Go to the FreeLists Home Page Home Signup Help Login
 



Browse freeroleplay: This Month's ArchiveMain Archive PageRelated postsPrevious by DateNext by Date

[freeroleplay] Generic Document Support

  • From: Samuel Penn <sam@xxxxxxxxxxxxxxxxxxx>
  • To: freeroleplay@xxxxxxxxxxxxx
  • Date: Mon, 24 May 2004 21:43:33 +0100
Currently, the Yagsbook Encyclopedia only supports Yagsbook XML
format, which in turns only supports Yags ruleset. It would be
nice if things were made more generic, to enable the support of
other document formats and other systems.

System support is just down to the XML and XSLT, so I'll forget
that for now and concentrate on the Java side - i.e. support for
different XML formats.

Obviously, each format will have to support certain information
structures. The Encyclopedia needs to know the URI of the document,
its title and topic sets to name just some of the basic information.
One way of doing this is to only support XML formats which implement
a certain format, i.e. the title of the document is always at 
/article/header/title. This would mean any support document format
would need at least, but probably no more than, a /article/header
block similar to Yagsbook.

A more flexible approach would be to support an interface for Java
classes, which handle the getting of information out of the document.
You could have an ISource interface which handles obtaining the
documents from some generic source (file system, CVS, webdav, SOAP),
which is passed to the core software. The ISource then provides a
list of IEntry interface objects, which define methods such as
getUri(), getTitle(), getCategory() for obtaining the necessary meta
information.

However, any supported format would need to have the concept of the
above meta data, even though how it represents it is implementation
dependent.

Just to make things clear, it would work by the Encyclopedia class
have an add(ISource src) method to add an (already instantiated)
data source. The ISource interface defines a
(IEntry[])getAllEntries() method to getting all the entries out.
Each IEntry represents an individual XML document, giving access
to it via getter methods.

I think this latter is the best way forward.


Would the CML format provide this sort of meta data? I need to flesh
out details about what I mean by TopicSet for instance (a concept I've
taken from NewsML) since currently it's all pretty much just in my head.


Is this sort of thing getting off-topic for the list? I've stuck this
email on the Yagsbook developer forum on SourceForge just in case.

-- 
Be seeing you,                          http://www.bifrost.demon.co.uk/
Sam.                                    jabber: samuel.penn@xxxxxxxxxx

Other related posts:

  • [freeroleplay] Generic Document Support
  • [freeroleplay] Re: Generic Document Support
  • [freeroleplay] Re: Generic Document Support




  • [ Home | Signup | Help | Login | Archives | Lists ]

    All trademarks and copyrights within the FreeLists archives are owned by their respective owners.
    Everything else ©2008 Avenir Technologies, LLC.