[Thread Prev][Thread Next][Index]

Re: [ferret_users] ..varnames



Hi Patrick,
Thanks for the report, and the example. You're right; this is not correct. It's not listing the coordinate variables themselves, which would be XAXLEVITR, etc.; but it is listing the variables that describe the coordinate bounds. In fact what should happen is that the bounds variables XAXLEVITR_bnds and so forth, should be among the dimensions in the header of the file.

Compare what happens if you save the file with the /EDGES qualifier instead of with the /BOUNDS qualifier:

yes? save/file=tyty_edge.nc/clobber/edges temp[k=1]

In this case, I think everything behaves as you would expect. We will correct this in the next version.

Ansley

Brockmann Patrick wrote:
Hi all,

I have found that ..varnames returns also bounds coordinate variables.
Try:

yes? use levitus_climatology.cdf
yes? list ..varnames
           VARIABLE : ..VARNAMES
            FILENAME : levitus_climatology.cdf
            SUBSET   : 2 points (X)
1   / 1:"TEMP"
2   / 2:"SALT"
yes? save/file=tyty.nc/clobber/bounds temp[k=1]
LISTing to file tyty.nc

yes? use tyty.nc
yes? list ..varnames
            VARIABLE : ..VARNAMES
            FILENAME : tyty.nc
            SUBSET   : 4 points (X)
1   / 1:"XAXLEVITR_bnds"
2   / 2:"YAXLEVITR_bnds"
3   / 3:"ZAXLEVITR1_1_bnds"
4   / 4:"TEMP"

That is not what I was expecting from command yes? list ..varnames
since I consider bound variables impied by the definition
of the bounds attribut as coordinate variables.

Let me know if this can be corrected in a next release.

tested with FERRET v6.2 on a x86_64

Thanks

Patrick



[Thread Prev][Thread Next][Index]

Contact Us
Dept of Commerce / NOAA / OAR / PMEL / TMAP

Privacy Policy | Disclaimer | Accessibility Statement