[mapserver-users] Java - Mapserver

[James Macgill]

At 21:30 03/09/2002, Steve Lime wrote:
>I'd be interested in seeing the new code base.

I've made a snapshot available for people not able to use CVS, its not a 
formal release but a handy stopgap.

>  I think it's critical
>that MapServer have capabilities in the Java world.  There are other
>options, perhaps the most intruiging would be leveraging GeoTools.
>There's no way, unless a bounty of developers appears, that MapServer
>will be re-written in Java. I'm not even sure that it's a good idea.
>What about the possibility of writing a set of classes on top of
>GeoTools to mimic MapScript?

I have to confess that, whilst we use MapServer a lot, its tends to be as a 
wms and not in conjunction with MapScript so I'm not yet familiar with it 
or how much would be involved.

>Sure, some things would have to be written
>from scratch, but I would imagine (having not seen the GT2 APIs) that
>much of the core functionality already exists in GeoTools (graphics
>creation/query).

GT2 is still quite new, graphics creation with good control over style is 
defiantly there, the exact nature of the query mechanism is being worked 
on. (Though, all base code for processing a query is in place)

>  That would give you a pure Java version of
>MapServer/MapScript. The advantage of mimicing MapScript is that there's
>already a user/documentation/application base to draw from.

That does sound like a very useful option providing the two api's could be 
kept close enough together.

>  You'd
>probably also have classes available to use the MapServer style of
>application development (for better or worse) with regular GeoTools
>components, thus extending MapScript and GeoTools.

GT2 has a modular architecture, so adding additional functionality like 
this would not bloat the core API.  (For example, the wmsserver is 
implemented as a separate module)
>Thoughts?

I'd be more than happy to assist in the creation of an API to mimic 
MapScript, especially as doing so would ensure that GT2 supported a good 
key functionality set.  I think development would be iterative, with more 
of the MapScript API being available as the capabilities of GT2 expanded.

My own time for such a project would be limited as I have to focus on the 
main development of GT2, however I could assist in explaining the GT2 
architecture, and extending the capabilities of GT2 when they fell behind 
what was needed to support MapScript.

All the best

James

-- 
James Macgill
Center for Computational Geography http://www.ccg.leeds.ac.uk
Spell Checker (c) Creative Spelling inc (aka my dyslexic brain)
http://www.geotools.org a client side java mapping toolkit.