Respect UTF-16 positions in LSP word lookup

Author: mgomes

cmd/vibes/lsp.go

@@ -364,13 +364,15 @@ func wordAtPosition(source string, line, character int) string {
 		return ""
 	}
 
-	runes := []rune(lines[line])
+	lineText := lines[line]
+	runes := []rune(lineText)
 	if len(runes) == 0 {
 		return ""
 	}
 	if character < 0 {
 		character = 0
 	}
+	character = utf16OffsetToRuneIndex(lineText, character)
 	if character > len(runes) {
 		character = len(runes)
 	}
@@ -401,6 +403,26 @@ func wordAtPosition(source string, line, character int) string {
 	return string(runes[start:end])
 }
 
+func utf16OffsetToRuneIndex(text string, utf16Offset int) int {
+	if utf16Offset <= 0 {
+		return 0
+	}
+	runeIndex := 0
+	consumed := 0
+	for _, r := range text {
+		if consumed >= utf16Offset {
+			break
+		}
+		if r > 0xFFFF {
+			consumed += 2
+		} else {
+			consumed++
+		}
+		runeIndex++
+	}
+	return runeIndex
+}
+
 func isWordRune(r rune) bool {
 	return unicode.IsLetter(r) || unicode.IsDigit(r) || r == '_' || r == '?' || r == '!'
 }

cmd/vibes/lsp_test.go

@@ -186,6 +186,14 @@ func TestWordAtPosition(t *testing.T) {
 	}
 }
 
+func TestWordAtPositionUsesUTF16CharacterOffsets(t *testing.T) {
+	source := "😀😀x y\n"
+	word := wordAtPosition(source, 0, 4)
+	if word != "x" {
+		t.Fatalf("expected x, got %q", word)
+	}
+}
+
 func rawID(value string) *json.RawMessage {
 	raw := json.RawMessage(value)
 	return &raw