XML Elves

M/Gateway Developments has released M/DB:X, an XPath-based XML Database with a non-RESTful HTTP interface based on DOM. There doesn't seem to be any support for queries across documents--you have to know the ID of the document you're querying--so it doesn't really replace a good XQuery database. There might be some use cases for this I'm ot seeing right now, but in general I think most projects would be better served by a traditional XQuery database such as eXist instead. M/DB:X is licensed under the GNU Affero General Public License Version 3.

Read the Complete News Item on Cafe con Leche...

Thanks for drawing attention to M/DB:X. The first version is very much just that: a first "bare-bones" version. It's actually a spin-off from two existing products, one (M/DB) which is an open source clone of SimpleDB and the other (EWD) where M/DB:X is exposing, via HTTP, its core XML DOM engine which is used as an HTML/XML transformation engine.

As such I've initially exposed the existing functionality of the EWD core DOM and used M/DB's (SimpleDB-based) HTTP and security mechanism to provide the interface. So it's got what EWD needed and not much more.

As it is released as a free open source product, I'm hoping that others will help to extend its functionality, XQuery being an obvious priority. There's no reason why this couldn't be done relatively straightforwardly, given time and a knowledge of XQuery's requirements: M/DB:X's core DOM has all the fundamental requirements to make it possible and the underlying GT.M database engine on which it's built would provide extremely good performance for this kind of thing.

As to use cases, I'd identify several contenders (although I'm hoping that the simple HTTP-based interfacing and lightweight nature of M/DB:X's approach to XML handling will inspire people to dream up novel new use cases):

- as I said, EWD (which is a web/ajax application development platform) uses the DOM as the basis of its HTML/XML compiler, for transforming high-level HTML/XML descriptions of web pages into PHP, ASP.Net, JSP, etc versions. It's a very effective and tried and tested platform for this kind of processing. M/DB:X makes that capability now available in a language-independent way via HTTP

- analysis, data extraction and re-packaging of HTML documents is another contender. M/DB:X's non-validating parser will convert any old HTML into a properly structured DOM (as best it can) that can then be manipulated using DOM techniques rather than crude text manipulation/pattern matching.

I'm sure there are other uses for M/DB:X. Programming using DOM techniques is very powerful and is something I've done a lot of within EWD, and I'm always amazed at the kind of sophisticated transformations that can be done in just a few simple DOM manipulations. I'm hoping that M/DB:X will make the kind of stuff I've been able to do with incredible ease over the last few years available to a lot more people.

Anyone interested in helping or finding out more, just contact me at rtweed@mgateway.com

Thanks

Rob