Re: bug in interpreter?

[Ansley Manke <Ansley.B.Manke@xxxxxxxx>]

Hi Hein,
I agree there seems to be a problem in parsing the definition
of dtdz_anom when Ferret goes to plot it.  Here's a workaround
that you can use until we track down what is happening. You
have the three definitions:

  LET dtdz_month = dtdz[gt=hope_month]
  LET dtdz_clim  = dtdz_month[gt=month_reg@mod]
  LET  dtdz_anom = dtdz_month - dtdz_clim[gt=dtdz_month@asn]

Note that in the definition of dtdz_anom, you regrid dtdz_clim to the
grid of dtdz_month. The time axis of dtdz_month is hope_month, so
you could replace

dtdz_clim[gt=dtdz_month@asn]   with    dtdz_clim[gt=hope_month@asn]

Now you have a simpler definition.  In a similar example I created,
this fixes the problem.

Ansley Manke

Hein Zelle wrote:

> Very strange ferret behaviour detected. A colleague of mine confirms
> that he ran into the same problem before, and thinks it is due to too
> many variables, a too long script, or something similar. Apparently
> rearranging variables or removing some can fix it.
>
> The whole script follows below, I'm just pasting three variable
> definitions here:
>
> let dtdz_month=dtdz[gt=hope_month]
> let dtdz_clim=dtdz_month[gt=month_reg@mod]
> let dtdz_anom = dtdz_month - dtdz_clim[gt=dtdz_month@asn]
>
> Follows the output from ferret when I run the script.
>
> yes? go upwelling-budget
>   (script runs, no errors visible)
> yes? set region/x=100w
> yes? plot dtdz_anom
>  **ERROR: command syntax: DTDZ_CLIM[GT=D                (error, no plot)
> yes? plot dtdz_month                                    (no error, plot
> ok)
> yes? plot dtdz_anom                                     (no error, plot
> ok)
> yes?
>
> I'm not doing anything fancy in the definition of dtdz_anom or
> dtdz_month, they just reference a bunch of other variables. No
> self-generating code or anything :)
>
> Does anyone have an idea what I am running into here? It could be
> anything from a buffer overrun to a limit in the number of variables
> reached. Any ideas/solutions are appreciated.
>
>          Hein zelle
>
> follows the complete upwelling-budget.jnl, the datafile nieuw1.nc is
> rather huge so I'm not sending it.
>
>  ! NOAA/PMEL TMAP
>  ! FERRET v5.50
>  ! Linux 2.4.3-12smp - 01/15/03
>  ! 02-May-03 13:14
>
> ! pacific region
> go pac
>
> ! axes
> use  climatological_axes
> cancel data 1
> define axis /t0=15-dec-1989 /t=1:120:1 /unit=months hope_month
>
> ! data files
> use nieuw1.nc
>
> ! define variables
>
> ! mixed layer depth
> let mixed = if (k eq 1) then 0 else pt_3[d=1,k=1] - pt_3[d=1]
>
> let upwelling_weights=wv * mixed[z=@weq:0.5]
> let upwelling=upwelling_weights[z=@sum]
>
> ! climatology
> let wv_month=upwelling[gt=hope_month]
> let wv_clim=wv_month[gt=month_reg@mod]
> let wv_anom = wv_month - wv_clim[gt=wv_month@asn]
> let climfold=wv_clim[gt=wv_month@asn]
>
> ! dtdz, + climatology
> let dtdz = 20/pt_3[z=@loc:20]
>
> ! climatology
> let dtdz_month=dtdz[gt=hope_month]
> let dtdz_clim=dtdz_month[gt=month_reg@mod]
> let dtdz_anom = dtdz_month - dtdz_clim[gt=dtdz_month@asn]
> let dtdz_climfold=dtdz_clim[gt=dtdz_month@asn]
>
> ! sst budget
> let wbartbar = climfold*dtdz_climfold
> let wbartprime = climfold*dtdz_anom
> let wprimetbar = wv_anom*dtdz_climfold
> let wprimetprime = wv_anom*dtdz_anom
> let primesum = wprimetbar + wprimetprime
>
> >-----------------------------------------------------------------<
>     Hein Zelle
>     Dept. of Oceographic Research
>     KNMI, The Netherlands
>     work:        zelle@knmi.nl     http://www.knmi.nl/~zelle
>     private:     hein@icce.rug.nl http://www.icce.rug.nl/~hein
>     Phone:       +31 (0)30 2206704
> >-----------------------------------------------------------------<
> --------------------------------------------------------------
> Zie ook/see also: http://www.knmi.nl/maildisclaimer.html

--
Ansley Manke  Pacific Marine Environmental Laboratory  Seattle WA  (206)526-6246