Repository navigation
T41 — Split the box manager's tests and handle into their own files - #2454
mitodrummer merged 1 commit into
Conversation
Move the test module to box_manager/tests.rs and the handle types (BoxManagerHandle, WeakBoxManagerHandle, BoxControl) to box_manager/handle.rs. The handles are re-exported at the old path. No behaviour change.
|
Important Review skippedBot user detected. To trigger a single review, invoke the ⚙️ Run configuration
You can disable this status message by setting the Use the checkbox below for a quick retry:
Comment |
|
Approved: no findings at Nothing to do. Verdict APPROVE, posted as a comment: the App cannot review its own pull request; the check and the label carry it. Clobber: none seen. |
There was a problem hiding this comment.
arch-review: PASS · the full ruling
This change moves the box manager's tests and handle types into box_manager/tests.rs and box_manager/handle.rs. The code stays in the same module and crate, and the change does not change behaviour. The design of record does not govern the module layout in minimald, so this is a PASS by absence. It follows the rulings on #2453 and #2452.
Ruled at 2e4ae77 against gominimal/arch@109541a48c31 by claude-opus-5-5.
|
Round 1 of 10 on a failed check suite: the coder changed nothing. Revise round 1: I made no change. The directive names two failed checks, ci-macos-success and e2e, but the runtime could not read either job log (token mint refused, HTTP 422). So I had no failure text to act on. I checked the move against the parent commit. A line-sorted diff of the old box_manager.rs against the three new files shows only the expected changes: the mod and use lines, pub(super) on the handle fields, super:: paths in doc links and calls, rustfmt line joins, and one added #[cfg(target_os = "linux")] on the RwLock import in handle.rs. That import matches the linux-gated field that uses it. Every cfg gate is kept, and the tests module is still #[cfg(test)]. On Linux, cargo clippy -p minimald --all-targets gives no warnings in the box_manager files, cargo fmt --check is clean, and the test line passes. minimald does not build natively on macOS (AGENTS.md platform matrix), so this move is not a likely cause of the macOS lane failure. I was not able to run the ma…[truncated] |
mitodrummer
left a comment
There was a problem hiding this comment.
Babysit review on 2e4ae77 against T41's plan block. This is a pure move: box_manager.rs -1950, box_manager/handle.rs +445, box_manager/tests.rs +1512. After stripping whitespace, the added lines match the removed lines except for import regrouping, pub(super) on the handle fields, doc links re-pathed to super::BoxManager, and rustfmt rewrapping. No logic changed. The macOS e2e is a runner rerun that affects every PR in the walk.
T41 — Split the box manager's tests and handle into their own files
slice: S0
kind: enablement
covers: none
size: S
est: 120 code, 60 tests
executor: agent
difficulty: routine
model: light
diagnostics: none: a pure move, no behaviour
observability: none: a pure move, no signal
layers:
- manager:
- crates/minimald/src/box_manager.rs:
mod tests;, handle out- crates/minimald/src/box_manager/tests.rs (new): moved
- crates/minimald/src/box_manager/handle.rs (new): moved
test: cargo nextest run -p minimald destroy_removes_a_non_running_session
produces: none
consumes: none
proof: The manager behaves as before, tests split out.
Proof
box_manager.rs is now about 1,600 lines in place of 3,537. Its tests are in box_manager/tests.rs and its handle types are in box_manager/handle.rs. All 44 box_manager tests pass unchanged, so the manager behaves as before.
Notes from the agent: This is a pure move. BoxManagerHandle, WeakBoxManagerHandle and BoxControl are re-exported from box_manager with
pub use, so callers keep their paths. Their fields are now pub(super), because the manager actor builds and reads them. The common::SpecHash and HashSet imports moved from box_manager.rs into handle.rs and tests.rs. Intra-doc links in handle.rs to BoxManager now use explicit super:: targets. I also rancargo nextest run -p minimald box_manager::once as an extra check: 44 passed. The clippy warnings that remain are in files this task did not touch (box_runtime.rs, net/switch.rs, minimald-rpc).Tests
cargo nextest run -p minimald destroy_removes_a_non_running_sessionVerify
min task run agent-verifyexited 101 under rustc 1.98.1, which the repository does not pin (1.97.0 in rust-toolchain.toml); the PR is not a draft on that alone — CI, which honours the pin, is the gate.Closes #2379