fix(compiler): Properly encode hex values in Bytes literals (#2088)

grain-lang · Apr 6, 2024 · 263993d · 263993d
1 parent 8d157f3
commit 263993d
Show file tree

Hide file tree

Showing 2 changed files with 8 additions and 1 deletion.
diff --git a/compiler/src/utils/literals.re b/compiler/src/utils/literals.re
@@ -385,8 +385,10 @@ let add_code_point = (buf, str, unicode) => {
     | "x" => Scanf.sscanf(numstr, "%x", x => x)
     | _ => Scanf.sscanf(esc ++ numstr, "%o", x => x)
     };
-  if (Uchar.is_valid(code_point)) {
+  if (unicode && Uchar.is_valid(code_point)) {
     Buffer.add_utf_8_uchar(buf, Uchar.of_int(code_point));
+  } else if (!unicode) {
+    Buffer.add_uint8(buf, code_point);
   } else {
     raise(IllegalUnicodeCodePoint(str));
   };

diff --git a/compiler/test/suites/strings.re b/compiler/test/suites/strings.re
@@ -376,6 +376,11 @@ bar", 1))|},
 
   // Bytes literals
   assertRun("bytes_literal", {|print(b"abc")|}, "<bytes: 61 62 63>\n");
+  assertRun(
+    "bytes_literal_hex",
+    {|print(b"\xc3\x81\x24\x24\x00\x24\x99\xc3")|},
+    "<bytes: c3 81 24 24 00 24 99 c3>\n",
+  );
   assertRun(
     "bytes_literal_long",
     {|print(b"ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefg")|},