Port §5 workingSetFit metric to Go + Rust source trees

Status: ⚠ lossy · Date: 2026-05-24 · PR: #32 · Commit: 43c6f7a

Related posts: sama-v2-go-project-dive · sama-v2-rust-project-ripgrep · sama-v2-workingset-cross-repo-baseline


Recovered partially. The /goal text below was reconstructed from conversation summary, not verbatim from the PR body. Original wording may differ from what the user typed.

Recovered opening (from prior-session conversation summary, paraphrased):

Goal: Port the §5 workingSetFit metric to Go and Rust source trees. Run it against /tmp/dive and /tmp/ripgrep at pinned SHAs; replace the hand-estimated workingSetFit values in the dive + ripgrep audit blog posts with measured numbers; convert the cross-repo argument from "n=1 measured + n=3 estimated" to "n=3 measured + n=1 estimated".

What is preserved from the original /goal: only the opening sentence above, recovered from the prior-session conversation summary at the start of this conversation. The full Done when clauses, Constraints (anti-fudge), and Load-bearing files sections did not survive summarization.

What landed: src/b32_working_set_polyglot.ts (pure Layer 1) + src/c14_working_set_walker.ts (Layer 2 adapter) + scripts/measure-working-set.ts (CLI). 24 new tests covering bound-edge inclusivity + Go-vs-Rust test-file asymmetry (Go excludes *_test.go; Rust includes all .rs per /sama/v2 §6.2 inline-tests dialect). Measured results: dive @d6c69194 = 52.17% (originally hand-estimated ~80%, a 28-point miss); ripgrep @4519153e = 54.00% (originally ~60%, a 6-point miss). See PR #32.