Andy -
I got this working by adding the "standard_name = station_id" attribute to the
"station_name" variable.
Use this (change path to NetCDF file as needed):
<?xml version="1.0" encoding="UTF-8"?>
<netcdf xmlns="http://www.unidata.ucar.edu/namespaces/netcdf/ncml-2.2"
location="C:/Users/kwilcox/Desktop/watlev_NOAA.F.C_IKE_VIMS_3D_WITHWAVE.nc">
<variable name="station_name">
<attribute name="standard_name" value="station_id" />
</variable>
</netcdf>
-----Original Message-----
From: Andy Bird
Sent: Monday, December 19, 2011 11:30 AM
To: Ethan Davis; Kyle Wilcox
Cc: Alexander Crosby; John Caron
Subject: RE: Netcdf CDM and ragged arrays
Hi Ethan,
Thank you for getting back to me. Please find attached a small example of an
IMEDS file netcdf ragged array. I can provide many more links via opendap if
required.
The data is typically multi station time series data. Alex can provide the
detailed information on the netcdf construction and outline if needed.
Best,
Andy
-----Original Message-----
From: Ethan Davis [mailto:address@hidden]
Sent: Friday, December 16, 2011 2:31 PM
To: Kyle Wilcox
Cc: Andy Bird; Alexander Crosby; John Caron
Subject: Re: Netcdf CDM and ragged arrays
Hi Andy,
Yes, a sample would be great. What type of data are you working with?
Point data, soundings, etc?
I'm CCing John Caron who wrote the FeatureDataset code. I believe there are
some fixes in 4.3 but I'll let John speak to that in more detail.
Thanks,
Ethan
On 12/16/2011 10:36 AM, Kyle Wilcox wrote:
Andy -
Can you send along the NetCDF file you are working with?
-----Original Message-----
*From:* Andy Bird [address@hidden]
*Received:* Friday, 16 Dec 2011, 11:57am
*To:* address@hidden [address@hidden]
*CC:* Alexander Crosby [address@hidden]; Kyle Wilcox
[address@hidden]
*Subject:* Netcdf CDM and ragged arrays
Hi Ethan,
I was wondering if you could help with a small problem we are having,
we are trying to update one of our THREDDS services to work with
ragged arrays but are having some issues with the
FeatureDatasetFactoryManager.wrap command returning a null object when
parsing the netcdf feature type information. We are using version 4.2
of the netcdf tools. Kyle's thoughts were that some of the issues we
were having might be fixed in version 4.3??
Happy to provide more information if needed.
Andy Bird
*Andrew Bird *| Software Engineer
*RPS | ASA*| *Science. Services. Solutions.*
55 Village Square Drive | South Kingstown, RI 02879 USA
p: +1 401 789-6224 ex. 372 *| *f: +1 401 789-1932**
e: address@hidden<mailto:address@hidden> |
www.asascience.com<http://www.asascience.com>