From 43a19f7bb5a6b5f9a521be341cfc42b97964b0ba Mon Sep 17 00:00:00 2001 From: leynos Date: Fri, 23 Oct 2026 20:03:08 +0200 Subject: [PATCH] Docs(execplans): correct stale codec path in issue 287 plan Commit 815dad65 ("Normalize Rust module layout (#710)") moved `src/codec.rs` to `src/codec/mod.rs`, but the issue-287 ExecPlan still cited the old path in two places: the "Context and orientation" summary of where `FrameCodec` is defined, and the "Surprises & Discoveries" observation that the codec layer is generic over `Self::Frame`. Both claims remain true; only the path had drifted. This contradicted the plan's own constraint to "reflect the current repository layout rather than historical issue context", and its published companion `docs/frame-vec-u8-inventory.md` already cited the current path, so the two artefacts disagreed with each other. Verified `FrameCodec` is declared at `src/codec/mod.rs:64`, that no `src/codec.rs` exists, and that the adjacent `src/hooks.rs` and `src/connection/mod.rs` evidence cited alongside it is unchanged. Recorded the correction in the plan's revision note. Co-Authored-By: Claude Code --- ...ventory-trait-bounds-expecting-frame-vec-u8.md | 15 +++++++++++---- 1 file changed, 11 insertions(+), 4 deletions(-) diff --git a/docs/execplans/issue-287-inventory-trait-bounds-expecting-frame-vec-u8.md b/docs/execplans/issue-287-inventory-trait-bounds-expecting-frame-vec-u8.md index d82b00045..0cc87fcf4 100644 --- a/docs/execplans/issue-287-inventory-trait-bounds-expecting-frame-vec-u8.md +++ b/docs/execplans/issue-287-inventory-trait-bounds-expecting-frame-vec-u8.md @@ -92,7 +92,7 @@ echo "markdownlint exit: $?" The current repository already shows a split between generic transport frames and owned payload-byte APIs: -- `src/codec.rs` defines `FrameCodec` generically over `Self::Frame`. +- `src/codec/mod.rs` defines `FrameCodec` generically over `Self::Frame`. - `src/hooks.rs` defines `WireframeProtocol` generically over `type Frame: FrameLike`. - `src/app/envelope.rs`, `src/middleware.rs`, `src/client/hooks.rs`, and @@ -157,9 +157,9 @@ conclusions. before proceeding. - Observation: the runtime no longer hard-codes `F::Frame = Vec` at the - codec or protocol trait layer. Evidence: `src/codec.rs` and `src/hooks.rs`. - Impact: the main migration scope is payload ownership and mutability, not a - blanket frame-type constraint. + codec or protocol trait layer. Evidence: `src/codec/mod.rs` and + `src/hooks.rs`. Impact: the main migration scope is payload ownership and + mutability, not a blanket frame-type constraint. - Observation: the strongest remaining coupling is public payload-oriented API shape. Evidence: `src/app/envelope.rs`, `src/middleware.rs`, @@ -207,3 +207,10 @@ Validation outcomes: This file was restored on 2026-04-11 from commit `c0cbc75` and updated to reflect completion of the documentation work in the current worktree. + +On 2026-10-23 the two `src/codec.rs` references above were corrected to +`src/codec/mod.rs`, the path this module has occupied since commit `815dad65` +("Normalize Rust module layout (#710)"). The inventory document itself already +cited the current path; only this ExecPlan had drifted, which contradicted its +own constraint to "reflect the current repository layout rather than historical +issue context".