Re: [ferret_users] shade/modulo every time

[Ansley Manke <ansley.b.manke@xxxxxxxx>]

Hi Patrick,
/HLIM is intended to set the axis limits without impacting the context 
of the expression you're plotting. It lets you set up the plot axes 
independent of the data.  It's also a way to handle curvilinear grids 
where specifying /X= isn't the right way to go.   To get what you want, 
for data on a rectilinear grid, it would be

   shade/x=-180:180:30 temp[k=1]

and then the modulo character of the longitudes is applied. We have 
plans to make commands more seamless for users, to make better use of 
the coordinates attributes, so that you won't have to use the 3-argument 
form of the plot commands when  your data is on a curvilinear grid.

For now, perhaps using symbols is a reasonable solution

   if ($curvilienar) then
      define symbol plot_qualifiers /hlim=-180:180:30
      define symbol plot_var = temp[k=1], x_coords, y_coords
   else
      define symbol plot_qualifiers /x=-180:180
      define symbol plot_var = temp[k=1]
   endif

   shade($plot_qualifiers)  ($plot_var)

Ansley


On 3/6/2012 9:00 AM, Brockmann Patrick wrote:
> Hi all,
>
> I am wondering if it could be possible to have the qualifier /modulo
> for the shade command even if I use this command with 1 arg
> (this qualifier is proposed only for the shade 3 args)
>
> For now, the only way to get a correct plot is to specify var with 
> var[x=-180:180]
>
> yes? use levitus_climatology
> yes? shade/hlim=-180:180:30 temp[k=1]
> --> missing some part of the plot
>
> You then must do
> yes? shade/hlim=-180:180:30 temp[k=1,x=-180:180]
>
> But I would like to do:
> yes? shade/hlim=-180:180:30/modulo temp[k=1]
>
> By this way, I can treat in the same way, variables that are with 
> rectlinear grid
> and the ones that are with curvilinear grids.
> In both case, I will use:
> shade/hlim=-180:180:30/modulo var [,...]
>
> I can even think to a transparent way to swith between shade1arg and 
> shade3args.
> Bellow the script that does the switch.
>
> let me know, if this request can be accpeted or if there is a good 
> workaround.
>
> Regards
> Patrick
>
> $ cat CMIP5_shade.jnl
> !======================================
> let allnames = ..varnames
>
> def sym isexist_lon=`IS_ELEMENT_OF_STR (allnames, "lon")`
> def sym isexist_lat=`IS_ELEMENT_OF_STR (allnames, "lat")`
>
> if `($isexist_lon) AND ($isexist_lat)` then
>        set variable/units="degrees_east"/title="Longitude" lon
>        set variable/units="degrees_north"/title="Latitude" lat
>        def sym nbpts=`lon,return=isize`
>        set axis/modulo `lon,return=xaxis`
>        ($01)/modulo/i=1:`($nbpts)+1` ($02),lon,lat
>
> else
>        ($01) ($02)
>
> endif
> !======================================
>
> $ ferret
> yes? go CMIP5_shade.jnl shade/lev=20 var2D