Hi Daniel,
This is a two part problem, one half of which Jess might be able to
help with. The first part is parsing the XML. Parsing a well-defined
computer language is a straightforward procedure that can and should
be done by a simple, deterministic parsing program. These abound for
XML; I've been using Sun's early-access XML classes (see
developer.javasoft.com), but there are many more; Microsoft's is
popular, if standards-ignorant. If you did in fact produce a set of rules
which somehow did the actual parsing of an XML document, it'd be
unbelievably slow (and downright silly, frankly.) A validating parser
can check the doc against a DTD, and this should properly handle
validation of most documents. But let's assume that your company has
special requirements that cannot be represented in a DTD.
Now. Given that you have a -parsed- XML document, this means you have
in-memory data structures to represent it. Jess could indeed be set up
to run a rulebase which reacted to these objects. You could write a
bit of Java code which walked these data structures and asserted a set
of facts into Jess which represented the XML document. Note that this
is easy once the document has been parsed; very hard beforehand.
Anyway, those are my 2 cents on the matter.
I think Daniel B. Davis wrote:
>
> I am about to start a project involving XML documents. In particular there
> will need to be rules for searching XML for particular tags and text.
>
> Although I have subscribed for some time, I dont remember that past
> correspondence covered anything like that; I believe not.
>
> The job involves verifying whether documents, expressed in XML, conform to
> certain standards.
> There are a lot of them and they are fussy. Also, they change from time to
> time.
>
> It is possible to assert the XML as facts, but in even a modest document,
> there could be a lot of facts. Also, some of the implied structure would be
> missing, and might need to be otherwise represented. And repeated. Java,
> alas, does not support XML directly, but through parsers. Yet the very
> attractive JESS features include:
> - the rule orientation
> - the RETE network
> - the simplicity of the language
> - the ability to instantiate java classes
>
> All combine to suggest that use of JESS is appropriate if matching can be
> worked out.
>
> Therefore, this query is whether anyone knows of or is doing current work
> along these lines.
>
>
> ---------------------------------------------------------------------
> To unsubscribe, send the words 'unsubscribe jess-users [EMAIL PROTECTED]'
> in the BODY of a message to [EMAIL PROTECTED], NOT to the
> list. List problems? Notify [EMAIL PROTECTED]
> ---------------------------------------------------------------------
>
>
---------------------------------------------------------
Ernest Friedman-Hill
Distributed Systems Research Phone: (510) 294-2154
Sandia National Labs FAX: (510) 294-2234
Org. 8920, MS 9214 [EMAIL PROTECTED]
PO Box 969 http://herzberg.ca.sandia.gov
Livermore, CA 94550
---------------------------------------------------------------------
To unsubscribe, send the words 'unsubscribe jess-users [EMAIL PROTECTED]'
in the BODY of a message to [EMAIL PROTECTED], NOT to the
list. List problems? Notify [EMAIL PROTECTED]
---------------------------------------------------------------------
