regardless of mechanism.

#3720) Static analysis of whether the SSH transport path warrants an async runtime. Maps every spawn, pipe, and I/O site in the subprocess and russh-embedded paths; lays out where async helps (multi-host fan-out, runtime co-location) versus where it does not (KEX, auth, single transfer); and gives four falsifiable decision criteria for adopting a runtime in the subprocess path.

Document runtime choice (tokio vs smol vs async-std) for any future async SSH transport. Distinct from #1593 (async I/O on the SSH path). Cross-references #1779 (tokio scope audit) and #1780 (single-runtime invariant); recommends tokio + russh, deferred behind --features embedded-ssh.

Expand the placeholder async migration plan into a comprehensive, opinionated design covering: current synchronous threading model per subsystem, per-subsystem cost/benefit, a phased incremental adoption strategy, tokio runtime choice with rejected alternatives, sync/async bridge patterns, backward-compat strategy, risk register, and open questions. Cross-references existing async-related tasks (#1367, #1411, #1591, #1593, #1595, #1674, #1751, #1779, #1780, #1782, #1796, #1797, #1805, #1806, #1889, #1890, #1891, #1892, #1934, #1935, #2136) and related design notes so future planners have a single anchor.

Add docs/design/async-ssh-evaluation.md, the focused evaluation that backs the Phase 3 SSH-transport recommendation in the async migration plan. Documents the current subprocess-pipe path, what async overlap actually buys, the integration boundary and spawn_blocking surfaces, and the embedded-russh-vs-subprocess tradeoff. Recommends embedded russh as the first async SSH target since the internal futures already exist and it eliminates fork+execve per connection; defers async subprocess until embedded is proven.

#3720) Static analysis of whether the SSH transport path warrants an async runtime. Maps every spawn, pipe, and I/O site in the subprocess and russh-embedded paths; lays out where async helps (multi-host fan-out, runtime co-location) versus where it does not (KEX, auth, single transfer); and gives four falsifiable decision criteria for adopting a runtime in the subprocess path.

Document runtime choice (tokio vs smol vs async-std) for any future async SSH transport. Distinct from #1593 (async I/O on the SSH path). Cross-references #1779 (tokio scope audit) and #1780 (single-runtime invariant); recommends tokio + russh, deferred behind --features embedded-ssh.

Expand the placeholder async migration plan into a comprehensive, opinionated design covering: current synchronous threading model per subsystem, per-subsystem cost/benefit, a phased incremental adoption strategy, tokio runtime choice with rejected alternatives, sync/async bridge patterns, backward-compat strategy, risk register, and open questions. Cross-references existing async-related tasks (#1367, #1411, #1591, #1593, #1595, #1674, #1751, #1779, #1780, #1782, #1796, #1797, #1805, #1806, #1889, #1890, #1891, #1892, #1934, #1935, #2136) and related design notes so future planners have a single anchor.

Add docs/design/async-ssh-evaluation.md, the focused evaluation that backs the Phase 3 SSH-transport recommendation in the async migration plan. Documents the current subprocess-pipe path, what async overlap actually buys, the integration boundary and spawn_blocking surfaces, and the embedded-russh-vs-subprocess tradeoff. Recommends embedded russh as the first async SSH target since the internal futures already exist and it eliminates fork+execve per connection; defers async subprocess until embedded is proven.

Expand the placeholder async migration plan into a comprehensive, opinionated design covering: current synchronous threading model per subsystem, per-subsystem cost/benefit, a phased incremental adoption strategy, tokio runtime choice with rejected alternatives, sync/async bridge patterns, backward-compat strategy, risk register, and open questions. Cross-references existing async-related tasks (#1367, #1411, #1591, #1593, #1595, #1674, #1751, #1779, #1780, #1782, #1796, #1797, #1805, #1806, #1889, #1890, #1891, #1892, #1934, #1935, #2136) and related design notes so future planners have a single anchor.

Add docs/design/async-ssh-evaluation.md, the focused evaluation that backs the Phase 3 SSH-transport recommendation in the async migration plan. Documents the current subprocess-pipe path, what async overlap actually buys, the integration boundary and spawn_blocking surfaces, and the embedded-russh-vs-subprocess tradeoff. Recommends embedded russh as the first async SSH target since the internal futures already exist and it eliminates fork+execve per connection; defers async subprocess until embedded is proven.

Allow workspace brand metadata aliases

841deb9

oferchen added the codex label Oct 28, 2025 — with ChatGPT Codex Connector

oferchen merged commit 94b5339 into master Oct 28, 2025

oferchen deleted the prepare-oc-rsync-for-production-release branch October 28, 2025 17:44

oferchen mentioned this pull request May 1, 2026

docs(rsync_io): audit async I/O for SSH transport path (#1593) #3508

Merged

9 tasks

oferchen mentioned this pull request May 7, 2026

docs(audits): async I/O for SSH transport evaluation #3819

Merged

2 tasks

oferchen mentioned this pull request May 16, 2026

docs(design): async migration plan with incremental adoption strategy (#1594) #4186

Merged

4 tasks

oferchen mentioned this pull request May 16, 2026

docs(design): async runtime evaluation for SSH transport (#1411) #4194

Merged

3 tasks

oferchen mentioned this pull request May 17, 2026

bench(transport): SSH sync vs async I/O on slow links (#1889) #4251

Merged

4 tasks

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Allow workspace brand metadata aliases#1411

Allow workspace brand metadata aliases#1411
oferchen merged 1 commit into
masterfrom
prepare-oc-rsync-for-production-release

oferchen commented Oct 28, 2025

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

Uh oh!

Conversation

oferchen commented Oct 28, 2025

Summary

Testing

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant