So I've discovered - but I still give me an answer to my question. The best path I've found thus far is to convert the DTDs into XML schema and use Sun's XSOM API to parse the XML Schema into memory.
I'm trying this out now. JAXB looked good from the surface, but then one learns this requires a manual compile of the XML Schema or DTD, and then you have to write code specific to the Java Objects generated from the compile. I need something dynamic, where the structure of the XML document is unknown until parse time, as if the user specifies the DTD. I want to, at run-time, iterate over the structure and provide the user with actions to be taken when those elements are encountered in the XML documents. _____ Scott Andrews Principle Software Engineer Concurrent Technologies Corporation (814) 269 6580 (Monday, Wednesday, Friday) (814) 632 9559 (Tuesday, Thursday) (814) 880 8522 (Cell) -----Original Message----- From: Anne Thomas Manes [mailto:[EMAIL PROTECTED] Sent: Tuesday, January 13, 2004 10:35 AM To: [EMAIL PROTECTED] Subject: Re: How do I parse a DTD in Java? That's because a DTD is NOT XML. At 09:54 AM 1/13/2004, you wrote: >How do I parse a DTD into an in-memory Java object, like a TreeMap or >perhaps some XML specific collection class? > >I asked this question the other day, and got an answer that the >DocumentBuilder parse method should handle the parsing of a DTD - since a >DTD IS XML. > >However, I get basic parsing errors when inputting a simple DTD. The code >works fine on XML documents, but not on DTDs. The code I'm using to parse >the DTD looks like this: > > public static void main( String argArgs[] ) { > { > File dtdFile = new File( "C:\\APIS\\WorkSpace\\tv.dtd" ); > DocumentBuilderFactory dbf = > DocumentBuilderFactory.newInstance(); > DocumentBuilder db = dbf.newDocumentBuilder(); > Document document = db.parse( dtdFile ); > parseChildrenRecursivly( document.getChildNodes(); ); > } > > public void parseChildrenRecursivly( NodeList argNodeList ) { > > if (argNodeList == null) { > return; > } > > Node node; > for (int i=0; i<argNodeList.getLength(); i++) { > node = argNodeList.item( i ); > if (node.getNodeType() != Node.TEXT_NODE) { > System.out.println( > "node.nodeName = " + node.getNodeName() + "; " > + > "node.nodeType = " + Short.toString( > node.getNodeType() ) + "; " + > "node.localName = " + node.getLocalName() + "; " + > "node.namespaceUri = " + node.getNamespaceURI() + > "; " + > "node.nodeValue = " + node.getNodeValue() + "; " + > "" > ); > parseChildrenRecursivly( node.getChildNodes() ); > } > } // for > } > >However, I get errors when making the attempt: > >[Fatal Error] :-1:-1: Premature end of file. >ERR:> Exception Premature end of file. > >The DTD I'm trying to parse is just an example. It looks like this, where >the elements are embedded inside the DOCTYPE tag: > ><!DOCTYPE TVSCHEDULE [ > ><!ELEMENT TVSCHEDULE (CHANNEL+)> ><!ELEMENT CHANNEL (BANNER, DAY+)> ><!ELEMENT BANNER (#PCDATA)> ><!ELEMENT DAY ((DATE, HOLIDAY) | (DATE, PROGRAMSLOT+))+> ><!ELEMENT HOLIDAY (#PCDATA)> ><!ELEMENT DATE (#PCDATA)> ><!ELEMENT PROGRAMSLOT (TIME, TITLE, DESCRIPTION?)> ><!ELEMENT TIME (#PCDATA)> ><!ELEMENT TITLE (#PCDATA)> ><!ELEMENT DESCRIPTION (#PCDATA)> > ><!ATTLIST TVSCHEDULE NAME CDATA #REQUIRED> ><!ATTLIST CHANNEL CHAN CDATA #REQUIRED> ><!ATTLIST PROGRAMSLOT VTR CDATA #IMPLIED> ><!ATTLIST TITLE RATING CDATA #IMPLIED> ><!ATTLIST TITLE LANGUAGE CDATA #IMPLIED> > >]> > >If I just parse the ELEMENTS, by removing the DOCTYPE tag, I still get errors: > >Exception The markup in the document preceding the root element must be >well-formed. >[Fatal Error] tv.dtd:3:3: The markup in the document preceding the root >element must be well-formed. > >Anybody have a clue how to parse a DTD, so I can get an in-memory >structure of the DTD in Java? > > > >---------- >Scott Andrews >Principle Software Engineer >Concurrent Technologies Corporation >(814) 269 6580 (Monday, Wednesday, Friday) >(814) 632 9559 (Tuesday, Thursday) >(814) 880 8522 (Cell) > > --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
