VERY glad to see your back. :-) It's been pretty lonely around here lately.
The most important thing to do right now is stabilize enough that we can do a alpha/beta release of 1.1. I'm hoping we can add a few of the more commonly asked for features that don't require major surgery, but overall getting a new release out that fixes as many of the common bugs as possible should probably be our priority.
I finally have access to bugzilla so as soon as I have the time I'll set it up so that we can start tracking issues.
Here's what I think we need to do before we can get 1.1 out for wider testing.
- Better test and stabilize the new XML-RPC based XML:DB API and the new embedded XML:DB API.
- Finish implementing compressed document support over XML-RPC. I implemented it for GetDocument to figure out how it works, so I just need to add it in other places as well.
- Make sure we get better / more useful error messages out of the server. From what I've seen of what you did with XML-RPC, it's already quite a bit better, but I'm hoping we can review this more so we stop seeing errors with just numbers or other unhelpful things.
- Update the documentation. I've been doing some of this, but we have no documentation on the new XML-RPC API or the embedded API and the docs still talk about CORBA.
- Decide about XMLObjects. In particular I'd like your opinion here, as I know you were using them. I'd like to replace them with a simpler custom message facility for XML-RPC. If we decide to do that, then we need to implement that facility. Otherwise we need to fix XMLObjects, which I completely broke again.
A few simple features that we should probably add.
- The ability to get query results without the document key and collection being added to the result.
- Through XML-RPC the ability to request partial results on queries. i.e. results 1-10, 11-20 and so on. There's no way to expose this through the XML:DB API, but having it in XML-RPC would be a start.
- Full parsing of documents on the client. Basically we'd just send the compressed byte array from the client to the server. This would also allow us to support simple schema and DTD validation on insertion by simply providing a way to utilize those features in Xerces. This would be implemented as part of the XML:DB API.
Whatever else we can do to make this thing more useful to people.
I've started looking at doing an update to the XML:DB API to fix the issues that it has. Unfortunately, it will be a pretty major change and won't at all be backwards compatible. I'm not sure how to go about this yet.
Anyway, let me know what you want to tackle or point out any things that I' m missing.
On Wednesday, July 17, 2002, at 04:03 PM, Kurt Ward wrote:
Hey guys-
I'm moved and back up and running for the most part. Looks like Kimbro has
been busy with changes for a 1.1 release, very cool. What can I do to help?
Command line tools? XML-RPC optimization? ???
Kurt
Kimbro Staken Java and XML Software, Consulting and Writing http://www.xmldatabases.org/ Apache Xindice native XML database http://xml.apache.org/xindice XML:DB Initiative http://www.xmldb.org
