decode html entities

brothersincode · Jan 1, 2019 · 98b79d1 · 98b79d1
1 parent 801a792
commit 98b79d1
Showing 1 changed file with 8 additions and 0 deletions.
diff --git a/lib/virastar.js b/lib/virastar.js
@@ -30,6 +30,7 @@
   Virastar.prototype.parseOptions = function (options) {
     var defaults = {
       normalize_eol: true,
+      decode_htmlentities: true,
       fix_dashes: true,
       fix_three_dots: true,
       fix_english_quotes_pairs: true,
@@ -127,6 +128,13 @@
       text = text.replace(/(\r?\n)|(\r\n?)/g, '\n');
     }
 
+    // Converts an html characterSet into its original character
+    if (opts.decode_htmlentities) {
+      text = text.replace(/&#(\d+);/g, function (matched, dec) {
+        return String.fromCharCode(dec);
+      });
+    }
+
     // replace double dash to ndash and triple dash to mdash
     if (opts.fix_dashes) {
       text = text.replace(/-{3}/g, '—');