From 36ece653f007ef6cd237d8f1b831fd4fda7b01bb Mon Sep 17 00:00:00 2001
From: Maxim <mvalcke@hubspot.com>
Date: Tue, 24 May 2022 06:59:59 +0200
Subject: [PATCH] Transform "NonEscapeCharacter" in Ast_utf8_string correct.

Fixes https://github.com/rescript-lang/syntax/issues/473

The [ecmascript spec](https://tc39.es/ecma262/#prod-CharacterEscapeSequence) states that both `SingleEscapeCharacter` and `NonEscapeCharacter` are valid escape sequences.
Previously escape sequences containing a `NonEscapeCharacter`, any regular char like `a` in`"\a"`, would throw the "Invalid escape code" error.
ReScript strings should have the same semantics as JS.
---
 jscomp/frontend/ast_utf8_string.ml        | 8 +++++---
 jscomp/ounit_tests/ounit_unicode_tests.ml | 3 +++
 2 files changed, 8 insertions(+), 3 deletions(-)

diff --git a/jscomp/frontend/ast_utf8_string.ml b/jscomp/frontend/ast_utf8_string.ml
index bf754ad89d..8d62246efd 100644
--- a/jscomp/frontend/ast_utf8_string.ml
+++ b/jscomp/frontend/ast_utf8_string.ml
@@ -25,7 +25,6 @@
 type error =
   | Invalid_code_point
   | Unterminated_backslash
-  | Invalid_escape_code of char
   | Invalid_hex_escape
   | Invalid_unicode_escape
   | Invalid_unicode_codepoint_escape
@@ -36,7 +35,6 @@ let pp_error fmt err =
   match err with
   | Invalid_code_point -> "Invalid code point"
   | Unterminated_backslash -> "\\ ended unexpectedly"
-  | Invalid_escape_code c -> "Invalid escape code: " ^ String.make 1 c
   | Invalid_hex_escape -> "Invalid \\x escape"
   | Invalid_unicode_escape -> "Invalid \\u escape"
   | Invalid_unicode_codepoint_escape ->
@@ -109,7 +107,11 @@ and escape_code loc buf s offset s_len =
   | 'x' ->
       Buffer.add_char buf cur_char;
       two_hex (loc + 1) buf s (offset + 1) s_len
-  | _ -> error ~loc (Invalid_escape_code cur_char)
+  | _ ->
+      (* Regular characters, like `a` in `\a`,
+       * are valid escape sequences *)
+      Buffer.add_char buf cur_char;
+      check_and_transform (loc + 1) buf s (offset + 1) s_len
 
 and two_hex loc buf s offset s_len =
   if offset + 1 >= s_len then error ~loc Invalid_hex_escape;
diff --git a/jscomp/ounit_tests/ounit_unicode_tests.ml b/jscomp/ounit_tests/ounit_unicode_tests.ml
index 97edf17f88..65826252e5 100644
--- a/jscomp/ounit_tests/ounit_unicode_tests.ml
+++ b/jscomp/ounit_tests/ounit_unicode_tests.ml
@@ -47,6 +47,9 @@ let suites =
             Ast_utf8_string.transform_test
             "\\n" =~ "\\n"
         end;
+        __LOC__ >:: begin fun _ ->
+            Ast_utf8_string.transform_test {|\h\e\l\lo \"world\"!|} =~ {|\h\e\l\lo \"world\"!|}
+        end;
         __LOC__ >:: begin fun _ ->
             Ast_utf8_string.transform_test "\\u{1d306}" =~ "\\u{1d306}"
         end;