emscripten-core · PtPrashantTripathi · Sep 5, 2025
diff --git a/AUTHORS b/AUTHORS
@@ -601,3 +601,4 @@ a license to everyone to use it as detailed in LICENSE.)
 * Artur Gatin <agatin@teladochealth.com> (copyright owned by Teladoc Health, Inc.)
 * Christian Lloyd <clloyd@teladochealth.com> (copyright owned by Teladoc Health, Inc.)
 * Sean Morris <sean@seanmorr.is>
+* Pt. Prashant Tripathi <ptprashanttripathi@outlook.com>
diff --git a/site/source/docs/tools_reference/settings_reference.rst b/site/source/docs/tools_reference/settings_reference.rst
@@ -2700,6 +2700,18 @@ of ENVIRONMENT since TextDecoder is not available in those environments).
 
 Default value: 1
 
+.. _textencoder:
+
+TEXTENCODER
+===========
+
+The default value of 1 means the generated code will use TextEncoder if
+available and fall back to custom encoding code when it is not available.
+If set to 2, we assume TextEncoder is always present and usable, and no
+fallback JS code will be emitted.
+
+Default value: 1
+
 .. _embind_std_string_is_utf8:
 
 EMBIND_STD_STRING_IS_UTF8

diff --git a/src/lib/libstrings.js b/src/lib/libstrings.js
@@ -8,6 +8,10 @@
 #error "TEXTDECODER must be either 1 or 2"
 #endif
 
+#if TEXTENCODER != 1 && TEXTENCODER != 2
+#error "TEXTENCODER must be either 1 or 2"
+#endif
+
 addToLibrary({
   // TextDecoder constructor defaults to UTF-8
 #if TEXTDECODER == 2
@@ -16,6 +20,13 @@ addToLibrary({
   $UTF8Decoder: "typeof TextDecoder != 'undefined' ? new TextDecoder() : undefined",
 #endif
 
+  // TextEncoder constructor defaults to UTF-8
+#if TEXTENCODER == 2
+  $UTF8Encoder: "new TextEncoder()",
+#else
+  $UTF8Encoder: "typeof TextEncoder != 'undefined' ? new TextEncoder() : undefined",
+#endif
+
   $findStringEnd: (heapOrArray, idx, maxBytesToRead, ignoreNul) => {
     var maxIdx = idx + maxBytesToRead;
     if (ignoreNul) return maxIdx;
@@ -147,9 +158,12 @@ addToLibrary({
    *                                   terminator.
    * @return {number} The number of bytes written, EXCLUDING the null terminator.
    */
+  $stringToUTF8Array__deps: [
+    '$UTF8Encoder',
 #if ASSERTIONS
-  $stringToUTF8Array__deps: ['$warnOnce'],
+    '$warnOnce',
 #endif
+  ],
   $stringToUTF8Array: (str, heap, outIdx, maxBytesToWrite) => {
 #if CAN_ADDRESS_2GB
     outIdx >>>= 0;
@@ -162,6 +176,28 @@ addToLibrary({
     if (!(maxBytesToWrite > 0))
       return 0;
 
+#if TEXTENCODER == 2
+    // Always use TextEncoder when TEXTENCODER == 2
+    var encoded = UTF8Encoder.encode(str);
+    var bytesToWrite = Math.min(encoded.length, maxBytesToWrite - 1); // -1 for null terminator
+    for (var i = 0; i < bytesToWrite; ++i) {
+      heap[outIdx + i] = encoded[i];
+    }
+    heap[outIdx + bytesToWrite] = 0;
+    return bytesToWrite;
+#else
+    // When using conditional TextEncoder
+    if (UTF8Encoder) {
+      var encoded = UTF8Encoder.encode(str);
+      var bytesToWrite = Math.min(encoded.length, maxBytesToWrite - 1); // -1 for null terminator
+      for (var i = 0; i < bytesToWrite; ++i) {
+        heap[outIdx + i] = encoded[i];
+      }
+      heap[outIdx + bytesToWrite] = 0;
+      return bytesToWrite;
+    }
+
+    // Fallback: manual UTF-8 encoding
     var startIdx = outIdx;
     var endIdx = outIdx + maxBytesToWrite - 1; // -1 for string null terminator.
     for (var i = 0; i < str.length; ++i) {
@@ -198,6 +234,7 @@ addToLibrary({
     // Null-terminate the pointer to the buffer.
     heap[outIdx] = 0;
     return outIdx - startIdx;
+#endif // TEXTENCODER == 2
   },
 
   /**
@@ -224,7 +261,18 @@ addToLibrary({
    * @param {string} str - JavaScript string to operator on
    * @return {number} Length, in bytes, of the UTF8 encoded string.
    */
+  $lengthBytesUTF8__deps: ['$UTF8Encoder'],
   $lengthBytesUTF8: (str) => {
+#if TEXTENCODER == 2
+    // Always use TextEncoder when TEXTENCODER == 2
+    return UTF8Encoder.encode(str).length;
+#else
+    // When using conditional TextEncoder
+    if (UTF8Encoder) {
+      return UTF8Encoder.encode(str).length;
+    }
+
+    // Fallback: manual calculation
     var len = 0;
     for (var i = 0; i < str.length; ++i) {
       // Gotcha: charCodeAt returns a 16-bit word that is a UTF-16 encoded code
@@ -243,6 +291,7 @@ addToLibrary({
       }
     }
     return len;
+#endif // TEXTENCODER == 2
   },
 
   $intArrayFromString__docs: '/** @type {function(string, boolean=, number=)} */',

diff --git a/src/settings.js b/src/settings.js
@@ -1776,6 +1776,13 @@ var EVAL_CTORS = 0;
 // [link]
 var TEXTDECODER = 1;
 
+// The default value of 1 means the generated code will use TextEncoder if
+// available and fall back to custom encoding code when it is not available.
+// If set to 2, we assume TextEncoder is always present and usable, and no
+// fallback JS code will be emitted.
+// [link]
+var TEXTENCODER = 1;
+
 // Embind specific: If enabled, assume UTF-8 encoded data in std::string binding.
 // Disable this to support binary data transfer.
 // [link]

diff --git a/test/code_size/test_codesize_cxx_ctors1.json b/test/code_size/test_codesize_cxx_ctors1.json
@@ -1,10 +1,10 @@
 {
-  "a.out.js": 19754,
-  "a.out.js.gz": 8162,
-  "a.out.nodebug.wasm": 129509,
-  "a.out.nodebug.wasm.gz": 49243,
-  "total": 149263,
-  "total_gz": 57405,
+  "a.out.js": 19941,
+  "a.out.js.gz": 8224,
+  "a.out.nodebug.wasm": 129513,
+  "a.out.nodebug.wasm.gz": 49252,
+  "total": 149454,
+  "total_gz": 57476,
   "sent": [
     "__cxa_throw",
     "_abort_js",

diff --git a/test/code_size/test_codesize_cxx_ctors2.json b/test/code_size/test_codesize_cxx_ctors2.json
@@ -1,10 +1,10 @@
 {
-  "a.out.js": 19732,
-  "a.out.js.gz": 8148,
-  "a.out.nodebug.wasm": 128936,
-  "a.out.nodebug.wasm.gz": 48884,
-  "total": 148668,
-  "total_gz": 57032,
+  "a.out.js": 19919,
+  "a.out.js.gz": 8209,
+  "a.out.nodebug.wasm": 128940,
+  "a.out.nodebug.wasm.gz": 48891,
+  "total": 148859,
+  "total_gz": 57100,
   "sent": [
     "__cxa_throw",
     "_abort_js",

diff --git a/test/code_size/test_codesize_cxx_except.json b/test/code_size/test_codesize_cxx_except.json
@@ -1,10 +1,10 @@
 {
-  "a.out.js": 23415,
-  "a.out.js.gz": 9145,
-  "a.out.nodebug.wasm": 171271,
-  "a.out.nodebug.wasm.gz": 57338,
-  "total": 194686,
-  "total_gz": 66483,
+  "a.out.js": 23604,
+  "a.out.js.gz": 9203,
+  "a.out.nodebug.wasm": 171275,
+  "a.out.nodebug.wasm.gz": 57346,
+  "total": 194879,
+  "total_gz": 66549,
   "sent": [
     "__cxa_begin_catch",
     "__cxa_end_catch",

diff --git a/test/code_size/test_codesize_cxx_except_wasm.json b/test/code_size/test_codesize_cxx_except_wasm.json
@@ -1,10 +1,10 @@
 {
-  "a.out.js": 19643,
-  "a.out.js.gz": 8112,
-  "a.out.nodebug.wasm": 144630,
-  "a.out.nodebug.wasm.gz": 54894,
-  "total": 164273,
-  "total_gz": 63006,
+  "a.out.js": 19831,
+  "a.out.js.gz": 8171,
+  "a.out.nodebug.wasm": 144634,
+  "a.out.nodebug.wasm.gz": 54899,
+  "total": 164465,
+  "total_gz": 63070,
   "sent": [
     "_abort_js",
     "_tzset_js",

diff --git a/test/code_size/test_codesize_cxx_except_wasm_legacy.json b/test/code_size/test_codesize_cxx_except_wasm_legacy.json
@@ -1,10 +1,10 @@
 {
-  "a.out.js": 19643,
-  "a.out.js.gz": 8112,
-  "a.out.nodebug.wasm": 142219,
-  "a.out.nodebug.wasm.gz": 54358,
-  "total": 161862,
-  "total_gz": 62470,
+  "a.out.js": 19831,
+  "a.out.js.gz": 8171,
+  "a.out.nodebug.wasm": 142223,
+  "a.out.nodebug.wasm.gz": 54362,
+  "total": 162054,
+  "total_gz": 62533,
   "sent": [
     "_abort_js",
     "_tzset_js",

diff --git a/test/code_size/test_codesize_cxx_lto.json b/test/code_size/test_codesize_cxx_lto.json
@@ -1,10 +1,10 @@
 {
-  "a.out.js": 19082,
-  "a.out.js.gz": 7841,
-  "a.out.nodebug.wasm": 106463,
-  "a.out.nodebug.wasm.gz": 42596,
-  "total": 125545,
-  "total_gz": 50437,
+  "a.out.js": 18725,
+  "a.out.js.gz": 7678,
+  "a.out.nodebug.wasm": 106473,
+  "a.out.nodebug.wasm.gz": 42629,
+  "total": 125198,
+  "total_gz": 50307,
   "sent": [
     "a (emscripten_resize_heap)",
     "b (_setitimer_js)",

diff --git a/test/code_size/test_codesize_cxx_mangle.json b/test/code_size/test_codesize_cxx_mangle.json
@@ -1,10 +1,10 @@
 {
-  "a.out.js": 23465,
-  "a.out.js.gz": 9164,
-  "a.out.nodebug.wasm": 235312,
-  "a.out.nodebug.wasm.gz": 78933,
-  "total": 258777,
-  "total_gz": 88097,
+  "a.out.js": 23654,
+  "a.out.js.gz": 9223,
+  "a.out.nodebug.wasm": 235316,
+  "a.out.nodebug.wasm.gz": 78938,
+  "total": 258970,
+  "total_gz": 88161,
   "sent": [
     "__cxa_begin_catch",
     "__cxa_end_catch",

diff --git a/test/code_size/test_codesize_cxx_noexcept.json b/test/code_size/test_codesize_cxx_noexcept.json
@@ -1,10 +1,10 @@
 {
-  "a.out.js": 19754,
-  "a.out.js.gz": 8162,
-  "a.out.nodebug.wasm": 131926,
-  "a.out.nodebug.wasm.gz": 50238,
-  "total": 151680,
-  "total_gz": 58400,
+  "a.out.js": 19941,
+  "a.out.js.gz": 8224,
+  "a.out.nodebug.wasm": 131930,
+  "a.out.nodebug.wasm.gz": 50245,
+  "total": 151871,
+  "total_gz": 58469,
   "sent": [
     "__cxa_throw",
     "_abort_js",

diff --git a/test/code_size/test_codesize_cxx_wasmfs.json b/test/code_size/test_codesize_cxx_wasmfs.json
@@ -1,10 +1,10 @@
 {
-  "a.out.js": 7143,
-  "a.out.js.gz": 3338,
-  "a.out.nodebug.wasm": 169797,
-  "a.out.nodebug.wasm.gz": 63087,
-  "total": 176940,
-  "total_gz": 66425,
+  "a.out.js": 7353,
+  "a.out.js.gz": 3407,
+  "a.out.nodebug.wasm": 169801,
+  "a.out.nodebug.wasm.gz": 63095,
+  "total": 177154,
+  "total_gz": 66502,
   "sent": [
     "__cxa_throw",
     "_abort_js",

diff --git a/test/code_size/test_codesize_file_preload.json b/test/code_size/test_codesize_file_preload.json
@@ -1,10 +1,10 @@
 {
-  "a.out.js": 22684,
-  "a.out.js.gz": 9375,
+  "a.out.js": 22869,
+  "a.out.js.gz": 9431,
   "a.out.nodebug.wasm": 1681,
   "a.out.nodebug.wasm.gz": 960,
-  "total": 24365,
-  "total_gz": 10335,
+  "total": 24550,
+  "total_gz": 10391,
   "sent": [
     "a (fd_write)"
   ],

diff --git a/test/code_size/test_codesize_files_js_fs.json b/test/code_size/test_codesize_files_js_fs.json
@@ -1,10 +1,10 @@
 {
-  "a.out.js": 18357,
-  "a.out.js.gz": 7470,
+  "a.out.js": 18536,
+  "a.out.js.gz": 7528,
   "a.out.nodebug.wasm": 381,
   "a.out.nodebug.wasm.gz": 260,
-  "total": 18738,
-  "total_gz": 7730,
+  "total": 18917,
+  "total_gz": 7788,
   "sent": [
     "a (fd_write)",
     "b (fd_read)",

diff --git a/test/code_size/test_codesize_files_wasmfs.json b/test/code_size/test_codesize_files_wasmfs.json
@@ -1,10 +1,10 @@
 {
-  "a.out.js": 5549,
-  "a.out.js.gz": 2591,
-  "a.out.nodebug.wasm": 50231,
+  "a.out.js": 5759,
+  "a.out.js.gz": 2665,
+  "a.out.nodebug.wasm": 50237,
   "a.out.nodebug.wasm.gz": 18094,
-  "total": 55780,
-  "total_gz": 20685,
+  "total": 55996,
+  "total_gz": 20759,
   "sent": [
     "a (emscripten_date_now)",
     "b (emscripten_err)",

diff --git a/test/code_size/test_codesize_hello_O0.json b/test/code_size/test_codesize_hello_O0.json
@@ -1,10 +1,10 @@
 {
-  "a.out.js": 22496,
-  "a.out.js.gz": 8322,
+  "a.out.js": 22508,
+  "a.out.js.gz": 8324,
   "a.out.nodebug.wasm": 15127,
   "a.out.nodebug.wasm.gz": 7450,
-  "total": 37623,
-  "total_gz": 15772,
+  "total": 37635,
+  "total_gz": 15774,
   "sent": [
     "fd_write"
   ],

diff --git a/test/code_size/test_codesize_hello_dylink.json b/test/code_size/test_codesize_hello_dylink.json
@@ -1,10 +1,10 @@
 {
-  "a.out.js": 26976,
-  "a.out.js.gz": 11458,
+  "a.out.js": 27164,
+  "a.out.js.gz": 11514,
   "a.out.nodebug.wasm": 18567,
   "a.out.nodebug.wasm.gz": 9199,
-  "total": 45543,
-  "total_gz": 20657,
+  "total": 45731,
+  "total_gz": 20713,
   "sent": [
     "__heap_base",
     "__indirect_function_table",

diff --git a/test/code_size/test_codesize_hello_dylink_all.json b/test/code_size/test_codesize_hello_dylink_all.json
@@ -1,7 +1,7 @@
 {
-  "a.out.js": 246331,
-  "a.out.nodebug.wasm": 597763,
-  "total": 844094,
+  "a.out.js": 246519,
+  "a.out.nodebug.wasm": 597778,
+  "total": 844297,
   "sent": [
     "IMG_Init",
     "IMG_Load",

diff --git a/test/code_size/test_codesize_mem_O3.json b/test/code_size/test_codesize_mem_O3.json
@@ -1,10 +1,10 @@
 {
-  "a.out.js": 4399,
-  "a.out.js.gz": 2112,
-  "a.out.nodebug.wasm": 5260,
-  "a.out.nodebug.wasm.gz": 2418,
-  "total": 9659,
-  "total_gz": 4530,
+  "a.out.js": 4581,
+  "a.out.js.gz": 2194,
+  "a.out.nodebug.wasm": 5267,
+  "a.out.nodebug.wasm.gz": 2422,
+  "total": 9848,
+  "total_gz": 4616,
   "sent": [
     "a (emscripten_resize_heap)"
   ],