diff --git a/std/assembly/util/string.ts b/std/assembly/util/string.ts
index 76e993ec96..e812a38c48 100644
--- a/std/assembly/util/string.ts
+++ b/std/assembly/util/string.ts
@@ -4,13 +4,13 @@ import { ipow32 } from "../math";
 // All tables are stored as two staged lookup tables (static tries)
 // because the full range of Unicode symbols can't be efficiently
 // represented as-is in memory (see Unicode spec ch 5, p.196):
-// https://www.unicode.org/versions/Unicode12.0.0/ch05.pdf
+// https://www.unicode.org/versions/Unicode13.0.0/ch05.pdf
 // Tables have been generated using these forked musl tools:
 // https://github.com/MaxGraey/musl-chartable-tools/tree/case-ignorable
 
 // Lookup table to check if a character is alphanumeric or not
 // See: https://git.musl-libc.org/cgit/musl/tree/src/ctype/alpha.h
-// size: 3904 bytes
+// size: 4032 bytes (compressed to ~3500 after binaryen)
 // @ts-ignore
 @inline @lazy const ALPHA_TABLE = memory.data<u8>([
   18,17,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,17,34,35,36,17,37,38,39,40,
@@ -191,7 +191,7 @@ import { ipow32 } from "../math";
   0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,255,3
 ]);
 
-// size: 1568 bytes (compressed to ~1380 bytes after binaryen)
+// size: 1568 bytes (compressed to ~1300 bytes after binaryen)
 // @ts-ignore: decorator
 @lazy @inline const CASED = memory.data<u8>([
   18,19,20,21,22,23,16,16,16,16,16,16,16,16,16,16,
@@ -274,7 +274,7 @@ import { ipow32 } from "../math";
   0,0,0,0,0,0,0,0
 ]);
 
-// size: 2976 bytes (compressed to ~2050 bytes after binaryen)
+// size: 2976 bytes (compressed to ~2000 bytes after binaryen)
 // @ts-ignore: decorator
 @lazy @inline const CASE_IGNORABLES = memory.data<u8>([
   18,16,19,20,21,22,23,24,25,26,27,28,29,30,31,32,
@@ -983,25 +983,25 @@ export function joinStringArray(dataStart: usize, length: i32, separator: string
     // @ts-ignore: type
     if (value !== null) estLen += value.length;
   }
-  var offset = 0;
-  var sepLen = separator.length;
-  var result = __alloc((estLen + sepLen * lastIndex) << 1, idof<string>());
+  var offset: usize = 0;
+  var sepLen = <usize>separator.length << 1;
+  var result = __alloc((estLen << 1) + sepLen * lastIndex, idof<string>());
   for (let i = 0; i < lastIndex; ++i) {
     value = load<string>(dataStart + (<usize>i << alignof<string>()));
     if (value !== null) {
-      let valueLen = value.length;
+      let valueLen = <usize>value.length << 1;
       memory.copy(
-        result + (<usize>offset << 1),
+        result + offset,
         changetype<usize>(value),
-        <usize>valueLen << 1
+        valueLen
       );
       offset += valueLen;
     }
     if (sepLen) {
       memory.copy(
-        result + (<usize>offset << 1),
+        result + offset,
         changetype<usize>(separator),
-        <usize>sepLen << 1
+        sepLen
       );
       offset += sepLen;
     }
@@ -1009,7 +1009,7 @@ export function joinStringArray(dataStart: usize, length: i32, separator: string
   value = load<string>(dataStart + (<usize>lastIndex << alignof<string>()));
   if (value !== null) {
     memory.copy(
-      result + (<usize>offset << 1),
+      result + offset,
       changetype<usize>(value),
       <usize>value.length << 1
     );
@@ -1017,6 +1017,38 @@ export function joinStringArray(dataStart: usize, length: i32, separator: string
   return changetype<string>(result); // retains
 }
 
+export function concat3(a: string, b: string, c: string): string {
+  var bytesLenA = <usize>a.length << 1;
+  var bytesLenB = <usize>b.length << 1;
+  var bytesLenC = <usize>c.length << 1;
+  var result = __alloc(bytesLenA + bytesLenB + bytesLenC, idof<string>());
+  var offset: usize = 0;
+  if (bytesLenA) {
+    memory.copy(
+      result,
+      changetype<usize>(a),
+      bytesLenA
+    );
+    offset += bytesLenA;
+  }
+  if (bytesLenB) {
+    memory.copy(
+      result + offset,
+      changetype<usize>(b),
+      bytesLenB
+    );
+    offset += bytesLenB;
+  }
+  if (bytesLenC) {
+    memory.copy(
+      result + offset,
+      changetype<usize>(c),
+      bytesLenC
+    );
+  }
+  return changetype<string>(result); // retains
+}
+
 export function joinReferenceArray<T>(dataStart: usize, length: i32, separator: string): string {
   var lastIndex = length - 1;
   if (lastIndex < 0) return "";
diff --git a/tests/compiler/std/array.untouched.wat b/tests/compiler/std/array.untouched.wat
index 347bea9cff..7c88527e6a 100644
--- a/tests/compiler/std/array.untouched.wat
+++ b/tests/compiler/std/array.untouched.wat
@@ -16072,15 +16072,17 @@
   local.set $10
   local.get $2
   call $~lib/string/String#get:length
+  i32.const 1
+  i32.shl
   local.set $11
   local.get $5
+  i32.const 1
+  i32.shl
   local.get $11
   local.get $3
   i32.mul
   i32.add
   i32.const 1
-  i32.shl
-  i32.const 1
   call $~lib/rt/tlsf/__alloc
   local.set $12
   i32.const 0
@@ -16117,16 +16119,14 @@
     if
      local.get $6
      call $~lib/string/String#get:length
+     i32.const 1
+     i32.shl
      local.set $9
      local.get $12
      local.get $10
-     i32.const 1
-     i32.shl
      i32.add
      local.get $6
      local.get $9
-     i32.const 1
-     i32.shl
      call $~lib/memory/memory.copy
      local.get $10
      local.get $9
@@ -16137,13 +16137,9 @@
     if
      local.get $12
      local.get $10
-     i32.const 1
-     i32.shl
      i32.add
      local.get $2
      local.get $11
-     i32.const 1
-     i32.shl
      call $~lib/memory/memory.copy
      local.get $10
      local.get $11
@@ -16182,8 +16178,6 @@
   if
    local.get $12
    local.get $10
-   i32.const 1
-   i32.shl
    i32.add
    local.get $6
    local.get $6