Merge pull request #1322 from mathjax/fix/replaceUnicode

Update `replaceUnicode()` to handle two `\U{}` calls in a row.

Author: dpvc

testsuite/tests/util/string.test.ts

@@ -48,15 +48,16 @@ describe('string functions', () => {
     expect(string.replaceUnicode(String.raw`\U006d`)).toBe('m');
     expect(string.replaceUnicode(String.raw`\U{6D}`)).toBe('m');
     expect(string.replaceUnicode(String.raw`\U{6d}`)).toBe('m');
+    expect(string.replaceUnicode(String.raw`\U{6d}\U{6d}`)).toBe('mm');
     expect(string.replaceUnicode(String.raw`a \U{62} c`)).toBe('a b c');
-    expect(string.replaceUnicode(String.raw`\\U{61}`)).toBe(String.raw`\\U{61}`);
-    expect(string.replaceUnicode(String.raw`\\\U{61}`)).toBe(String.raw`\\a`);
-    expect(string.replaceUnicode(String.raw`\\\\U{61}`)).toBe(String.raw`\\\\U{61}`);
-    expect(string.replaceUnicode(String.raw`\\\\\U{61}`)).toBe(String.raw`\\\\a`);
-    expect(string.replaceUnicode(String.raw`x\\U{61}`)).toBe(String.raw`x\\U{61}`);
-    expect(string.replaceUnicode(String.raw`x\\\U{61}`)).toBe(String.raw`x\\a`);
-    expect(string.replaceUnicode(String.raw`x\\\\U{61}`)).toBe(String.raw`x\\\\U{61}`);
-    expect(string.replaceUnicode(String.raw`x\\\\\U{61}`)).toBe(String.raw`x\\\\a`);
+    expect(string.replaceUnicode(String.raw`\\U{61}`)).toBe(String.raw`\U{61}`);
+    expect(string.replaceUnicode(String.raw`\\\U{61}`)).toBe(String.raw`\a`);
+    expect(string.replaceUnicode(String.raw`\\\\U{61}`)).toBe(String.raw`\\U{61}`);
+    expect(string.replaceUnicode(String.raw`\\\\\U{61}`)).toBe(String.raw`\\a`);
+    expect(string.replaceUnicode(String.raw`x\\U{61}`)).toBe(String.raw`x\U{61}`);
+    expect(string.replaceUnicode(String.raw`x\\\U{61}`)).toBe(String.raw`x\a`);
+    expect(string.replaceUnicode(String.raw`x\\\\U{61}`)).toBe(String.raw`x\\U{61}`);
+    expect(string.replaceUnicode(String.raw`x\\\\\U{61}`)).toBe(String.raw`x\\a`);
   });
 
   test('toEntity()', () => {

ts/util/string.ts

@@ -42,7 +42,7 @@ export function sortLength(a: string, b: string): number {
  * Quote a string for use in regular expressions
  *
  * @param {string} text  The text whose regex characters are to be quoted
- * @returns {string}  The quoted string
+ * @returns {string}     The quoted string
  */
 export function quotePattern(text: string): string {
   return text.replace(/([\^$(){}.+*?\-|[\]:\\])/g, '\\$1');
@@ -52,7 +52,7 @@ export function quotePattern(text: string): string {
  * Convert a UTF-8 string to an array of unicode code points
  *
  * @param {string} text  The string to be turned into unicode positions
- * @returns {number[]}  Array of numbers representing the string's unicode character positions
+ * @returns {number[]}   Array of numbers representing the string's unicode character positions
  */
 export function unicodeChars(text: string): number[] {
   return Array.from(text).map((c) => c.codePointAt(0));
@@ -62,7 +62,7 @@ export function unicodeChars(text: string): number[] {
  * Convert an array of unicode code points to a string
  *
  * @param {number[]} data   The array of unicode code points
- * @returns {string}         The string consisting of the characters at those points
+ * @returns {string}        The string consisting of the characters at those points
  */
 export function unicodeString(data: number[]): string {
   return String.fromCodePoint(...data);
@@ -71,7 +71,7 @@ export function unicodeString(data: number[]): string {
 /**
  * Test if a value is a percentage
  *
- * @param {string} x   The string to test
+ * @param {string} x    The string to test
  * @returns {boolean}   True if the string ends with a percent sign
  */
 export function isPercent(x: string): boolean {
@@ -81,23 +81,24 @@ export function isPercent(x: string): boolean {
 /**
  * Split a space-separated string of values
  *
- * @param {string} x   The string to be split
+ * @param {string} x    The string to be split
  * @returns {string[]}  The list of white-space-separated "words" in the string
  */
 export function split(x: string): string[] {
   return x.trim().split(/\s+/);
 }
 
 /**
- * Replace \U{...} with the specified unicode character
+ * Replace \U{...} with the specified unicode character and \\ with \
  *
- * @param {string} text   The string to be scanned for \U{...}
- * @returns {string}       The string with the unicode characters in place of \U{...}
+ * @param {string} text   The string to be scanned for \U{...} and \\
+ * @returns {string}      The string with the unicode characters in place of \U{...}
  */
 export function replaceUnicode(text: string): string {
   return text.replace(
-    /((?:^|[^\\])(?:\\\\)*)\\U(?:([0-9A-Fa-f]{4})|\{\s*([0-9A-Fa-f]{1,6})\s*\})/g,
-    (_m, pre, h1, h2) => pre + String.fromCodePoint(parseInt(h1 || h2, 16))
+    /\\U(?:([0-9A-Fa-f]{4})|\{\s*([0-9A-Fa-f]{1,6})\s*\})|\\./g,
+    (m, h1, h2) =>
+      m === '\\\\' ? '\\' : String.fromCodePoint(parseInt(h1 || h2, 16))
   );
 }