Re: error message

From: Mary Haley <haley_at_nyahnyahspammersnyahnyah>
Date: Wed Oct 13 2010 - 08:53:12 MDT

Dear Fan,

I believe this error can be safely ignored, even though it claims it is fatal. Internally, NCL is trying to use the nice_mnmxintvl function to calculate a nice min and max for that third Y axis, and because your data is constant, it issues an error.

I will fix gsn_csm_xy3 so that the error is not printed, but this isn't going to change the end result of your script. Internally, NCL is using the trYMinF and trYMaxF that you have set for that third Y axis. I assume that the script is working as expected? I got what looked like a valid plot from running it.

--Mary

On Sep 29, 2010, at 4:34 PM, David Brown wrote:

> Hi Fan,
> Does this involve a contour plot? You say it is a time series? Do you
> mean there are several plots produced? Anyway, normally if you have
> constant data you should get a message saying to the effect no contour
> plot possible, but somehow your data is slipping through without
> getting this message. What are the actual min and max values and what
> is the missing value? We may need to see the actual data and your
> script to properly diagnose this issue.
> -dave
>
> On Sep 29, 2010, at 2:24 PM, Fang, Fan (GSFC-610.2)[ADNET SYSTEMS INC]
> wrote:
>
>> I don't seem to be able to find this topic in your archive but I got
>> an error message from an NCL script:
>>
>> fatal:_NhlGetEndpointsAndStepSize: Max value less than or equal to
>> min value
>>
>> I was plotting a time series and the data were constant with missing
>> values. How to avoid or fix the error?
>>
>> Thanks.
>>
>> -Fan
>> _______________________________________________
>> ncl-talk mailing list
>> List instructions, subscriber options, unsubscribe:
>> http://mailman.ucar.edu/mailman/listinfo/ncl-talk
>
> _______________________________________________
> ncl-talk mailing list
> List instructions, subscriber options, unsubscribe:
> http://mailman.ucar.edu/mailman/listinfo/ncl-talk

_______________________________________________
ncl-talk mailing list
List instructions, subscriber options, unsubscribe:
http://mailman.ucar.edu/mailman/listinfo/ncl-talk
Received on Wed Oct 13 08:53:16 2010

This archive was generated by hypermail 2.1.8 : Tue Oct 19 2010 - 14:38:00 MDT