Built by Maven

Usage

commons-text consists of a number of text filters. They receive a CharSequence with the text to filter, and return a CharSequence with the filtered text. If a StringBuilder instance is passed in as CharSequence, it may be reused for performance reasons, and should not be used after the invocation any more.

commons-text v2.0 is optimized for the new Java 8 streaming API.

Example

The KeepFilter is the most simple filter, as it just keeps the text unchanged.

KeepFilter filter = new KeepFilter();
CharSequence result = filter.apply("Hello World");
System.out.println(result.toString());

All filters implement the Function functional interface and can thus be used in Java 8 streams:

String result = Arrays.asList("foo", "<b>bar</b>", "bla")
        .stream()
        .map(new StripHtmlFilter())
        .map(new ParagraphFilter())
        .collect(Collectors.joining(" "));

This example strips HTML from all texts in the stream, then adding paragraphs and finally joining all texts into a single String.

Available Filters

The filters can be roughly divided into two groups: one that expects plaintext as input, and another that expects HTML markup.

Text input

  • KeepFilter: Keeps the text unchanged.
  • GithubFlavoredMarkdownFilter: Like MarkdownFilter, but also accepts GitHub Flavored Markdown extensions like tables, task lists, strikethrough, and heading anchors.
  • HtmlEscapeFilter: Escapes all characters with special meaning in HTML. This is a basic text to HTML converter.
  • LinkToUrlFilter: Detects http, https and ftp links within a plain text, and converts them to HTML links. Optionally, rel=“nofollow” can be added to each link, and a link target can be chosen.
  • MarkdownFilter: Converts CommonMark markdown to HTML. Optionally, a LinkAnalyzer can be applied.
  • NormalizeFilter: Converts Windows (CRLF) and Mac (CR) line endings to Un*x line endings (LF).
  • ParagraphFilter: Surrounds all paragraphs with <p> tags. A paragraph ends with a double line feed. Additionally, <br> tags are added at each simple line feed. Can be used for simply paragraphing a text in HTML.
  • SmilyFilter: Detects a set of text emoticons and replaces them by <img> tags.
  • TextileFilter: Converts textile markup to HTML. Optionally, a LinkAnalyzer can be applied.

HTML input

  • NofollowLinksFilter: Adds rel=“nofollow” to all links in the given HTML document.
  • SimplifyHtmlFilter: Only keeps a defined set of HTML tags (and attributes). All other HTML tags are removed. Can be used to offer simple HTML markup, but remove everything that could inflict XSS.
  • StripHtmlFilter: Removes all HTML markup from the text. Whitespaces are inserted where needed (e.g. “Hello<br>World” is converted to “Hello World”).