Support \u{...} outside of character classes

demo/index.html

@@ -57,8 +57,8 @@ <h2>Themes</h2>
     <div class="info no-margin-collapse">
       <p id="info-theme-default">You don't need to add a stylesheet to your page to use the default theme. Just run <code>RegexColorizer.loadStyles()</code>.</p>
       <p id="info-theme-nobg" class="hidden">The No BG theme avoids background colors as part of highlighting.</p>
-      <p id="info-theme-regexpal" class="hidden">In 2007, <a href="https://stevenlevithan.com/regexpal/">RegexPal</a> was the first web-based regex tester with regex syntax highlighting. Regex Colorizer started out by extracting RegexPal's highlighting code into a standalone library. This theme uses all black text, because RegexPal's implementation used highlighted text underneath a <code>textarea</code> with a transparent background. The RegexPal theme doesn't uniquely distinguish: escaped literal tokens, backreferences, and character class boundaries.</p>
-      <p id="info-theme-regexbuddy" class="hidden">OG <a href="https://www.regexbuddy.com/">RegexBuddy</a>'s highlighting styles. RegexBuddy inspired RegexPal and numerous regex testers that came after it. Currently, the RegexBuddy theme doesn't uniquely distinguish: escaped literal tokens, backreferences, character class boundaries, and range-hyphens.</p>
+      <p id="info-theme-regexpal" class="hidden">In 2007, <a href="https://stevenlevithan.com/regexpal/">RegexPal</a> was the first web-based regex tester with regex syntax highlighting. Regex Colorizer started out by extracting RegexPal's highlighting code into a standalone library. This theme uses all black text, because RegexPal's implementation used highlighted text underneath a <code>textarea</code> with a transparent background. The RegexPal theme doesn't <em>uniquely</em> distinguish the following: escaped literal tokens, backreferences, and character class boundaries.</p>
+      <p id="info-theme-regexbuddy" class="hidden">OG <a href="https://www.regexbuddy.com/">RegexBuddy</a>'s highlighting styles. RegexBuddy inspired RegexPal and some other regex testers that came after it. Currently, the RegexBuddy theme doesn't <em>uniquely</em> distinguish the following: escaped literal tokens, backreferences, character class boundaries, and range-hyphens.</p>
     </div>
 
     <h2>More examples</h2>

regex-colorizer.js

@@ -94,7 +94,7 @@ const RegexColorizer = (() => {
 
   /**
    * Returns the character code for the provided regex token. Supports tokens used within character
-   * classes only, since that's all it's currently needed for.
+   * classes only.
    * @param {string} token Regex token.
    * @returns {number} Character code of the provided token, or NaN.
    */
@@ -585,6 +585,33 @@ const RegexColorizer = (() => {
             lastToken = {
               quantifiable: false,
             };
+          // '\u{...}'
+          } else if (m.startsWith('\\u{')) {
+            if (flagsObj.unicode) {
+              const charCode = getTokenCharCode(m);
+              output += charCode <= 0x10FFFF ?
+                to.metasequence(m) :
+                to.error(m, error.INDEX_OVERFLOW);
+              lastToken = {
+                quantifiable: true,
+              };
+            // Non-Unicode mode and '\u{...}' includes only decimal digits, so treat as an escaped
+            // literal 'u' followed by a quantifier
+            } else if (/^\\u{\d+}$/.test(m)) {
+              // If there's a following `?` it will be handled as the next token which technically
+              // isn't correct, but everything is still highlighted correctly apart from the gap in
+              // tokens that might be visible depending on styling
+              output += to.error('\\u', error.INCOMPLETE_TOKEN) + to.error(m.slice(2), error.UNQUANTIFIABLE);
+              lastToken = {
+                quantifiable: false,
+              };
+            // Non-Unicode mode and '\u{...}' includes hex digits A-F/a-f
+            } else {
+              output += to.error('\\u', error.INCOMPLETE_TOKEN) + m.slice(2);
+              lastToken = {
+                quantifiable: true,
+              };
+            }
           // Unquantifiable metasequence
           } else if ('bB'.includes(char1)) {
             output += to.metasequence(m);