Testing for a spectral dataset

One recipe for testing for a spectrum is to look at the axis labels. (whereas a modern approach might use WCS information). Here is a longer example showing how this might be implemented. Suppose the name of the dataset being probed is stored in variable ndf.

     # Get the full attributes.
     ndftrace $ndf fullaxis accept > /dev/null

     # Assign the axis labels and number of dimensions to variables.
     set axlabel = `parget atype ndftrace`
     set nodims = `parget ndim`

     # Exit the script when there are too many dimensions to handle.
     if ( $nodims > 2 ) then
        echo Cannot process a $nodims-dimensional dataset.
        goto exit
     endif

     # Loop for each dimension or until a spectral axis is detected.
     set i = 1
     set spectrum = FALSE
     while ( $i <= nodims && $spectrum == FALSE )

     # For simplicity the definition of a spectral axis is that
     # the axis label is one of a list of acceptable values.  This
     # test could be made more sophisticated.  The toupper converts the
     # label to uppercase to simplify the comparison.  Note the \ line
     # continuation.
        set uaxlabel = `echo $axlabel[$i] | awk '{print toupper($0)}'`
        if ( $uaxlabel == "WAVELENGTH" || $uaxlabel == "FREQUENCY" \
             $uaxlabel == "VELOCITY" ) then

     # Record that the axis is found and which dimension it is.
           set spectrum = TRUE
           set spaxis = $i
        endif
        @ i++
     end

     # Process the spectrum.
     if ( $spectrum == TRUE ) then

     # Rotate the dataset to make the spectral axis along the first
     # dimension.
        if ( $spaxis == 2 ) then
           irot90 $file $file"_rot" accept

     # Fit the continuum.
           sfit spectrum=$file"_rot" order=2 output=$file"_fit" accept
        else
           sfit spectrum=$file order=2 output=$file"_fit accept
        end if
     endif