Correctly convert surrogate UTF-16 to rune on MS-Windows

internal/win32/util.go

@@ -15,26 +15,13 @@ func hiword(x uint32) uint16 {
 	return uint16((x >> 16) & 0xFFFF)
 }
 
-func decodeUtf16(s uint16) rune {
-	const (
-		// 0xd800-0xdc00 encodes the high 10 bits of a pair.
-		// 0xdc00-0xe000 encodes the low 10 bits of a pair.
-		// the value is those 20 bits plus 0x10000.
-		surr1 = 0xd800
-		surr3 = 0xe000
-
-		// Unicode replacement character
-		replacementChar = '\uFFFD'
-	)
+func isSurrogatedCharacter(x rune) bool {
+	return x > 0xd800 // Surrogate characters are mounted after 0xd800
+}
 
-	var a rune
-	switch r := s; {
-	case r < surr1, surr3 <= r:
-		// normal rune
-		a = rune(r)
-	default:
-		// invalid surrogate sequence
-		a = replacementChar
-	}
-	return a
+// surrogatedUtf16toRune recovers code points from high and low surrogates
+func surrogatedUtf16toRune(high rune, low rune) rune {
+	high -= 0xd800
+	low -= 0xdc00
+	return (high << 10) + low + 0x10000
 }

internal/win32/window.go

@@ -37,6 +37,8 @@ type Window struct {
 	cursorCaptureCount int                         // non-shared
 	modifiers          events.ModifiersState       // non-shared
 
+	highSurrogated rune // non-shared
+
 	// callbacks
 	resizedCb           atomicx.Pointer[events.WindowResizedCallback]
 	closeRequestedCb    atomicx.Pointer[events.WindowCloseRequestedCallback]
@@ -809,11 +811,21 @@ func windowProc(window, msg, wparam, lparam uintptr) uintptr {
 			}
 		}
 		return 0
-
 	case procs.WM_CHAR, procs.WM_SYSCHAR:
+		// Most UTF16 without surrogates can simply be considered rune.
+		ch := rune(wparam)
+		// The surrogated UTF16 character is POSTed as two consecutive WM_CHARs: high surrogate and low surrogate.
+		if isSurrogatedCharacter(ch) {
+			if w.highSurrogated == 0 {
+				w.highSurrogated = ch
+				return 0
+			}
+			ch = surrogatedUtf16toRune(w.highSurrogated, ch)
+			w.highSurrogated = 0
+		}
 		if cb := w.receivedCharacterCb.Load(); cb != nil {
 			if cb := (*cb); cb != nil {
-				cb(decodeUtf16(uint16(wparam)))
+				cb(ch)
 			}
 		}
 		return 0