Re: [ferret_users] @Sum Transformation: Feature (Or bug)?

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

Hi Roman and all,
The variable slat_sum = slat[j=@sum] is still defined as having original 
grid of the underlying variable slat.  I'm going to change to a dataset 
we can all use to give an example

yes? use levitus_climatology
yes? let tsum = temp[x=@sum]

yes? sh grid temp
   GRID GMS1
name       axis              # pts   start                end
XAXLEVITR LONGITUDE          360mr   20.5E                19.5E(379.5)
YAXLEVITR LATITUDE           180 r   89.5S                89.5N
ZAXLEVITR DEPTH (m)           20 i-  0                    5000
normal    T

yes? show grid tsum
   GRID GMS1
name       axis              # pts   start                end
XAXLEVITR LONGITUDE          360mr   20.5E                19.5E(379.5)
YAXLEVITR LATITUDE           180 r   89.5S                89.5N
ZAXLEVITR DEPTH (m)           20 i-  0                    5000
normal    T

yes? say `temp,return=shape`
!-> MESSAGE/CONTINUE XYZ
XYZ

yes? say `tsum,return=shape`
!-> MESSAGE/CONTINUE XYZ
XYZ

The transformed variable you've defined does get To get information 
about a variable you've defined


The `RETURN=` keywords don't evaluate anything; they just go and look at 
the grid of the variable.  So this is a bit of a "feature" I guess. As 
Peter showed, running an "action" command like any plotting command, or 
LOAD or STAT, does apply the transformation, reducing the x dimension in 
the result of the command.  If the data isn't too large, STAT can be 
quite useful for seeing the shape of a variable you've defined.


For a bit more discussion of transformations, take a look at the 
"Expressions" chapter in the Ferret Users Guide,
http://ferret.pmel.noaa.gov/Ferret/documentation/users-guide/variables-xpressions/XPRESSIONS.  
Find Chapter 3, section 2.4 to read about transformations, and in 
particular, look up "transformations, general information" in the index

Ansley

Roman Tonkonojenkov wrote:
> Hi ferreters,
> FYI:
>
>
>
> ---------------- Forwarded Message ----------------
> From: "Peter Szabo" <szabpet83@xxxxxxxxx>
> Date: Wed, Sep 24, 2008 02:46 AM
> Subject: Re: [ferret_users] @Sum Transformation: Feature (Or bug)?
> To: "Roman Tonkonojenkov" <romantonk@xxxxxxx>
>
> Hey,
> i tested your file and turned out that there is no problem with your 
> data or ferret either. I experienced the same
> if you enter after "let slat_sum = slat[i=@sum]"  or "let slat_sum = 
> slat[j=@sum]" the command "fill slat_sum", ferret says
> **ERROR: dimensions improperly specified: must be a 2D region
>           CONTOUR/FILL slat_sum
> SO it changed the dimensions. And works also the "plot" command and 
> the "list slat_sum[j=80]" fine. But that must be a bug that ferret 
> doesnt warn you when you "list slat_sum[i=30,j=80]" or enter "show 
> grid slat_sum" and get the same dimensions as the original variable.
>
> Cheers, Peter
>
>
> On Mon, Sep 8, 2008 at 11:07 PM, Roman Tonkonojenkov 
> <romantonk@xxxxxxx <https://webmail.psu.edu/webmail/blankIe.html#>> wrote:
>
>      Dear Ferret Users,
>     I have discovered that applying the @sum transformation to a
>     variable doesn't change the dimension of the variable in some
>     instances. Do you think this is a feature or a bug? Here is an
>     example. I am using a file test.nc <http://test.nc/> (attached)
>     and here is what I get:
>     ============================================================================
>
>     use test.nc <http://test.nc/>
>      *** NOTE: Ignored modulo length exceeding axis length: XT
>     yes? sh da
>          currently SET data sets:
>         1> ./test.nc <http://test.nc/>  (default)
>      name     title              &nb! sp;              I        
>     J         K         L
>      SLAT     sea level atmospheric temperatu  1:100     1:100    
>     ...       1:1
>      
>     yes? let slat_sum = slat[i=@sum]
>     yes? say `slat_sum, return = shape`
>     XY
>     yes? list slat[i=30,j=80]
>         &nbsp! ;        VARIABLE : sea level atmospheric temperature (K)
>                  FILENAME : test.nc <http://test.nc/>
>                  LONGITUDE: 106.2E(106.2)
>          !         LATITUDE : 53.1N & nbsp;           TIME     :
>     31-DEC-2609 22:12
>               275.8
>     yes? list slat_sum[i=30,j=80]
>                  VARIABLE : SLAT[I=@SUM]
>                  FILENAME : test.nc <http://test.nc/>
>                  LONGITUDE: 104.4E to 108E
>                  LATITUDE : ! 53.1N
>             &nbsp! ; & nbsp;  TIME     : 31-DEC-2609 22:12
>               275.8
>     yes? list slat_sum[j=80]
>               !    VARIABLE : SLAT[I=@SUM]
>                  FILENAME : test.nc <http://test.nc/>
>                  LONGITUDE: 0E(0) to 0E(360)
>                  LATITUDE : 53.1N
>                  TIME     : 31-DEC-2609 22:12
>               27775.