strip LTM and RTM entities when normalize text

To support bidirectional text alphabetization (https://en.wikipedia.org/wiki/Bidirectional_text) Via EbookFoundation/free-programming-books#6714
davorpa · Feb 11, 2022 · f0c365d · f0c365d
1 parent ee5f968
commit f0c365d
Showing 1 changed file with 2 additions and 2 deletions.
diff --git a/lib/alphabetical-list-items.js b/lib/alphabetical-list-items.js
@@ -3,8 +3,8 @@ const visit = require('unist-util-visit');
 const toString = require('mdast-util-to-string');
 
 function normalize(text) {
-  const removeAtBeginning = /^([-._(《"'])*/;
-  const removeInside = /[,:]/;
+  const removeAtBeginning = /^([-._(《"'\u200e\u200f])*/;
+  const removeInside = /[,:\u200e\u200f]/;
   const replaceWithSpace = /-/;
   return text.toLowerCase()
     .trim()