Hi Aravind

----- Original Message -----
From: "aravind subramanian" <[EMAIL PROTECTED]>
> Hi,
>
> i am evaluating java xml api's for use in the following task:
>
> I need to display a large (potentially 30000 top-level nodes, each with
> 2-5 child nodes, all text) xml document that is periodically generated
> from a database. A JTree seems to be the best display mechanism, but for
> memory reasons it would have to have a custom TreeModel that is somehow
> bound to the XMl document.

Firstly an XML document can only have 1 top level node. I guess you mean
that the root node contains 30000 immediate children.

Will users actually scroll through these 30,000 nodes in a JTree? Maybe some
kind of table might be better?

Either way some kind of special TreeModel sounds like a requirement which
tries to only keep in memory the part of the tree being viewed at any point
in time.

Maybe some kind of persistent dom4j implementation might help; though maybe
just doing a database-specific TreeModel to deal with the size issue might
help.

> I would appreciate any advice on:
>
> 1) Is dom4j suitable for this task and if so how ( i read about a
> "pruning mode" in the FAQ but am not certain if that would apply, so any
> hints will help). Would SAX be better for this task (compared to the
> dom4j "pruning mode")?

'pruning mode' is designed for processing XML documents as they are parsed,
in a kind of streaming way similar to SAX.

> 2) Is anyone aware of an open source xml viewer of this sort?

Not that I know of.

James

>
>
> Thanks,
>
> aravind
>
>
>
>
> _______________________________________________
> dom4j-user mailing list
> [EMAIL PROTECTED]
> https://lists.sourceforge.net/lists/listinfo/dom4j-user
>



_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com


_______________________________________________
dom4j-user mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/dom4j-user

Reply via email to