[gdal-dev] Scale and Offset, datatype and UMN MapServer

[Maarten Plieger]

Hi,

When using a file with scale and an offset to transform 8bit BYTE raster 
values into meaningful values, what happens with the datatype? I would 
like to implement the scale and offset parameters in our ADAGUC netcdf4 
driver.
I have seen some drivers to change the datatype to float. What happens 
when translating from such a file to another file? Will the datatype be 
changed to float always?

What happens with UMN MapServer in combination with such a file? I can 
imagine that mapserver does not see a scale and offset parameter, but 
just a raster with datatype float?

The problem is that I would like to pass the reduced BYTE raster data 
with accompaning scale and offset through the Web Coverage Service of 
Mapserver. I do not want to pass the unpacked float raster data.

Is there a standard and neat way to do this?

Best regards,
 Maarten Plieger


-- 
Maarten Plieger
KNMI, R&D Information and Observation Technology, De Bilt
(t) +31 30 2206330