RANGE support incorporated for rasters

[Frank Warmerdam]

Bill Binko wrote:
> Frank,
> 
> I have no need to backport this to 4.8.2 -- I am currently working off
> of a patched CVS HEAD.
> 
> That said, my offer still stands to finish the
> Gradient/ColorMap/ColorRange/ColorRamp/whatever support as soon as the
> RFC is finalized and approved.

Bill,

Hmm, the RFC has languished for quite a while. With 4.8 in the can, I think
it makes sense to revisit the RFC and get on with things.

Reviewing the RFC I think we should go with the COLORRAMP object in
the mapfile and mapscript.  Addition of the INTERVALS and METHOD is good.

On the LEGEND, is there any issue with a legend icon that is taller than
normal?  I don't know too much about how legends.  Also, can we make the
light grey top/bottom values optional?  Do we need a new keyword to set
the color for the top/bottom text?  Calling LEGEND experts!

> Frank, did your changes require any changes to that RFC?  Also, did you
> make any changes on the legend support?

I think the RFC should explain that the RANGEITEM is the name of the
attribute field used for evaluation in the range for vectors and that
for rasters it is not required since the implicit value is the raw pixel
value.

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    | Geospatial Programmer for Rent