fix(color-contrast): ignore format unicode characters (#4102)

* fix(color-contrast): ignore format unicode characters * comment * typo * Apply suggestions from code review Co-authored-by: Wilco Fiers <WilcoFiers@users.noreply.github.com> --------- Co-authored-by: Wilco Fiers <WilcoFiers@users.noreply.github.com>
dequelabs · Jul 26, 2023 · 049522e · 049522e
1 parent c6e07be
commit 049522e
Show file tree

Hide file tree

Showing 4 changed files with 130 additions and 76 deletions.
diff --git a/lib/commons/text/has-unicode.js b/lib/commons/text/has-unicode.js
@@ -1,7 +1,8 @@
 import {
   getUnicodeNonBmpRegExp,
   getSupplementaryPrivateUseRegExp,
-  getPunctuationRegExp
+  getPunctuationRegExp,
+  getCategoryFormatRegExp
 } from './unicode';
 import emojiRegexText from 'emoji-regex';
 
@@ -20,19 +21,22 @@ import emojiRegexText from 'emoji-regex';
  */
 function hasUnicode(str, options) {
   const { emoji, nonBmp, punctuations } = options;
+  let value = false;
+
   if (emoji) {
-    return emojiRegexText().test(str);
+    value ||= emojiRegexText().test(str);
   }
   if (nonBmp) {
-    return (
+    value ||=
       getUnicodeNonBmpRegExp().test(str) ||
-      getSupplementaryPrivateUseRegExp().test(str)
-    );
+      getSupplementaryPrivateUseRegExp().test(str) ||
+      getCategoryFormatRegExp().test(str);
   }
   if (punctuations) {
-    return getPunctuationRegExp().test(str);
+    value ||= getPunctuationRegExp().test(str);
   }
-  return false;
+
+  return value;
 }
 
 export default hasUnicode;
diff --git a/lib/commons/text/remove-unicode.js b/lib/commons/text/remove-unicode.js
@@ -1,7 +1,8 @@
 import {
   getUnicodeNonBmpRegExp,
   getSupplementaryPrivateUseRegExp,
-  getPunctuationRegExp
+  getPunctuationRegExp,
+  getCategoryFormatRegExp
 } from './unicode.js';
 import emojiRegexText from 'emoji-regex';
 
@@ -25,8 +26,10 @@ function removeUnicode(str, options) {
     str = str.replace(emojiRegexText(), '');
   }
   if (nonBmp) {
-    str = str.replace(getUnicodeNonBmpRegExp(), '');
-    str = str.replace(getSupplementaryPrivateUseRegExp(), '');
+    str = str
+      .replace(getUnicodeNonBmpRegExp(), '')
+      .replace(getSupplementaryPrivateUseRegExp(), '')
+      .replace(getCategoryFormatRegExp(), '');
   }
   if (punctuations) {
     str = str.replace(getPunctuationRegExp(), '');

diff --git a/lib/commons/text/unicode.js b/lib/commons/text/unicode.js
@@ -83,3 +83,15 @@ export function getSupplementaryPrivateUseRegExp() {
   //      ┏━━━━━━┻━━━━━━┓┏━━━━━━┻━━━━━━┓
   return /[\uDB80-\uDBBF][\uDC00-\uDFFF]/g;
 }
+
+/**
+ * Get regular expression for unicode format category.
+ * When we drop IE11 we can instead use unicode character escape `/p{Cf}/gu`
+ * Reference:
+ * - https://www.compart.com/en/unicode/category/Cf
+ *
+ * @returns {RegExp}
+ */
+export function getCategoryFormatRegExp() {
+  return /[\xAD\u0600-\u0605\u061C\u06DD\u070F\u08E2\u180E\u200B-\u200F\u202A-\u202E\u2060-\u2064\u2066-\u206F\uFEFF\uFFF9-\uFFFB]|\uD804[\uDCBD\uDCCD]|\uD80D[\uDC30-\uDC38]|\uD82F[\uDCA0-\uDCA3]|\uD834[\uDD73-\uDD7A]|\uDB40[\uDC01\uDC20-\uDC7F]/g;
+}