remove useless slash-escapes of non-ascii characters. Closes #1506

terser · Mar 28, 2024 · 1306a72 · 1306a72
1 parent 1161ff1
commit 1306a72
Show file tree

Hide file tree

Showing 2 changed files with 19 additions and 1 deletion.
diff --git a/lib/parse.js b/lib/parse.js
@@ -814,7 +814,12 @@ function tokenizer($TEXT, filename, html5_comments, shebang) {
         while ((ch = next(true))) if (NEWLINE_CHARS.has(ch)) {
             parse_error("Unexpected line terminator");
         } else if (prev_backslash) {
-            source += "\\" + ch;
+            if (/^[\u0000-\u007F]$/.test(ch)) {
+                source += "\\" + ch;
+            } else {
+                // Remove the useless slash before the escape, but only for characters that won't be added to regexp syntax
+                source += ch;
+            }
             prev_backslash = false;
         } else if (ch == "[") {
             in_class = true;

diff --git a/test/compress/regexp.js b/test/compress/regexp.js
@@ -63,6 +63,19 @@ unsafe_nul_byte: {
     }
 }
 
+double_escape: {
+    format = {
+        ascii_only: true
+    }
+    input: {
+        /\🏳0\🌈️\☺/
+    }
+    expect: {
+        /\ud83c\udff30\ud83c\udf08\ufe0f\u263a/
+    }
+    expect_stdout: true
+}
+
 inline_script: {
     options = {}
     beautify = {