html_unescape

Build Status

A Dart library for unescaping HTML-encoded strings.

Supports:

  • Named Character References ( )
    • 2099 of them
  • Decimal Character References (á)
  • Hexadecimal Character References (ã)

The idea is that while you seldom need encoding to such a level (most of the time, all you need to escape is <, >, /, & and "), you do want to make sure that you cover the whole spectrum when decoding from HTML-escaped strings.

Inspired by Java's unbescape library.

Usage

A simple usage example:

import 'package:html_unescape/html_unescape.dart';

main() {
  var unescape = HtmlUnescape();
  var text = unescape.convert("&lt;strong&#62;This &quot;escaped&quot; string");
  print(text);
}

You can also use the converter to transform a stream. For example, the code below will transform a POSIX stdin into an HTML-unencoded stdout.

await stdin
    .transform(Utf8Decoder())
    .transform(HtmlUnescape())
    .transform(Utf8Encoder())
    .pipe(stdout);

Full versus small

If you're sure you will only encounter the most common escaped characters, you can import 'package:html_unescape/html_unescape_small.dart' instead of the full version. This will decrease code size and increase performance. The only difference is in the size of the Named Character Reference dictionary. The full set includes the likes of &DownLeftRightVector; or &UpArrowBar; while the small set only includes the first 255 charcodes.

Issues

Please use GitHub tracker. Don't hesitate to create pull requests, too.

Libraries

html_unescape
Unescapes HTML5-escaped strings.
html_unescape_small
Unescapes HTML5-escaped strings. This version doesn't support all runes to decrease code size and increase performance.