Victor Olaya Ferrero ha scritto:
> Andrea,
> 
> The uDig integration is not bridging the GeoTools process API with
> Sextante. It just calls the Sextante algorithms, feeding them with
> GeoTools data objects. The bridge is between the geotools data model and
> the Sextante data model(which is just a set of interfaces). This is the
> way Sextante can be linked to any GIS or geodata library, just wrapping
> data objects so they implement the corresponding Sextante interfaces and
> Sextante can take from them the data it needs to perform analysis.

Right. From my perspective, this is only half of the bridge I'm looking
for. Let me explain what I'm looking for.
In GeoServer I'm trying to setup an improved WPS module that can deal
with both vector data and raster data. This effort is not sponsored
by anyone so it will have to be developed in my spare time, this 
basically means it'll take time and at the moment I'm just trying
to figure out the moving parts of it, and how they connect together.
Now, the module is already using the GT2 process API, and my current
plan is to develop vector modules against it.
For raster modules Sextante is there, and it makes a lot of sense to
use it instead of reinventing the wheel. This could be done in a lot
of ways, but bottom line, I would have to find an integration between
the GeoServer WPS module and the Sextante API.
This could be done in many ways. I could for example teach the
WPS module in GeoServer to speak the gt2 process api and the sextante
api. This would basically mean placing the bridge inside the wps
module.
Yet, since some code has to be written in order to use Sextante, I
would prefer to keep the GeoServer WPS module clean of any Sextante
specific knoledge, and find a way to make the Sextante processes
look exactly like a GeoTools one. This basically means moving the
bridge in gt2, where the reuse potential is higher.
Now, at the moment I still haven't looked into the Sextante API,
but I imagine there should be some similarity between the two
APIs, and that wrapping the Sextante one into the gt2 one should
not be that difficult.

> Bindings for GT2.2, GT2.4 and GT2.5 are already developed and available in
> the SVN repository. 

You mean, in the Sextante repository?

> Also, there is a package that contains classes that
> implement the GeoTools process API and are based on Sextante algorithms,
> wrapping them, but as far as I know, this is not being used in uDig (or
> anywhere else...).

Aah, this is great, GeoServer WPS could be the first user of that one :)

> These classes haven't been tested much, but they could be used for that
> Sextante-based GeoTools module that you propose (which I think is a great
> idea). If so, I would be pleased to contribute and work improving those
> classes or whatever you might consider necessary.

Indeed this is great, and I'm looking forward to it... in time :)
The thing is, the WPS module in GeoServer needs quite some massaging,
the gt2 xml parser/encoder api needs changes too in order to deal 
properly with some WPS oddities, and as I said, I'm working in my spare
time, so no matter how much I love spatial analysis, I will have to deal
with it small step by small step, a weekend after the other,
and this might take a painfully long time.
Anyways, in the meantime I'll check out Sextante and have a look at
the processes, the api and the like.

Thanks a lot for everything, it's much appreciated :)
Cheers
Andrea

-- 
Andrea Aime
OpenGeo - http://opengeo.org
Expert service straight from the developers.

-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
_______________________________________________
Geotools-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/geotools-devel

Reply via email to