test(04-01): RED — gazetteer schema/license gates (G1-G11) + three-environment loader proof (L1-L3)

- G1-G6: dataset invariants — schema, slug-id uniqueness, canonical-name
  law ('KLCC' is an alias, never a name), DUKE-as-alias, KL bbox, the
  ODbL header traveling inside the JSON
- G7-G9: the licensing gate — verbatim GPL-3.0 LICENSE, LICENSES.md,
  the README no longer claiming 'all rights reserved'
- G10-G11: no logging/storage anywhere in the module (D-15); the loader
  builds its index inside a function, never at module top level
- L1-L3: the phase's top mechanical risk — bare-Node loading with the
  SRCH-01/negative probes, the static JSON import law, esbuild INLINING
  the dataset + the wrangler dry-run build guard

13 cases fail while the dataset, loader, LICENSE files, and README fix
are absent — the RED gate.

search/test/loader-env.test.mjs

@@ -0,0 +1,135 @@
+#!/usr/bin/env node
+/*
+ * loader-env.test.mjs — the THREE-ENVIRONMENT loader proof (D-07).
+ * ----------------------------------------------------------------
+ * THE phase's top mechanical risk (RESEARCH): the JSON dataset must be
+ * reachable identically from (1) the bare-Node corpus runner, (2) the
+ * Vite bundle, and (3) the wrangler/esbuild Worker bundle — Workers have
+ * no fs, so the dataset MUST be a static build-time import, never a
+ * runtime fetch.
+ *
+ *   L1  bare Node (type-stripping): the module graph loads, the dataset
+ *       parses, search('KLCC') resolves suria-klcc, out-of-area is refused
+ *   L2  static JSON import in the loader + (once plan 04-04 lands) the
+ *       Vite bundle carries the dataset marker; until then that half is
+ *       SKIPPED visibly while the static-import half is still asserted
+ *   L3  esbuild — the bundler family wrangler uses — INLINES the dataset
+ *       (its bundle carries the 'suria-klcc' marker), and the Worker
+ *       dry-run build guard exits 0
+ *
+ * Run:  npm run search:test     (exit 0 = all green)
+ */
+import { strict as assert } from "node:assert";
+import { spawnSync } from "node:child_process";
+import { existsSync, mkdirSync, readdirSync, readFileSync } from "node:fs";
+import path from "node:path";
+import { fileURLToPath } from "node:url";
+import { test } from "./harness.mjs";
+
+const here = path.dirname(fileURLToPath(import.meta.url)); // search/test
+const searchDir = path.resolve(here, ".."); // search/
+const repoRoot = path.resolve(searchDir, ".."); // repo root
+const esbuildBin = path.join(repoRoot, "node_modules", ".bin", "esbuild");
+const wranglerBin = path.join(repoRoot, "node_modules", ".bin", "wrangler");
+
+// Lazy index load — missing module => per-case ERR_MODULE_NOT_FOUND (the RED state).
+let indexPromise = null;
+function loadIndexModule() {
+  if (!indexPromise) {
+    indexPromise = import("../src/index.ts");
+  }
+  return indexPromise;
+}
+
+test("L1 bare Node (type-stripping) — the loader works with zero infrastructure", async () => {
+  const mod = await loadIndexModule();
+  assert.ok(mod.getEntries().length >= 30, "the dataset parses in bare Node");
+  const source = mod.createGazetteerSource();
+  assert.equal(source.name, "gazetteer");
+  const hits = await source.search({ text: "KLCC" });
+  assert.ok(hits.length > 0, "KLCC must resolve");
+  assert.equal(hits[0].entry.id, "suria-klcc", "SRCH-01: KLCC resolves to suria-klcc at rank 1");
+  assert.equal((await source.search({ text: "Kuching" })).length, 0, "out-of-area is refused honestly (D-13)");
+  assert.equal((await source.search({ text: "Penang" })).length, 0, "out-of-area is refused honestly (D-13)");
+  assert.equal(source.byId("suria-klcc")?.name, "Suria KLCC", "byId is synchronous and honest");
+  assert.equal(source.byId("../etc/passwd"), null, "byId rejects non-slug ids");
+});
+
+test("L2 static JSON import in the loader (Vite-bundle half guarded until 04-04)", async () => {
+  const loaderPath = path.join(searchDir, "src", "gazetteer.ts");
+  const loaderSrc = readFileSync(loaderPath, "utf8");
+  assert.match(
+    loaderSrc,
+    /import raw from "\.\/data\/kl-gazetteer\.json"/,
+    "the dataset must be a STATIC import (build-time inlined)",
+  );
+  assert.ok(!/fetch\(/.test(loaderSrc), "a runtime fetch would hard-fail in a Worker");
+  assert.ok(!/readFileSync/.test(loaderSrc), "Workers have no fs — never read the dataset at runtime");
+
+  // The Vite-bundle half is only assertable once plan 04-04 wires web/ to
+  // @araharah/search. Until then: visible skip, static-import half asserted.
+  const webSrcDir = path.join(repoRoot, "web", "src");
+  let webImportsSearch = false;
+  try {
+    for (const entry of readdirSync(webSrcDir, { recursive: true })) {
+      const rel = String(entry);
+      if (rel.endsWith(".ts") || rel.endsWith(".tsx")) {
+        if (readFileSync(path.join(webSrcDir, rel), "utf8").includes("@araharah/search")) {
+          webImportsSearch = true;
+          break;
+        }
+      }
+    }
+  } catch {
+    // no web/src — treat as not-yet-landed
+  }
+  if (!webImportsSearch) {
+    console.log("SKIP L2 (web-bundle marker) — web/ does not import @araharah/search yet (lands in plan 04-04); the static-import half is asserted above");
+    return;
+  }
+  const distAssets = path.join(repoRoot, "web", "dist", "assets");
+  assert.ok(existsSync(distAssets), "web/dist/assets must exist once web imports the search module");
+  let marker = false;
+  for (const file of readdirSync(distAssets)) {
+    if (file.endsWith(".js") && readFileSync(path.join(distAssets, file), "utf8").includes("suria-klcc")) {
+      marker = true;
+      break;
+    }
+  }
+  assert.ok(marker, "the Vite bundle must inline the gazetteer (build-time, never fetched)");
+});
+
+test("L3 esbuild inlines the gazetteer; the wrangler dry-run build guard exits 0", () => {
+  // — the inline half (provable at wave 1): bundle the search module with the
+  //   SAME bundler family wrangler uses and grep the output for the marker —
+  const bundleDir = "/tmp/opencode/araharah-search-bundle";
+  mkdirSync(bundleDir, { recursive: true });
+  const outfile = path.join(bundleDir, "index.js");
+  const bundle = spawnSync(
+    esbuildBin,
+    [path.join(searchDir, "src", "index.ts"), "--bundle", `--outfile=${outfile}`, "--log-level=error"],
+    { encoding: "utf8", timeout: 60000 },
+  );
+  assert.equal(bundle.status, 0, `esbuild bundle failed: ${bundle.stderr || bundle.stdout}`);
+  const bundled = readFileSync(outfile, "utf8");
+  assert.ok(
+    bundled.includes("suria-klcc"),
+    "a static JSON import MUST be inlined at build time — a runtime fetch would be a hard Worker failure",
+  );
+
+  // — the Worker dry-run half (build regression guard ONLY): the bundle-
+  //   marker grep for the WORKER bundle is plan 04-03 task 2's criterion,
+  //   after bot.ts imports createGazetteerSource() —
+  if (!existsSync(path.join(repoRoot, "web", "dist", "index.html"))) {
+    console.log("SKIP L3 (wrangler dry-run) — web/dist is not built; run npm run build first");
+    return;
+  }
+  const dryrunDir = "/tmp/opencode/araharah-worker-dryrun";
+  mkdirSync(dryrunDir, { recursive: true });
+  const dry = spawnSync(wranglerBin, ["deploy", "--dry-run", `--outdir=${dryrunDir}`], {
+    cwd: path.join(repoRoot, "bot"),
+    encoding: "utf8",
+    timeout: 180000,
+  });
+  assert.equal(dry.status, 0, `wrangler dry-run failed: ${(dry.stderr || dry.stdout).slice(-800)}`);
+});