test(04-01): RED — OSA distance vectors (D1-D9) + disjoint-tier scorer suite (S1-S7)

- 9 hand-computed Damerau-Levenshtein vectors (transpositions, empties, kitten/sitting sanity, the corpus's klcc/kclc + sunway/sunwya typos) - 7 scorer laws: tier literals + disjointness, min-length guard before exact, exact-outranks-fuzzy, mid-word PREFIX vs token-boundary TOKEN_SET, fuzzy residual 0.75*0.60 arithmetic, the 0.72 admission boundary (Penang/Padang 0.667 refuses), matchedAlias raw provenance All 16 new cases fail while src/distance.ts + src/score.ts are absent.
2026-10-24 00:25:19 +08:00
parent a9f9ab380c
commit 25ee1e9c61
2 changed files with 196 additions and 0 deletions
@@ -0,0 +1,80 @@
+#!/usr/bin/env node
+/*
+ * distance.test.mjs — OSA Damerau-Levenshtein vector suite (SRCH-02).
+ * -------------------------------------------------------------------
+ * RED phase (plan 04-01 task 2): every case fails while
+ * ../src/distance.ts does not exist (lazy import => per-case FAIL).
+ *
+ * The vectors are the hand-computed ones the plan pins (RESEARCH
+ * §"The Damerau-Levenshtein (OSA) distance"). The 3-band row rotation
+ * is the classic implementation trap — these vectors are written FIRST
+ * so a wrong rotation shows up as a red suite, not a silent mis-rank.
+ *
+ *   D1  adjacent transposition          ab       -> ba        = 1
+ *   D2  two transpositions              a cat    -> an act    = 2
+ *   D3  empty vs empty                                        = 0
+ *   D4  empty vs abc                                          = 3
+ *   D5  abc vs empty                                          = 3
+ *   D6  identical strings                                     = 0
+ *   D7  KLCC transposition typo         klcc     -> kclc      = 1
+ *   D8  real corpus typo                sunway   -> sunwya    = 1
+ *   D9  classic Levenshtein sanity      kitten   -> sitting   = 3
+ *
+ * Run:  npm run search:test     (exit 0 = all green)
+ */
+import { strict as assert } from "node:assert";
+import { test } from "./harness.mjs";
+
+// Lazy distance load — missing module => per-case ERR_MODULE_NOT_FOUND (the RED state).
+let distancePromise = null;
+function loadDamerau() {
+  if (!distancePromise) {
+    distancePromise = import("../src/distance.ts").then((m) => m.damerau);
+  }
+  return distancePromise;
+}
+
+test("D1 adjacent transposition: damerau('ab', 'ba') === 1", async () => {
+  const damerau = await loadDamerau();
+  assert.equal(damerau("ab", "ba"), 1);
+});
+
+test("D2 two transpositions: damerau('a cat', 'an act') === 2", async () => {
+  const damerau = await loadDamerau();
+  assert.equal(damerau("a cat", "an act"), 2);
+});
+
+test("D3 damerau('', '') === 0", async () => {
+  const damerau = await loadDamerau();
+  assert.equal(damerau("", ""), 0);
+});
+
+test("D4 damerau('', 'abc') === 3", async () => {
+  const damerau = await loadDamerau();
+  assert.equal(damerau("", "abc"), 3);
+});
+
+test("D5 damerau('abc', '') === 3", async () => {
+  const damerau = await loadDamerau();
+  assert.equal(damerau("abc", ""), 3);
+});
+
+test("D6 identical strings: damerau('same', 'same') === 0", async () => {
+  const damerau = await loadDamerau();
+  assert.equal(damerau("same", "same"), 0);
+});
+
+test("D7 KLCC transposition typo: damerau('klcc', 'kclc') === 1", async () => {
+  const damerau = await loadDamerau();
+  assert.equal(damerau("klcc", "kclc"), 1);
+});
+
+test("D8 real corpus typo: damerau('sunway', 'sunwya') === 1", async () => {
+  const damerau = await loadDamerau();
+  assert.equal(damerau("sunway", "sunwya"), 1);
+});
+
+test("D9 classic Levenshtein sanity: damerau('kitten', 'sitting') === 3", async () => {
+  const damerau = await loadDamerau();
+  assert.equal(damerau("kitten", "sitting"), 3);
+});