You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository has been archived by the owner on Sep 1, 2022. It is now read-only.
We are using NcML to apply CF 1.6 discrete sampling geometry conventions to some time series data. This involves removing some variables that have a depth dimension so that we have only featureType="timeSeries" variables in the dataset.
When we try the NcML locally, it works fine, but when we put the NcML in our TDS, it fails to remove the depth-dependent variables.
It seems to be a problem with the order of the dimensions, because if we write the NcML with dimension order (station,time) instead of (time,station) it works.
We are using NcML to apply CF 1.6 discrete sampling geometry conventions to some time series data. This involves removing some variables that have a depth dimension so that we have only featureType="timeSeries" variables in the dataset.
When we try the NcML locally, it works fine, but when we put the NcML in our TDS, it fails to remove the depth-dependent variables.
It seems to be a problem with the order of the dimensions, because if we write the NcML with dimension order (station,time) instead of (time,station) it works.
XML TDS catalog with two identical datasets, but with (time,station) and (station,time)
https://gist.github.com/rsignell-usgs/5484245
NcML: (time,station)
https://gist.github.com/rsignell-usgs/5484269
-Rich
The text was updated successfully, but these errors were encountered: