is it possible to draw grd file using Mapserver

[Frank Warmerdam]

Murat Beyhan wrote:
> So I draw both layer by using it's .shp .dbf file.
> As you see both are point file.
> But I need to know if I draw both on the one map, Which villages  
> placed  which zones.
> Zones could created by using acceleration. For example 0-100 gal is zone 
> 1, 101-200gal is zone2, etc.
> 
> I guess, If you draw acceleration using point data you could not 
> distinguish data as zone.
> May be by using point data I can obtain grid file or polygon file  than 
> I can determine the village point place on the polygon.
> When you draw acceleration distribution the shape of the distribution
> seems circular area then when you classify zones  you can get  result of 
> the distribution which are divided  into several  level.

Murat,

It sounds like you need an algorithm to generate a raster by interpolation
from your acceleration point data.  There are many options to do this,
but three that come to mind are GRASS, GMT or the gdal_grid program new
in GDAL 1.5.

Once you have an interpolated raster grid of the accelerations you could
draw each class/range in a distinct color using classes to give a visual
indication of the zones.

If you want to know what the zone is for each village I would suggest
doing this as a distinct processing step.  Something where you sample the
grid for each village location and attach a new column with the actual
value, or the corresponding zone.  This might be accomplished as a
GDAL/OGR script (eg. in python).  I'm sure this is also something that
can be accomplished in GRASS and perhaps in GMT.

I guess my main point is that you need to do various preprocessing steps
before visualization and that these are best done in other packages more
suited to the task.  As we say "MapServer is not a GIS" which means we aren't
trying to replicate all sorts of GIS processing in MapServer.

All that said, in theory you could accomplish something vaguely similar in
MapServer by writing a complicated mapscript script that for each village
did a search for the nearest point and apply a zonation on that basis.  But
I wouldn't advise this approach for performance, complexity and inexactness
reasons.

Best regards,
-- 
---------------------------------------+--------------------------------------
I set the clouds in motion - turn up   | Frank Warmerdam, warmerdam at pobox.com
light and sound - activate the windows | http://pobox.com/~warmerdam
and watch the world go round - Rush    | President OSGeo, http://osgeo.org