NullPointerException during aggregation

[Ken Knapp]

Hello-
I was about to ask the same question.
To expand on your problem, when aggregating 8 files, I encounter the 
error only when accessing the last file.
That is, an ncdump in time of 0:6:1 has no error, but 0:7:1 causes the 
error.
Thinking the eighth file was the problem, I removed it and got the error 
when accessing the last file again. It appears that there is some 
problem when trying to access the last file in the aggregation?
Funny thing is the last file is being read properly (at least in part) 
because the time variable is properly assigned by grabbing it from all 8 
files, but any of the other variables from the 8th file causes problems.
-Ken Knapp

NCML file:

<netcdf xmlns="http://www.unidata.ucar.edu/namespaces/netcdf/ncml-2.2">
  <aggregation dimName="time" type="joinExisting">
    <scan location="/home/kknapp/java/data/" suffix=".GPC"/>
  </aggregation>
</netcdf>



Jon Blower wrote:
> Hi all,
>
> I'm having problems working with NcML aggregations and the 2.2.16
> version of the Java NetCDF libraries.  I would like to aggregate
> files:
>
> NATL_1_28082006.nc
> NATL_1_29082006.nc
> NATL_1_30082006.nc
> ...
>
> along the time dimension.  I tried an NcML file like this:
>
> <netcdf xmlns="http://www.unidata.ucar.edu/namespaces/netcdf/ncml-2.2">
>   <aggregation dimName="time" type="joinNew">
>     <variableAgg name="temperature"/>
>     <variableAgg name="salinity"/>
>     <scan location="/path/to/files/" suffix=".nc"
> dateFormatMark="NATL_1_#ddMMyyyy" />
>   </aggregation>
> </netcdf>
>
> This gave me a NullPointerException:
>
> java.lang.NullPointerException
>     java.util.Calendar.setTime(Calendar.java:1032)
>     java.text.SimpleDateFormat.format(SimpleDateFormat.java:785)
>     java.text.SimpleDateFormat.format(SimpleDateFormat.java:778)
>     java.text.DateFormat.format(DateFormat.java:314)
>     ucar.nc2.units.DateFormatter.toDateTimeStringISO(DateFormatter.java:184) 
>
>     ucar.nc2.ncml.Aggregation.crawlDirectory(Aggregation.java:817)
>
> I think I know why this is.  There are other files in the same
> directory of the form "NATL_120_28082006.nc" and I think the
> aggregator is trying to match these.  Can I arrange it so that only
> files that match the pattern "NATL_1_#ddMMyyyy.nc" are aggregated, and
> all others are simply ignored?
>
> I am reluctant to change the directory structure of my data because
> other applications may depend on it.
>
> Thanks in advance,
> Jon
>

-- 
Ken Knapp, Ph.D.      Ken.Knapp at noaa.gov
Remote Sensing and Applications Division
National Climatic Data Center
151 Patton Ave
Asheville, NC 28801-5001
828-271-4339 (voice) 828-271-4328 (fax)

==============================================================================
To unsubscribe netcdf-java, visit:
http://www.unidata.ucar.edu/mailing-list-delete-form.html
==============================================================================