DIHtmlParser is a lightening fast and flexible Html Parser for Borland Delphi 4/5/6/7. It fully support Unicode / WideStrings throughout and is the perfect development tool to quickly extract various information from Html documents.

The core TDIHtmlParser component recognizes 12 different pieces of Html, like tags, text, comments, scripts, styles, titles, among others. Tags are parsed into the TDIHtmlTag object and can easily be queried for attributes and values.

If applications do not need to see a particular tag or piece of Html, they can simply turn it off and the parser will quickly skip it and not report it to the application. This Html piece and tag filtering is not only a feature, but significantly speeds up the parsing process.

Plugins are available to extend the Html Parser's functionality. Any number of plugins can be assigned to a parser, and each TDIHtmlParserPlugin implements its own filtering mechanism just like the main parser.

The plugins allow to extract additional information "on the fly" next to the main parsing process. The DIHtmlParser package ships with 6 plugins, which extract e-mail addresses, links and titles, implement Html events, track Html tables, and automatically write Html data.

The Html Parser reads Html data from memory or from a TStream object. It correctly handles preformatted text and converts Html entities, including custom ones. "Smart parsing", which jumps over filtered Html pieces, makes the Html Parser amazingly fast despite of its rich set of features.

The DIHtmlParser package includes full Html Help plus various demo projects. The Demo version shows a once-only reminder when run outside the IDE. Commercial and Source Code licenses are available for a small fee.