Add \O, \N

README.md

@@ -381,7 +381,7 @@ Notice that nearly every feature below has at least subtle differences from Java
   </tr>
 
   <tr valign="top">
-    <th align="left" rowspan="5">Character sets</th>
+    <th align="left" rowspan="7">Character sets</th>
     <td>Digit, word</td>
     <td><code>\d</code>, <code>\w</code>, etc.</td>
     <td align="middle">✅</td>
@@ -417,6 +417,26 @@ Notice that nearly every feature below has at least subtle differences from Java
       ✔ Excludes only <code>\n</code> (unlike JS)<br>
     </td>
   </tr>
+  <tr valign="top">
+    <td>Any</td>
+    <td><code>\O</code></td>
+    <td align="middle">✅</td>
+    <td align="middle">✅</td>
+    <td>
+      ✔ Any char (any flags)<br>
+      ✔ Identity escape in char class<br>
+    </td>
+  </tr>
+  <tr valign="top">
+    <td>Non-newline</td>
+    <td><code>\N</code></td>
+    <td align="middle">✅</td>
+    <td align="middle">✅</td>
+    <td>
+      ✔ Excludes only <code>\n</code> (any flags)<br>
+      ✔ Identity escape in char class<br>
+    </td>
+  </tr>
   <tr valign="top">
     <td>Unicode property</td>
     <td>
@@ -860,7 +880,7 @@ Notice that nearly every feature below has at least subtle differences from Java
     <td align="middle">✅</td>
     <td>
       ✔ <code>\u{…}</code> is an error<br>
-      ✔ <code>[\q{…}]</code> matches one of <code>q</code>, <code>{</code>, etc.<br>
+      ✔ <code>[\q{…}]</code> matches <code>q</code>, etc.<br>
       ✔ <code>[a--b]</code> includes the invalid reversed range <code>a</code> to <code>-</code><br>
     </td>
   </tr>

scripts/onig-compare.js

@@ -24,15 +24,15 @@ compare([
   [r`\N`, `\r`],
   [r`[\N]`, `\r`],
   [r`[\N]`, `N`],
-  [r`\o`, `o`],
-  [r`[\o]`, `o`],
-  [r`\o{1}`, `\u{1}`],
-  [r`[\o{1}]`, `\u{1}`],
   [r`\O`, `\n`],
   [r`[\O]`, `\n`],
   [r`\O`, `\r`],
   [r`[\O]`, `\r`],
   [r`[\O]`, `O`],
+  [r`\o`, `o`],
+  [r`[\o]`, `o`],
+  [r`\o{1}`, `\u{1}`],
+  [r`[\o{1}]`, `\u{1}`],
   [r`\p`, `p`],
   [r`[\p]`, `p`],
   [r`\p{`, `p{`],

scripts/utils.js

@@ -70,30 +70,30 @@ function getMatchDetails(match) {
 }
 
 /**
-@type {MatchDetailsFn<Promise<MatchDetails>>}
+@type {MatchDetailsFn}
 */
-const onigurumaResult = async (pattern, str, pos) => {
+const transpiledRegExpResult = (pattern, str, pos) => {
   let result;
   try {
-    result = await onigurumaExec(pattern, str, pos);
+    const options = pos ? {global: true} : undefined;
+    const re = toRegExp(pattern, '', options);
+    if (pos) {
+      re.lastIndex = pos;
+    }
+    result = re.exec(str);
   } catch (err) {
     result = err;
   }
   return getMatchDetails(result);
 };
 
 /**
-@type {MatchDetailsFn}
+@type {MatchDetailsFn<Promise<MatchDetails>>}
 */
-const transpiledRegExpResult = (pattern, str, pos) => {
+const onigurumaResult = async (pattern, str, pos) => {
   let result;
   try {
-    const options = pos ? {global: true} : undefined;
-    const re = toRegExp(pattern, '', options);
-    if (pos) {
-      re.lastIndex = pos;
-    }
-    result = re.exec(str);
+    result = await onigurumaExec(pattern, str, pos);
   } catch (err) {
     result = err;
   }

spec/match-char-set.spec.js

@@ -1,3 +1,4 @@
+import {r} from '../src/utils.js';
 import {matchers} from './helpers/matchers.js';
 
 beforeEach(() => {
@@ -6,6 +7,23 @@ beforeEach(() => {
 
 describe('CharacterSet', () => {
   describe('any', () => {
+    it('should match any character', () => {
+      expect([
+        '\0', '\n', '\r', 'a', '\x85', '\u2028', '\u2029', '\u{10000}', '\u{10FFFF}',
+      ]).toExactlyMatch(r`\O`);
+    });
+
+    it('should match line feed with flag m disabled', () => {
+      expect('\n').toExactlyMatch(r`(?-m)\O`);
+    });
+
+    it('should be identity escape within a char class', () => {
+      expect('O').toExactlyMatch(r`[\O]`);
+      expect('a').not.toFindMatch(r`[\O]`);
+    });
+  });
+
+  describe('dot', () => {
     it('should match any character except line feed', () => {
       expect('\n').not.toFindMatch('.');
       expect([
@@ -16,6 +34,29 @@ describe('CharacterSet', () => {
     it('should match line feed with flag m enabled', () => {
       expect('\n').toExactlyMatch({pattern: '.', flags: 'm'});
     });
+
+    it('should be literal within a char class', () => {
+      expect('.').toExactlyMatch('[.]');
+      expect('a').not.toFindMatch('[.]');
+    });
+  });
+
+  describe('non_newline', () => {
+    it('should match any character except line feed', () => {
+      expect('\n').not.toFindMatch('.');
+      expect([
+        '\0', '\r', 'a', '\x85', '\u2028', '\u2029', '\u{10000}', '\u{10FFFF}',
+      ]).toExactlyMatch(r`\N`);
+    });
+
+    it('should not match line feed with flag m enabled', () => {
+      expect('\n').not.toFindMatch({pattern: r`\N`, flags: 'm'});
+    });
+
+    it('should be identity escape within a char class', () => {
+      expect('N').toExactlyMatch(r`[\N]`);
+      expect('a').not.toFindMatch(r`[\N]`);
+    });
   });
 
   // TODO: Add remaining

src/generate.js

@@ -322,7 +322,7 @@ function genCharacterClassRange(node, state) {
 }
 
 function genCharacterSet({kind, negate, value, key}, state) {
-  if (kind === AstCharacterSetKinds.any) {
+  if (kind === AstCharacterSetKinds.dot) {
     return state.currentFlags.dotAll ?
       ((state.appliedGlobalFlags.dotAll || state.useFlagMods) ? '.' : '[^]') :
       // Onig's only line break char is line feed, unlike JS

src/tokenize.js

@@ -25,7 +25,9 @@ const TokenTypes = /** @type {const} */ ({
 const TokenCharacterSetKinds = {
   any: 'any',
   digit: 'digit',
+  dot: 'dot',
   hex: 'hex',
+  non_newline: 'non_newline',
   posix: 'posix',
   property: 'property',
   space: 'space',
@@ -228,6 +230,20 @@ function getTokenWithDetails(context, pattern, m, lastIndex) {
         }),
       };
     }
+    if (m1 === 'N') {
+      return {
+        token: createToken(TokenTypes.CharacterSet, m, {
+          kind: TokenCharacterSetKinds.non_newline,
+        }),
+      };
+    }
+    if (m1 === 'O') {
+      return {
+        token: createToken(TokenTypes.CharacterSet, m, {
+          kind: TokenCharacterSetKinds.any,
+        }),
+      };
+    }
     if ('RX'.includes(m1)) {
       return {
         token: createToken(TokenTypes.VariableLengthCharacterSet, m, {
@@ -335,7 +351,7 @@ function getTokenWithDetails(context, pattern, m, lastIndex) {
   if (m === '.') {
     return {
       token: createToken(TokenTypes.CharacterSet, m, {
-        kind: TokenCharacterSetKinds.any,
+        kind: TokenCharacterSetKinds.dot,
       }),
     };
   }

src/transform.js

@@ -145,8 +145,12 @@ const FirstPassVisitor = {
 
   CharacterSet({node, replaceWith}, {allowBestEffort, minTargetEs2024}) {
     const {kind, negate, value} = node;
-    if (kind === AstCharacterSetKinds.hex) {
+    if (kind === AstCharacterSetKinds.any) {
+      replaceWith(createUnicodeProperty('Any'));
+    } else if (kind === AstCharacterSetKinds.hex) {
       replaceWith(createUnicodeProperty('AHex', {negate}));
+    } else if (kind === AstCharacterSetKinds.non_newline) {
+      replaceWith(parseFragment(r`[^\n]`));
     } else if (kind === AstCharacterSetKinds.posix) {
       if (!minTargetEs2024 && (value === 'graph' || value === 'print')) {
         if (!allowBestEffort) {