fix(atlas): give every sub-national region a distinct code (#1217)

geoBoundaries fills shapeISO with the bare country code for some countries (every
Spanish region got "ESP", every Chinese "CHN", also Chile/Oman), so marking one
region lit up the whole country. build-atlas-geo.mjs now keeps shapeISO only when
it is a real "XX-..." subdivision code and otherwise synthesizes a unique
per-country id from the region name. Regenerated admin1.geojson.gz: Spain/China/
Chile/Oman now carry distinct region codes (countries with real codes, e.g.
Germany, are unchanged).

server/scripts/build-atlas-geo.mjs

@@ -151,18 +151,37 @@ function normalizeAdm0Feature(f) {
 
 function normalizeAdm1(geo, a3, countryName) {
   if (!geo?.features) return []
+  const a2 = A3_TO_A2[a3] || null
+  // Ensure every region in a country ends up with a distinct iso_3166_2 — the Atlas
+  // marks/unmarks regions by this code, so duplicates make one mark light up the whole
+  // country.
+  const used = new Set()
+  const uniq = (base) => {
+    let code = base, n = 2
+    while (used.has(code)) code = `${base}-${n++}`
+    used.add(code)
+    return code
+  }
   return geo.features.map(f => {
     const name = f.properties?.shapeName || ''
     const geometry = quantizeGeometry(f.geometry, ADM1_DECIMALS)
     if (!geometry) return null
-    const a2 = A3_TO_A2[a3] || null
+    // shapeISO is a real ISO 3166-2 code for most features, but geoBoundaries sometimes
-    // shapeISO is a real ISO 3166-2 code for ~90% of features; geoBoundaries leaves the
+    // fills it with the bare country code instead of a subdivision code — e.g. every
-    // rest blank or uses an `XX_YYY` placeholder. Keep real/placeholder codes as-is
+    // Spanish region gets "ESP", every Chinese "CHN" (also CL/OM). Keep it only when it
-    // (stable per polygon → manual mark/unmark works, real ones match Nominatim). For
+    // is a real `XX-…` subdivision code and not already taken; otherwise synthesize a
-    // blank codes, synthesize a stable id mirroring the server's geocode fallback so
+    // stable, unique-per-country id from the region name so each region is independently
-    // every region is still markable.
+    // markable.
-    let code = f.properties?.shapeISO || ''
+    const raw = f.properties?.shapeISO || ''
-    if (!code && a2) code = `${a2}-${name.replace(/[^A-Za-z0-9]/g, '').substring(0, 3).toUpperCase()}`
+    let code
+    if (/^[A-Za-z]{2}-[A-Za-z0-9]+$/.test(raw) && !used.has(raw)) {
+      code = raw
+      used.add(code)
+    } else if (a2) {
+      code = uniq(`${a2}-${name.replace(/[^A-Za-z0-9]/g, '').toUpperCase() || 'RGN'}`)
+    } else {
+      code = raw
+    }
     return {
       type: 'Feature',
       // Property names the Atlas region layer + server getRegionGeo already read.

server/tests/unit/services/atlasBundle.test.ts

@@ -0,0 +1,36 @@
+import { describe, it, expect } from 'vitest';
+import fs from 'fs';
+import path from 'path';
+import zlib from 'zlib';
+
+// Data-integrity guard for the shipped Atlas region bundle. geoBoundaries fills
+// shapeISO with the bare country code for some countries (every Spanish region got
+// "ESP", every Chinese "CHN", also CL/OM), which made marking one region light up the
+// whole country (#1217). build-atlas-geo.mjs now synthesizes a unique per-region code
+// for those; this asserts the shipped bundle actually carries distinct codes.
+describe('Atlas admin1 region bundle (#1217)', () => {
+  const bundlePath = path.join(__dirname, '..', '..', '..', 'assets', 'atlas', 'admin1.geojson.gz');
+  const features = JSON.parse(zlib.gunzipSync(fs.readFileSync(bundlePath)).toString()).features as {
+    properties: { iso_a2: string | null; iso_3166_2: string };
+  }[];
+
+  const regions = (a2: string) => features.filter(f => f.properties.iso_a2 === a2);
+
+  it('ATLAS-BUNDLE-001 — previously-broken countries now have distinct region codes', () => {
+    for (const a2 of ['ES', 'CN', 'CL', 'OM']) {
+      const f = regions(a2);
+      expect(f.length, `${a2} should ship regions`).toBeGreaterThan(1);
+      expect(new Set(f.map(r => r.properties.iso_3166_2)).size, `${a2} region codes must be unique`).toBe(f.length);
+    }
+  });
+
+  it('ATLAS-BUNDLE-002 — countries with real ISO codes keep them and stay unique', () => {
+    for (const a2 of ['DE', 'FR', 'US']) {
+      const f = regions(a2);
+      expect(f.length).toBeGreaterThan(1);
+      // real ISO 3166-2 form, e.g. DE-BW
+      expect(f.some(r => /^[A-Z]{2}-[A-Z0-9]+$/.test(r.properties.iso_3166_2))).toBe(true);
+      expect(new Set(f.map(r => r.properties.iso_3166_2)).size).toBe(f.length);
+    }
+  });
+});