Using the ExtData component: Difference between revisions

Bmauer (talk | contribs)
No edit summary
Bmauer (talk | contribs)
No edit summary
Line 61: Line 61:
If you enter a refresh template to update at a certain time each day, month, etc ... as described above there are a few considerations. When ExtData trieds to fill the field at the time described by the template it will once again apply the current model time to the file template, and then try to find data on file at that time. It will not do time interpolation.
If you enter a refresh template to update at a certain time each day, month, etc ... as described above there are a few considerations. When ExtData trieds to fill the field at the time described by the template it will once again apply the current model time to the file template, and then try to find data on file at that time. It will not do time interpolation.


The most complex case is if you enter '0' for the represh template. In this case the field gets updated every step and ExtData will try to interpolate to that time using the data on file and can interpolate between data points in different files. In order to accomplish this if you select a refresh template of '0' ExtData tries to find two times that bracket the start time the first time ExtData runs. These get updated as time advances in your application. To find the bracketing data when starting ExtData looks for two times on the file that bracket the current time. But how does it know what time to apply to the file template is the data is spread across multiple files. This is where the idea of a file frequency and reference time comes in. You as the user can specify a file frequency either with the 3 optional key words to the primary export or by letting ExtData "guess" from the file template. The frequency tells ExtData to advance the time by the frequency if it can't find what it needs using the current time, apply this new time to the file template, and finally check
The most complex case is if you enter '0' for the refresh template. In this case the field gets updated every step and ExtData will try to interpolate to that time using the data on file and can interpolate between data points in different files. In order to accomplish this if you select a refresh template of '0' ExtData tries to find two times that bracket the start time the first time ExtData runs. These get updated as time advances in your application. But how does it know what time to apply to the file template if the data is spread across multiple files? This is where the idea of a file frequency and reference time comes in. The user can specify a file frequency and a reference time either with the 3 optional key words to the primary export or by letting ExtData "guess" from the file template. The idea is that starting from a reference time and the given frequency ExtData tries to find the bracketing data on the file with the time closest to the current time. If it does not find the data there it checks the next time. The user can specify a reference time and the frequency. If these are not given it tries to guess this from the file template. For example if the last token in the template is an hour then the frequency is an hour and the reference time is whatever the current data is with anything after the hour set to zero. If the file template is myfile_%y4%m2%d2_%h2.nc4 and ExtData will assume the frequency is 1 hour and the reference time is the top of the hour on whatever date you start the application.
=== Masks ===
=== Masks ===


=== Derived Exports ===
=== Derived Exports ===