dtddoc step 1: Parsing a DTD (1/2) - exploring XML

dtddoc step 1: Parsing a DTD

Now that we have looked at the -- to my knowledge -- only tool for documenting DTDs, let's get started with our own development effort. This can be broken down into the following steps:

Parsing a DTD

Creating the internal data structure

Adding element and attribute descriptions

Generating HTML

We will look at the first step today and leave the others to later installments of this column.

Parsing a DTD

A DTD parser has the job of reading a textual representation of a DTD and transforming it into an in-memory representation suitable for further manipulation in the respective programming language. Instead of writing a DTD parser ourselves, we find these software packages on the Web:

All of these parsers have a straightforward way of calling them on an input stream and creating an internal data structure in return.

Bourret's DTD parser

With Bourret's parser you simply create a DTDParser and call either parseXMLDocument() or parseExternalSubset(), depending on whether the DTD is embedded in the XML document or external to it:

import org.sax.InputSource; import org.xmlmiddleware.schemas.dtds.DTDParser; import org.xmlmiddleware.schemas.dtds.DTD; DTDParser parser = new DTDParser(); InputSource is = new InputSource("html4.dtd"); DTD dtd = parser.parseExternalSubset(is, null);