spec(port): port specs/port/trinity/src/vibeec/soul_collector.t27 (Closes #9719)

[dmitrii-f-t27]

Summary

Port gHashTag/trinity:src/vibeec/soul_collector.zig (at ee512359254b, 69 lines) as specs/port/trinity/src/vibeec/soul_collector.t27 — Roadmap Stage 5 (goal #4547), target t27c gen-rust. Closes #9719.

The original walks the project tree and gathers every sacred text into trinity_corpus.txt.

Per the issue's rule ("Port the decision, not the plumbing"), the one listed function is ported under its name as a pure function (no transports — the directory walk, the reads and the corpus writes are the callers' plumbing) with 5 tests (33 runtime asserts, 0 vacuous, 100% function coverage):

• the suffix scan (original line 31-32's endsWith): byte-indexed, compares the path's last suffix.len bytes;
• the marker scan (lines 35-37's indexOf != null): a sliding window across the path;
• the sacred filter (line 39): a file enters the corpus exactly when it ends with .tri or .zig and mentions none of .git, zig-cache, temp_generated anywhere in its path — including the original's quirk that .gitignore.zig is profane too (its indexOf(".git") matches);
• the 10 MB read cap (line 47): a file at most the cap in size is absorbed; a larger one fails the capped read and the walk continues without it (the original's catch |continue|);
• main: the walk's per-file decision — the filter and the cap together decide ABSORBED or SKIPPED.

Verification (all run live from the worktree root, zig 0.16.0)

• t27c parse: OK. t27c spec-status: IMPLEMENTED. t27c typecheck: 0 errors, 0 warnings.
• t27c coverage: 5/5 tested (100%) — no transports in this port.
• t27c test-report: 5 tests / 5 pass / 0 FAIL / 0 BLOCKED, 0 vacuous passes — the suffix scan on match/mismatch/shorter-than-suffix/empty cases, the marker scan including the .gitignore quirk and a mid-string needle, the filter on eight paths covering every exclusion and both extensions, the cap boundary at exactly 10485760 and one byte over, and the walk decision end-to-end.
• All four generators were run before the push: gen (138 lines, 0 'not yet implemented'), gen-rust, gen-c, gen-verilog.
• grep -cE '^\s*(pub )?fn (main)\(': 1. grep -cE '^[[:space:]]*test[[:space:]]+("|[A-Za-z_])': 5 (≥ 1).
• Boundary: git status --short — exactly this one file. The repo's lefthook pre-commit passed.
• dupe_scan: all 5 function names unique across the corpus (checked before the commit); the bare scan reports zero error lines — no new duplicate group, none grew.

Dmitrii Fedorov (@dmitrii-f-t27)

[dmitrii-f-t27]

@gHashTag Please merge this PR into master via the golden chain — rights on gHashTag/t27 are read-only for me (403), so only you can merge it.

Verification of this head (285b6bd, on current master — no extra merge needed):

• Required CI trio green: validate, check-linked-issue, parse-ratchet; also Check L1 TRACEABILITY, duplicate-bodies and cli-tri green.
• The t27b-native-run artifact of this head (commit 285b6bd, run 39069710520) row for specs/port/trinity/src/vibeec/soul_collector.t27: reference pass, t27b pass, tests 5, invariants 0, asserts 33, blockers [], detail empty, reference_disagree: [].
• All four generators were run locally before the push (gen, gen-rust, gen-c, gen-verilog) — this spec emits cleanly everywhere.
• The only red checks are proven base-red on master WITHOUT this PR:
  • Corpus ratchet: the failing step is 'Every conflicted type name has a written verdict' — docs/reports/type_conflicts_classified.json disagrees with the tree (UNJUDGED/STALE). The identical error fails on master itself (workflow runs on 18c98cc and b655dce, and every PR of this series). This spec declares no struct or enum, so it cannot add a conflicted type name.
  • t27b-native-ratchet: the registration debt, red on master across this whole port series.

After the merge I will: verify the spec file live on the merge SHA, confirm #9719 auto-closed, and file the 14-step report.

Dmitrii Fedorov (@dmitrii-f-t27)

[gHashTag]

Landed unchanged (as a merge commit) in #10174, together with ten other green ports, on top of the master fixes #10160/#10172. Closing this as superseded; the linked issue closes when #10174 merges.