[Thread Prev][Thread Next][Index]

Re: Use of 'X' with Contour in draw_2d.tmpl in LAS 6.3



Steve,

The first thing to do is try to do your contouring in Ferret rather than through LAS so we can isolate the problem.  Will Ferret 5.3 give you an XZ contour plot of your data with the following?
use ~dataset~
set region ...
fill ~variable~,  'X', depth_at_sigma
For the preceding to work, the dataset must have the following variables defined on the same grid:  ~variable~, X, depth_at_sigma.  Note that Ferret treats anything in single quotes as case sensitive whereas Ferret is normally case insensitive. 

If the commands above don't work then we've narrowed things down a lot.

If the simple commands above work then I'd try going into the las/server/ directory, making sure you have the same Ferret environment variables as your LAS, copying the /tmp/lasgo31108960517.jnl file to bug_a.jnl and then copying the Ferret command lines from las/server/log/debug to bug.jnl and modify it to reference bug_a.jnl.

Then you can start Ferret and run exactly the same session that LAS was running.  You will reproduce the same error and can comment out lines in your bug.jnl and bug_a.jnl scripts until you get something that does work.  In the process you should find the offending lines in the script and we can figure out what's going wrong from there.

Good luck sleuthing this one.


-- Jon


Steve Cousins wrote:
Hi again,

I just thought I'd check again to see if anybody has any ideas about being
able to customize the draw_2d.tmpl file in LAS 6.3?  Is there a reason
that this should work in LAS 5 but not in 6?

It boils down to this line:

   [% ELSIF args.view == 'xz' %] 
     [% args.fill_type %]/levels=[% args.fill_levels %]/set [% args.variable_name.0 %],'X',depth_at_sigma

where I've added ,'X',depth_at_sigma to the CONTOUR line.  The error (all
details are below) comes up as:

  yes? GO "/tmp/lasgo31108960517.jnl"
   **ERROR: variable unknown or not in data set: 'X'
  CONTOUR/FILL/levels=30/set no3,'X',depth_at_sigma
  Command file, command group, or REPEAT execution aborted

Thanks,

Steve 

______________________________________________________________________
 Steve Cousins, Ocean Modeling Group    Email: cousins@umit.maine.edu
 Marine Sciences, 208 Libby Hall        http://rocky.umeoce.maine.edu
 Univ. of Maine, Orono, ME 04469        Phone: (207) 581-4302

On Mon, 12 Jul 2004, Steve Cousins wrote:

  
Thanks Joe (and Cyndy).  This has allowed me to continue but I now get a
different error:

LAS Error

 The following error message was received from LAS: 

 Adding an acceptable error string: "*** NOTE: ".
         NOAA/PMEL TMAP
         FERRET v5.53  
         Linux(g77) 2.4.x - 09/26/03
         12-Jul-04 11:46     

 yes? cancel mode verify
 yes? cancel mode interp
 yes? define symbol refmap_xhi = 290.071990966797
 yes? define symbol refmap_yhi = 10.030584335327148
 yes? define symbol land_type = shade
 yes? define symbol refmap_xlo = 99.7431030273438
 yes? define symbol jnl_t = ARRAY(0x8cde450)
 yes? define symbol refmap_view = xline
 yes? define symbol refmap_ylo = 10.030584335327148
 yes? define symbol variable_name_0 = no3
 yes? define symbol fill_type = fill
 yes? define symbol jnl_x = ARRAY(0x8cde2c4)
 yes? define symbol variable_name = ARRAY(0x886064c)
 yes? define symbol jnl_y = ARRAY(0x8cde3fc)
 yes? define symbol jnl_z = ARRAY(0x8cde414)
 yes? define symbol diag_direction = down
 yes? define symbol dataset_name_0 =
/usr2/fer_dsets/descr/pacific-roms-90-01.des
 yes? define symbol dataset_name = ARRAY(0x8860670)
 yes? define symbol rank = 2
 yes? canc data/all
 yes? go std_initialize "/usr2/fer_dsets/descr/pacific-roms-90-01.des" "1"
"1" "no3"
 yes? set
region/x="99.7431030273438":"290.071990966797"/y="10.030584335327148"/z="1":"20"/l="1":"1"
 yes? set win/size=.5
 yes? GO "/tmp/lasgo31108960517.jnl"
  **ERROR: variable unknown or not in data set: 'X'
 CONTOUR/FILL/levels=30/set no3,'X',depth_at_sigma
 Command file, command group, or REPEAT execution aborted


I'm just adding the 'X', and depth_at_sigma variables to the end of the
CONTOUR line.  'X' is supposed to refer to the X dimension variable as I
understand it.  It works this way in LAS 5.  

Any ideas?

Thanks,

Steve



______________________________________________________________________
 Steve Cousins, Ocean Modeling Group    Email: cousins@umit.maine.edu
 Marine Sciences, 208 Libby Hall        http://rocky.umeoce.maine.edu
 Univ. of Maine, Orono, ME 04469        Phone: (207) 581-4302

On Sun, 11 Jul 2004, Joe McLean wrote:

    
Hi Steve,

As of LAS v6.0 template access to args.variable_name refers to an array 
rather than a string variable .  Easy fix though.
You'll notice in 'draw_2d.tmpl' which you sent that most references to 
the variable name are in the following syntax:  args.variable_name.0

There is still one place in your template which refers to 'args.variable_name' instead of 'args.variable_name.0'. Make this change and all should be well.

Joe
---------------------------
[...]
! SDC: Dealing with Sigma (actually S-coordinates in this case) 
[...] 
   DEFINE AXIS/Z=-5000.0:0.0:10.0/UNIT=meters/DEPTH zdepth 
!   LET varonsigma= ZAXREPLACE([% args.variable_name%],depth_x_sigma,z[gz=zdepth])
   LET varonsigma= ZAXREPLACE([% args.variable_name%],depth_at_sigma,z[gz=zdepth])
   LET var_at_depth = varonsigma[ [% args.z_lo %] ]
[...]
   let var_kend = `var_at_depth,RETURN=KEND`
[% END %]
--------------------------


Steve Cousins wrote:

      
Hi All,

In LAS 5 I was able to edit the draw2d.tmpl file to get LAS to show data
correctly that is on a curvilinear grid and with Sigma coordinate depth.
I'm now trying to get the Sigma depth part to work in LAS 6.3.  When I
try it though, it produces:

----------------------------------------------------------------------

LAS Error

The following error message was received from LAS: 

Adding an acceptable error string: "*** NOTE: ".
        NOAA/PMEL TMAP
        FERRET v5.53  
        Linux(g77) 2.4.x - 09/26/03
         9-Jul-04 18:03     

yes? cancel mode verify
yes? cancel mode interp
yes? define symbol refmap_xhi = 290.0
yes? define symbol refmap_yhi = -17.0
yes? define symbol land_type = shade
yes? define symbol refmap_xlo = 100.0
yes? define symbol jnl_t = ARRAY(0x8cde4c4)
yes? define symbol refmap_view = xline
yes? define symbol refmap_ylo = -17.0
yes? define symbol variable_name_0 = no3
yes? define symbol fill_type = fill
yes? define symbol jnl_x = ARRAY(0x8cde488)
yes? define symbol variable_name = ARRAY(0x88605dc)
yes? define symbol jnl_y = ARRAY(0x8cde4b8)
yes? define symbol jnl_z = ARRAY(0x8cde2b4)
yes? define symbol diag_direction = down
yes? define symbol dataset_name_0 =
/juno/usr4/lshi/RESULT_NP_1990_2001_NCEP_daily/NP_avg.001.nc
yes? define symbol dataset_name = ARRAY(0x8860600)
yes? define symbol rank = 2
yes? canc data/all
yes? go std_initialize
"/juno/usr4/lshi/RESULT_NP_1990_2001_NCEP_daily/NP_avg.001.nc" "1" "1"
"no3"
yes? set
region/x="100.0":"290.0"/y="-17.0"/z="1":"20"/t="01-Jan-1990":"01-Jan-1990"
yes? set win/size=.5
yes? GO "/tmp/lasgo223198511390.jnl"
 **ERROR: command syntax:
ZAXREPLACE(ARRAY(0x88605dc),depth_at_sigma,z[gz=zdepth])
          RRAY(0x88
DEFINE VARIABLE varonsigma=ZAXREPLACE(ARRAY(0x88605dc),depth_at_sigma,z[gz=zdepth])
Command file, command group, or REPEAT execution aborted

----------------------------------------------------------------------

The thing that sticks out is the ARRAY(0x88605dc).  I hadn't seen this
before. The /tmp/lasgo... file looks like:

----------------------------------------------------------------------

  can view

  define view/x=0,1/y=0.25,1 vt
  define view/x=0.029,.975/y=0,0.25 vb

  set view vt

  set MEMORY/SIZE=20.0

  DEFINE AXIS/Z=-5000.0:0.0:10.0/UNIT=meters/DEPTH zdepth 
  LET varonsigma=ZAXREPLACE(ARRAY(0x88605cc),depth_at_sigma,z[gz=zdepth])
  LET var_at_depth = varonsigma[  ]

  let var_kend = `var_at_depth,RETURN=KEND`

  go preplot_setup
    
  fill/levels=30/set no3,'X',depth_at_sigma
   
  go postplot_setup
  palette default
    
  ppl title,0.15,averaged nitrate (mMol N m-3)
  ppl fill
  ppl %range 'PPL$ZMIN' 'PPL$ZMAX' 30
  ppl set PPL$RANGE_INC2 'PPL$RANGE_INC' * 2

contour/over/levels=('PPL$RANGE_LOW','PPL$RANGE_HIGH','PPL$RANGE_INC2')/nolabel no3

  set view vb
  ppl origin ,.7
  ppl axlen ,2

----------------------------------------------------------------------

In order to compare it with my LAS 5 server, I put a typo in the
draw_2d.tmpl file so the /tmp/lasgo... file would not be deleted.  Here is
what I get with LAS 5:

----------------------------------------------------------------------

  can view

  define view/x=0,1/y=0.25,1 vt
  define view/x=0.029,.975/y=0,0.25 vb

  set view vt

  set MEMORY/SIZE=20.0
     
  DEFINE AXIS/Z=-5000.0:0.0:10.0/UNIT=meters/DEPTH zdepth 
  LET varonsigma= ZAXREPLACE(temp,depth_at_sigma,z[gz=zdepth])
  LET var_at_depth = varonsigma[  ]

  let var_kend = `var_at_depth,RETURN=KEND`

   go preplot_setup
    
   fill/levels=30/set temp,'X',depth_at_igma
   
   go postplot_setup
   palette default

   ppl title,0.2,temp
   
   ppl fill
   ppl %range 'PPL$ZMIN' 'PPL$ZMAX' 30
   ppl set PPL$RANGE_INC2 'PPL$RANGE_INC' * 2
	
   contour/over/levels=1/nolabel temp,'X',depth_at_sigma
   set view vb

   ppl origin ,.7
   ppl axlen ,2

----------------------------------------------------------------------

So you see that it doesn't use ARRAY(0x88605cc) in the ZAXREPLACE line.

Is this the problem?  Ferret doesn't know what ARRAY(0x88605cc) is?  Maybe
I'm going down the wrong path.  Just to make this email even longer, I'll
include the draw_2d.tmpl file:

----------------------------------------------------------------------

[% DEFAULT
   args.fill_type = 'fill'
   args.fill_levels = '30'
   args.palette = 'default'
   args.do_shade = 1
   args.do_contour = 0
   args.do_overlay = 0
%]


[% IF args.fill_type == 'default';
   args.fill_type = 'fill' ;
ELSIF args.fill_type == 'none';
  args.do_shade = 0 ;
END %]

! Support for <contour_levels>none</contour_levels>
! Must be overridden for overlay plots.
[% IF args.contour_levels == 'none';
   args.do_contour = 0 ;
   args.contour_levels = '' ;
END %]



! NetCDF strides should be used only for XY views
! NetCDF strides should not be used for comparison.
[% IF args.view == 'xy' && args.variable_name.0 != 'diff_' &&
args.variable_name.0 != 'data1_' && args.variable_name.0 != 'data2_' %]
   go stride_create_symbols [% args.variable_name.0 %] 300 200
[% END %]

! SDC: Dealing with Sigma (actually S-coordinates in this case) 

[% IF args.view == 'xy' %]
     [% needz = 'NO' %]
  [% ELSIF args.view == 'xz' %]
     [% needz = 'YES' %]
  [% ELSIF args.view == 'yz' %]
     [% needz = 'YES' %]
  [% ELSE %]
     [% needz = 'NO' %]
[% END %]
   
[% IF needz == 'YES' %]
  set MEMORY/SIZE=20.0

!   let sigma = z[gz=zpos]
!   let depth_x_sigma = sigma * depth
     
  DEFINE AXIS/Z=-5000.0:0.0:10.0/UNIT=meters/DEPTH zdepth 
!   LET varonsigma= ZAXREPLACE([% args.variable_name
%],depth_x_sigma,z[gz=zdepth])
  LET varonsigma= ZAXREPLACE([% args.variable_name
%],depth_at_sigma,z[gz=zdepth])
  LET var_at_depth = varonsigma[ [% args.z_lo %] ]

  let var_kend = `var_at_depth,RETURN=KEND`
[% END %]

[% IF args.do_shade %]
   go preplot_setup
   [% IF args.view == 'xy' && args.variable_name.0 != 'diff_' &&
args.variable_name.0 != 'data1_' && args.variable_name.0 != 'data2_' %]
       [% args.fill_type %]/levels=[% args.fill_levels %]/set [%
args.variable_name.0 %][i=($STRIDE_irange),j=($STRIDE_jrange)]
   [% ELSIF args.view == 'xz' %] 
       [% args.fill_type %]/levels=[% args.fill_levels %]/set [%
args.variable_name.0 %],'X',depth_at_sigma
   [% ELSIF args.view == 'yz' %]
       [% args.fill_type %]/levels=[% args.fill_levels %]/set [%
args.variable_name.0 %],'Y',depth_at_sigma    
   [% ELSE %]
       [% args.fill_type %]/levels=[% args.fill_levels %]/set [%
args.variable_name.0 %]
   [% END %]
   go postplot_setup
   palette [% args.palette %]
   [% IF args.title %] 
       ppl title,0.15,[% args.title %]
   [% END %]
   [% INCLUDE labels.tmpl args = args %]
   ppl [% args.fill_type %]
[% END %]

[% IF args.do_contour %]
   [% IF args.do_shade %]
	ppl %range 'PPL$ZMIN' 'PPL$ZMAX' 30
	ppl set PPL$RANGE_INC2 'PPL$RANGE_INC' * 2
	[% IF ! args.contour_levels %]
	    [% args.contour_levels =
"('PPL\$RANGE_LOW','PPL\$RANGE_HIGH','PPL\$RANGE_INC2')" %]
	[% END %]
   [% ELSE %]
	[% IF ! args.contour_levels %]
	    [% args.contour_levels = "20" %]
	[% END %]
   [% END %]
   [% IF args.view == 'xy' && args.variable_name.0 != 'diff_' &&
args.variable_name.0 != 'data1_' && args.variable_name.0 != 'data2_' %]
       contour/over/levels=[% args.contour_levels %]/nolabel [%
args.variable_name.0 %][i=($STRIDE_irange),j=($STRIDE_jrange)]
   [% ELSE %]
       contour/over/levels=[% args.contour_levels %]/nolabel [%
args.variable_name.0 %]
   [% END %]
[% END %]

[% IF args.do_overlay %]
   [% IF ! args.contour_levels %]
       [% args.contour_levels = '20' %]
   [% END %]
   [% IF args.view == 'xy' && args.variable_name.0 != 'diff_' &&
args.variable_name.0 != 'data1_' && args.variable_name.0 != 'data2_' %]
       contour/over/levels=[% args.contour_levels %]/nolabel [%
args.overlay_variable_name.0 %][i=($STRIDE_irange),j=($STRIDE_jrange)]
   [% ELSE %]
       contour/over/levels=[% args.contour_levels %]/nolabel [%
args.overlay_variable_name.0 %]
   [% END %]

[% END %]
----------------------------------------------------------------------

Unfortunately it doesn't wrap very well.  If you want me to just attach it
let me know.

Can anyone tell me what the problem is?

Thanks very much and have a good weekend.

Steve
______________________________________________________________________
Steve Cousins, Ocean Modeling Group    Email: cousins@umit.maine.edu
Marine Sciences, 208 Libby Hall        http://rocky.umeoce.maine.edu
Univ. of Maine, Orono, ME 04469        Phone: (207) 581-4302



 

        



    

  

[Thread Prev][Thread Next][Index]

Dept of Commerce / NOAA / OAR / PMEL / TMAP
Contact Us | Privacy Policy | Disclaimer | Accessibility Statement