thaizip 0.7.3 → 0.7.4

package/README.md

@@ -1,9 +1,17 @@
 # thaizip
 
+[![npm version](https://img.shields.io/npm/v/thaizip.svg)](https://www.npmjs.com/package/thaizip)
+[![npm downloads](https://img.shields.io/npm/dm/thaizip.svg)](https://www.npmjs.com/package/thaizip)
+[![license](https://img.shields.io/npm/l/thaizip.svg)](https://github.com/naay99999/thai-zip/blob/main/LICENSE)
+
 Fast fuzzy autocomplete for Thai addresses — subdistrict, district, province, postal code. Thai and English input, zero runtime dependencies, React optional.
 
 **Docs:** https://naay99999.github.io/thai-zip/ (ไทย / [English](https://naay99999.github.io/thai-zip/en/))
 
+| Thai input | English input (romanization alias) |
+|---|---|
+| ![thaizip demo — typing "ลาดพร้าว" and getting fuzzy Thai suggestions](.github/assets/demo.png) | ![thaizip demo — typing "ladprao" and matching the same Thai suggestions](.github/assets/demo-en.png) |
+
 ```bash
 npm install thaizip
 ```
@@ -59,6 +67,8 @@ lookupByZipCode(index, '45000') // exact
 lookupByZipCode(index, '450')   // prefix
 ```
 
+Direct ZIP lookup rejects raw inputs longer than 1000 characters before scanning them. A finite `zipLimit` also bounds how many matching records are collected.
+
 ### English input
 
 The dataset indexes official RTGS spellings, so `bang rak` and `chatuchak` work directly. 87 common non-RTGS spellings are mapped on first: `lardprao` → `lat phrao`, `ladkrabang` → `lat krabang`, `krungthep` → `bangkok`. It is a curated dictionary, not a transliterator — unlisted spellings still miss. Opt out with `romanizationAliases: false`.

package/dist/data.cjs

@@ -2,8 +2,13 @@
 
 var __defProp = Object.defineProperty;
 var __getOwnPropNames = Object.getOwnPropertyNames;
-var __esm = (fn, res) => function __init() {
-  return fn && (res = (0, fn[__getOwnPropNames(fn)[0]])(fn = 0)), res;
+var __esm = (fn, res, err) => function __init() {
+  if (err) throw err[0];
+  try {
+    return fn && (res = (0, fn[__getOwnPropNames(fn)[0]])(fn = 0)), res;
+  } catch (e) {
+    throw err = [e], e;
+  }
 };
 var __export = (target, all) => {
   for (var name in all)

package/dist/data.js

@@ -1,7 +1,12 @@
 var __defProp = Object.defineProperty;
 var __getOwnPropNames = Object.getOwnPropertyNames;
-var __esm = (fn, res) => function __init() {
-  return fn && (res = (0, fn[__getOwnPropNames(fn)[0]])(fn = 0)), res;
+var __esm = (fn, res, err) => function __init() {
+  if (err) throw err[0];
+  try {
+    return fn && (res = (0, fn[__getOwnPropNames(fn)[0]])(fn = 0)), res;
+  } catch (e) {
+    throw err = [e], e;
+  }
 };
 var __export = (target, all) => {
   for (var name in all)

package/dist/index.cjs

@@ -308,14 +308,6 @@ function compareZip(a, b) {
   if (a > b) return 1;
   return 0;
 }
-function sortZipMatches(matches, exactZip) {
-  matches.sort((a, b) => {
-    if (a.zipCode === exactZip && b.zipCode !== exactZip) return -1;
-    if (b.zipCode === exactZip && a.zipCode !== exactZip) return 1;
-    return compareZip(a.zipCode, b.zipCode);
-  });
-  return matches;
-}
 function normalizeLimit(value) {
   if (Number.isNaN(value) || value === -Infinity) return 0;
   if (value === Infinity) return value;
@@ -323,9 +315,11 @@ function normalizeLimit(value) {
 }
 function lookupByZipCode(index, zip, options) {
   if (!index || !index.zipIndex || !index.records || typeof zip !== "string") return [];
+  if (zip.length > 1e3) return [];
+  const zipLimit = normalizeLimit(options?.zipLimit ?? Infinity);
+  if (zipLimit === 0) return [];
   const normalized = zip.trim();
   if (!ZIP_CODE_RE.test(normalized) || normalized.length < 2) return [];
-  const zipLimit = normalizeLimit(options?.zipLimit ?? Infinity);
   const matches = [];
   const keys = index.sortedZipKeys;
   const postings = index.sortedZipPostings;
@@ -337,17 +331,19 @@ function lookupByZipCode(index, zip, options) {
       if (keys[mid] < normalized) lo = mid + 1;
       else hi = mid;
     }
-    for (let i = lo; i < keys.length && keys[i].startsWith(normalized); i++) {
+    for (let i = lo; i < keys.length && keys[i].startsWith(normalized) && matches.length < zipLimit; i++) {
       const indices = postings[i];
-      for (let j = 0; j < indices.length; j++) matches.push(index.records[indices[j]]);
+      for (let j = 0; j < indices.length && matches.length < zipLimit; j++) matches.push(index.records[indices[j]]);
     }
   } else {
-    for (const [z, indices] of index.zipIndex) {
-      if (z.startsWith(normalized)) {
-        for (const idx of indices) matches.push(index.records[idx]);
+    const matchingKeys = [...index.zipIndex.keys()].filter((z) => z.startsWith(normalized)).sort(compareZip);
+    for (const z of matchingKeys) {
+      if (matches.length >= zipLimit) break;
+      for (const idx of index.zipIndex.get(z)) {
+        if (matches.length >= zipLimit) break;
+        matches.push(index.records[idx]);
       }
     }
-    sortZipMatches(matches, normalized);
   }
   return matches.slice(0, zipLimit).map((record) => ({ ...record }));
 }
@@ -389,6 +385,7 @@ function searchThaiAddress(index, query, options) {
     });
   }
   if (normalized.length < 3) return [];
+  if (limit === 0) return [];
   const searchText = options?.romanizationAliases !== false && HAS_LATIN_RE.test(normalized) ? applyRomanizationAliases(normalized) : normalized;
   const queryTrigrams = extractTrigramsNormalized(searchText);
   const { counts, touched } = getScratch(index);
@@ -415,7 +412,11 @@ function searchThaiAddress(index, query, options) {
     if (b.score !== a.score) return b.score - a.score;
     return b.matchRank - a.matchRank;
   });
-  const windowSize = Math.max(limit * 4, 50);
+  let windowSize = Math.min(limit, scored.length);
+  const boundary = scored[windowSize - 1];
+  while (windowSize < scored.length && scored[windowSize].score === boundary.score && scored[windowSize].matchRank === boundary.matchRank) {
+    windowSize++;
+  }
   const window = scored.slice(0, windowSize);
   window.sort((a, b) => {
     if (b.score !== a.score) return b.score - a.score;

package/dist/index.d.cts

@@ -184,7 +184,8 @@ declare function buildThaiAddressIndex(data: RawData, options?: BuildIndexOption
  * prefix range — O(log n_zips + matches). Ascending key order places an
  * exact match before any longer prefix match automatically, so no explicit
  * exact-first sort is needed. Indexes without the sorted arrays (older
- * shape, hand-built) fall back to the original O(n_zips) scan + sort.
+ * shape, hand-built) scan and sort matching keys before collecting records.
+ * Both paths stop collecting when `zipLimit` is reached.
  *
  * Returned records are shallow copies: mutating them cannot corrupt the
  * shared index (see `searchThaiAddress`).
@@ -193,8 +194,8 @@ declare function buildThaiAddressIndex(data: RawData, options?: BuildIndexOption
  * only by `options.zipLimit` (defaults to `Infinity`), since a zip code can
  * legitimately map to dozens of tambons (e.g. `45000` → 33 records).
  *
- * Malformed runtime input (non-string `zip`, missing index shape) returns
- * `[]` rather than crashing deep inside the library.
+ * Malformed runtime input (non-string `zip`, missing index shape) and raw
+ * strings longer than 1000 characters return `[]`.
  */
 declare function lookupByZipCode(index: TrigramIndex, zip: string, options?: SearchOptions): ThaiAddressRecord[];
 declare function searchThaiAddress(index: TrigramIndex, query: string, options?: SearchOptions): ThaiAddressRecord[];

package/dist/index.d.ts

@@ -184,7 +184,8 @@ declare function buildThaiAddressIndex(data: RawData, options?: BuildIndexOption
  * prefix range — O(log n_zips + matches). Ascending key order places an
  * exact match before any longer prefix match automatically, so no explicit
  * exact-first sort is needed. Indexes without the sorted arrays (older
- * shape, hand-built) fall back to the original O(n_zips) scan + sort.
+ * shape, hand-built) scan and sort matching keys before collecting records.
+ * Both paths stop collecting when `zipLimit` is reached.
  *
  * Returned records are shallow copies: mutating them cannot corrupt the
  * shared index (see `searchThaiAddress`).
@@ -193,8 +194,8 @@ declare function buildThaiAddressIndex(data: RawData, options?: BuildIndexOption
  * only by `options.zipLimit` (defaults to `Infinity`), since a zip code can
  * legitimately map to dozens of tambons (e.g. `45000` → 33 records).
  *
- * Malformed runtime input (non-string `zip`, missing index shape) returns
- * `[]` rather than crashing deep inside the library.
+ * Malformed runtime input (non-string `zip`, missing index shape) and raw
+ * strings longer than 1000 characters return `[]`.
  */
 declare function lookupByZipCode(index: TrigramIndex, zip: string, options?: SearchOptions): ThaiAddressRecord[];
 declare function searchThaiAddress(index: TrigramIndex, query: string, options?: SearchOptions): ThaiAddressRecord[];

package/dist/index.js

@@ -306,14 +306,6 @@ function compareZip(a, b) {
   if (a > b) return 1;
   return 0;
 }
-function sortZipMatches(matches, exactZip) {
-  matches.sort((a, b) => {
-    if (a.zipCode === exactZip && b.zipCode !== exactZip) return -1;
-    if (b.zipCode === exactZip && a.zipCode !== exactZip) return 1;
-    return compareZip(a.zipCode, b.zipCode);
-  });
-  return matches;
-}
 function normalizeLimit(value) {
   if (Number.isNaN(value) || value === -Infinity) return 0;
   if (value === Infinity) return value;
@@ -321,9 +313,11 @@ function normalizeLimit(value) {
 }
 function lookupByZipCode(index, zip, options) {
   if (!index || !index.zipIndex || !index.records || typeof zip !== "string") return [];
+  if (zip.length > 1e3) return [];
+  const zipLimit = normalizeLimit(options?.zipLimit ?? Infinity);
+  if (zipLimit === 0) return [];
   const normalized = zip.trim();
   if (!ZIP_CODE_RE.test(normalized) || normalized.length < 2) return [];
-  const zipLimit = normalizeLimit(options?.zipLimit ?? Infinity);
   const matches = [];
   const keys = index.sortedZipKeys;
   const postings = index.sortedZipPostings;
@@ -335,17 +329,19 @@ function lookupByZipCode(index, zip, options) {
       if (keys[mid] < normalized) lo = mid + 1;
       else hi = mid;
     }
-    for (let i = lo; i < keys.length && keys[i].startsWith(normalized); i++) {
+    for (let i = lo; i < keys.length && keys[i].startsWith(normalized) && matches.length < zipLimit; i++) {
       const indices = postings[i];
-      for (let j = 0; j < indices.length; j++) matches.push(index.records[indices[j]]);
+      for (let j = 0; j < indices.length && matches.length < zipLimit; j++) matches.push(index.records[indices[j]]);
     }
   } else {
-    for (const [z, indices] of index.zipIndex) {
-      if (z.startsWith(normalized)) {
-        for (const idx of indices) matches.push(index.records[idx]);
+    const matchingKeys = [...index.zipIndex.keys()].filter((z) => z.startsWith(normalized)).sort(compareZip);
+    for (const z of matchingKeys) {
+      if (matches.length >= zipLimit) break;
+      for (const idx of index.zipIndex.get(z)) {
+        if (matches.length >= zipLimit) break;
+        matches.push(index.records[idx]);
       }
     }
-    sortZipMatches(matches, normalized);
   }
   return matches.slice(0, zipLimit).map((record) => ({ ...record }));
 }
@@ -387,6 +383,7 @@ function searchThaiAddress(index, query, options) {
     });
   }
   if (normalized.length < 3) return [];
+  if (limit === 0) return [];
   const searchText = options?.romanizationAliases !== false && HAS_LATIN_RE.test(normalized) ? applyRomanizationAliases(normalized) : normalized;
   const queryTrigrams = extractTrigramsNormalized(searchText);
   const { counts, touched } = getScratch(index);
@@ -413,7 +410,11 @@ function searchThaiAddress(index, query, options) {
     if (b.score !== a.score) return b.score - a.score;
     return b.matchRank - a.matchRank;
   });
-  const windowSize = Math.max(limit * 4, 50);
+  let windowSize = Math.min(limit, scored.length);
+  const boundary = scored[windowSize - 1];
+  while (windowSize < scored.length && scored[windowSize].score === boundary.score && scored[windowSize].matchRank === boundary.matchRank) {
+    windowSize++;
+  }
   const window = scored.slice(0, windowSize);
   window.sort((a, b) => {
     if (b.score !== a.score) return b.score - a.score;

package/dist/react.cjs

@@ -152,14 +152,6 @@ function compareZip(a, b) {
   if (a > b) return 1;
   return 0;
 }
-function sortZipMatches(matches, exactZip) {
-  matches.sort((a, b) => {
-    if (a.zipCode === exactZip && b.zipCode !== exactZip) return -1;
-    if (b.zipCode === exactZip && a.zipCode !== exactZip) return 1;
-    return compareZip(a.zipCode, b.zipCode);
-  });
-  return matches;
-}
 function normalizeLimit(value) {
   if (Number.isNaN(value) || value === -Infinity) return 0;
   if (value === Infinity) return value;
@@ -167,9 +159,11 @@ function normalizeLimit(value) {
 }
 function lookupByZipCode(index, zip, options) {
   if (!index || !index.zipIndex || !index.records || typeof zip !== "string") return [];
+  if (zip.length > 1e3) return [];
+  const zipLimit = normalizeLimit(options?.zipLimit ?? Infinity);
+  if (zipLimit === 0) return [];
   const normalized = zip.trim();
   if (!ZIP_CODE_RE.test(normalized) || normalized.length < 2) return [];
-  const zipLimit = normalizeLimit(options?.zipLimit ?? Infinity);
   const matches = [];
   const keys = index.sortedZipKeys;
   const postings = index.sortedZipPostings;
@@ -181,17 +175,19 @@ function lookupByZipCode(index, zip, options) {
       if (keys[mid] < normalized) lo = mid + 1;
       else hi = mid;
     }
-    for (let i = lo; i < keys.length && keys[i].startsWith(normalized); i++) {
+    for (let i = lo; i < keys.length && keys[i].startsWith(normalized) && matches.length < zipLimit; i++) {
       const indices = postings[i];
-      for (let j = 0; j < indices.length; j++) matches.push(index.records[indices[j]]);
+      for (let j = 0; j < indices.length && matches.length < zipLimit; j++) matches.push(index.records[indices[j]]);
     }
   } else {
-    for (const [z, indices] of index.zipIndex) {
-      if (z.startsWith(normalized)) {
-        for (const idx of indices) matches.push(index.records[idx]);
+    const matchingKeys = [...index.zipIndex.keys()].filter((z) => z.startsWith(normalized)).sort(compareZip);
+    for (const z of matchingKeys) {
+      if (matches.length >= zipLimit) break;
+      for (const idx of index.zipIndex.get(z)) {
+        if (matches.length >= zipLimit) break;
+        matches.push(index.records[idx]);
       }
     }
-    sortZipMatches(matches, normalized);
   }
   return matches.slice(0, zipLimit).map((record) => ({ ...record }));
 }
@@ -233,6 +229,7 @@ function searchThaiAddress(index, query, options) {
     });
   }
   if (normalized.length < 3) return [];
+  if (limit === 0) return [];
   const searchText = options?.romanizationAliases !== false && HAS_LATIN_RE.test(normalized) ? applyRomanizationAliases(normalized) : normalized;
   const queryTrigrams = extractTrigramsNormalized(searchText);
   const { counts, touched } = getScratch(index);
@@ -259,7 +256,11 @@ function searchThaiAddress(index, query, options) {
     if (b.score !== a.score) return b.score - a.score;
     return b.matchRank - a.matchRank;
   });
-  const windowSize = Math.max(limit * 4, 50);
+  let windowSize = Math.min(limit, scored.length);
+  const boundary = scored[windowSize - 1];
+  while (windowSize < scored.length && scored[windowSize].score === boundary.score && scored[windowSize].matchRank === boundary.matchRank) {
+    windowSize++;
+  }
   const window = scored.slice(0, windowSize);
   window.sort((a, b) => {
     if (b.score !== a.score) return b.score - a.score;

package/dist/react.js

@@ -150,14 +150,6 @@ function compareZip(a, b) {
   if (a > b) return 1;
   return 0;
 }
-function sortZipMatches(matches, exactZip) {
-  matches.sort((a, b) => {
-    if (a.zipCode === exactZip && b.zipCode !== exactZip) return -1;
-    if (b.zipCode === exactZip && a.zipCode !== exactZip) return 1;
-    return compareZip(a.zipCode, b.zipCode);
-  });
-  return matches;
-}
 function normalizeLimit(value) {
   if (Number.isNaN(value) || value === -Infinity) return 0;
   if (value === Infinity) return value;
@@ -165,9 +157,11 @@ function normalizeLimit(value) {
 }
 function lookupByZipCode(index, zip, options) {
   if (!index || !index.zipIndex || !index.records || typeof zip !== "string") return [];
+  if (zip.length > 1e3) return [];
+  const zipLimit = normalizeLimit(options?.zipLimit ?? Infinity);
+  if (zipLimit === 0) return [];
   const normalized = zip.trim();
   if (!ZIP_CODE_RE.test(normalized) || normalized.length < 2) return [];
-  const zipLimit = normalizeLimit(options?.zipLimit ?? Infinity);
   const matches = [];
   const keys = index.sortedZipKeys;
   const postings = index.sortedZipPostings;
@@ -179,17 +173,19 @@ function lookupByZipCode(index, zip, options) {
       if (keys[mid] < normalized) lo = mid + 1;
       else hi = mid;
     }
-    for (let i = lo; i < keys.length && keys[i].startsWith(normalized); i++) {
+    for (let i = lo; i < keys.length && keys[i].startsWith(normalized) && matches.length < zipLimit; i++) {
       const indices = postings[i];
-      for (let j = 0; j < indices.length; j++) matches.push(index.records[indices[j]]);
+      for (let j = 0; j < indices.length && matches.length < zipLimit; j++) matches.push(index.records[indices[j]]);
     }
   } else {
-    for (const [z, indices] of index.zipIndex) {
-      if (z.startsWith(normalized)) {
-        for (const idx of indices) matches.push(index.records[idx]);
+    const matchingKeys = [...index.zipIndex.keys()].filter((z) => z.startsWith(normalized)).sort(compareZip);
+    for (const z of matchingKeys) {
+      if (matches.length >= zipLimit) break;
+      for (const idx of index.zipIndex.get(z)) {
+        if (matches.length >= zipLimit) break;
+        matches.push(index.records[idx]);
       }
     }
-    sortZipMatches(matches, normalized);
   }
   return matches.slice(0, zipLimit).map((record) => ({ ...record }));
 }
@@ -231,6 +227,7 @@ function searchThaiAddress(index, query, options) {
     });
   }
   if (normalized.length < 3) return [];
+  if (limit === 0) return [];
   const searchText = options?.romanizationAliases !== false && HAS_LATIN_RE.test(normalized) ? applyRomanizationAliases(normalized) : normalized;
   const queryTrigrams = extractTrigramsNormalized(searchText);
   const { counts, touched } = getScratch(index);
@@ -257,7 +254,11 @@ function searchThaiAddress(index, query, options) {
     if (b.score !== a.score) return b.score - a.score;
     return b.matchRank - a.matchRank;
   });
-  const windowSize = Math.max(limit * 4, 50);
+  let windowSize = Math.min(limit, scored.length);
+  const boundary = scored[windowSize - 1];
+  while (windowSize < scored.length && scored[windowSize].score === boundary.score && scored[windowSize].matchRank === boundary.matchRank) {
+    windowSize++;
+  }
   const window = scored.slice(0, windowSize);
   window.sort((a, b) => {
     if (b.score !== a.score) return b.score - a.score;

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "thaizip",
-  "version": "0.7.3",
+  "version": "0.7.4",
   "description": "Fast, fuzzy Thai address autocomplete library (headless + React)",
   "keywords": [
     "thai",
@@ -58,14 +58,17 @@
     "@testing-library/react": "^16.3.2",
     "@types/node": "^25.5.2",
     "@types/react": "^19.2.14",
-    "@vitest/coverage-v8": "^4.1.2",
+    "@vitest/coverage-v8": "^4.1.11",
     "jsdom": "^29.0.1",
     "react": "^19.2.4",
     "react-dom": "^19.2.4",
     "tsup": "^8.5.1",
-    "tsx": "^4.21.0",
+    "tsx": "^4.23.13",
     "typescript": "^6.0.2",
-    "vitest": "^4.1.2"
+    "vitest": "^4.1.11"
+  },
+  "overrides": {
+    "esbuild": "^0.28.1"
   },
   "peerDependencies": {
     "react": ">=18.0.0"