Syntax: linear (wavelength calib. FITS file)
           [-image] [-loglin]
           [disp=d]  [zero=w]  [-1d]  [-nozero]  [ef=filename]
           [row=]  [-listdisp]  [-lgi]  [table=filename]
   Produces linear wavelength calibrated FITS spectra.  Input:
   file.fits  Output (each row): file-##.fits where ## is the
   row number.  Output (whole image): file-L.fits .
 Options:
   -image : Create a 2 dimensional image of linear spectra with
            IRAF compatible header cards.
  -loglin : use a log-linear relation. In this case, disp= is
            equal to the velocity width of each bin in km/s .
   disp=d : align such that the Angstroms per pixel is "d".
            (Default is the average pixel width for each row.)
   zero=w : specifies that the spectra should be aligned such
            the wavelength "w" would be at the center of a bin.
            (Default value is 3000.)
     -1d  : Write out one dimensional spectra, i.e. NAXIS=1.
  -nozero : Ignore zero values, if any pixel within the new
            bin is zero, the new bin flux will be zero.
    ef=   : Error file. Align this data in a consistent matter.
            Any new bin which contains a zero or negative pixel
            will be set to -1.
   row=   : Only produce spectrum file for specified row.
 -listdisp: List default dispersion for all rows, no writing.
   -lgi   : Do lagrangian (area polynomial) interpolation (not
            recommended.)
  table=  : Specify alignment for row(s) in a file.  The format
            is: (row#) (starting wavelength) (dispersion)
            (number of pixels) . One entry per line.