Stricter Onig group name validation

slevithan · Jan 20, 2025 · f01febd · f01febd
1 parent f993078
commit f01febd
Show file tree

Hide file tree

Showing 2 changed files with 8 additions and 9 deletions.
diff --git a/src/parse.js b/src/parse.js
@@ -738,11 +738,10 @@ function getOptimizedGroup(node) {
 }
 
 function isValidGroupNameOniguruma(name) {
-  // Although Onig group name rules are permissive, the validation here is incomplete and can be
-  // improved. The more restrictive rules for valid JS group names are accurately captured in
-  // `isValidGroupNameJs`. Note that backrefs and subroutines might contextually use `-` and `+` to
-  // indicate relative index or recursion level
-  return /^[^-+\d)][^)]*$/.test(name);
+  // The more restrictive rules for valid JS group names are captured in `isValidGroupNameJs`. Note
+  // that backrefs and subroutines might contextually use `-` and `+` to indicate relative index or
+  // recursion level
+  return /^[\p{Alpha}\p{Pc}][^)]*$/u.test(name);
 }
 
 // For any intersection classes that contain only a class, swap the parent with its (modded) child

diff --git a/src/transform.js b/src/transform.js
@@ -179,17 +179,17 @@ const FirstPassVisitor = {
 
   CharacterSet({node, replaceWith}, {accuracy, minTargetEs2024, digitIsAscii, spaceIsAscii, wordIsAscii}) {
     const {kind, negate, value} = node;
-    // Flag D with `\d`, `\p{Digit}`, `[[:digit:]]``
+    // Flag D with `\d`, `\p{Digit}`, `[[:digit:]]`
     if (digitIsAscii && (kind === AstCharacterSetKinds.digit || value === 'digit')) {
       replaceWith(createCharacterSet(AstCharacterSetKinds.digit, {negate}));
       return;
     }
-    // Flag S with `\s`, `\p{Space}`, `[[:space:]]``
+    // Flag S with `\s`, `\p{Space}`, `[[:space:]]`
     if (spaceIsAscii && (kind === AstCharacterSetKinds.space || value === 'space')) {
       replaceWith(setNegate(parseFragment(asciiSpaceChar), negate));
       return;
     }
-    // Flag W with `\w`, `\p{Word}`, `[[:word:]]``
+    // Flag W with `\w`, `\p{Word}`, `[[:word:]]`
     if (wordIsAscii && (kind === AstCharacterSetKinds.word || value === 'word')) {
       replaceWith(createCharacterSet(AstCharacterSetKinds.word, {negate}));
       return;
@@ -838,7 +838,7 @@ function isLoneGLookaround(node, options) {
 }
 
 function isValidGroupNameJs(name) {
-  // JS group names are more restrictive than Onig; see `isValidGroupNameOniguruma`,
+  // JS group names are more restrictive than Onig; see `isValidGroupNameOniguruma` and
   // <developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Lexical_grammar#identifiers>
   return /^[$_\p{IDS}][$\u200C\u200D\p{IDC}]*$/u.test(name);
 }