We need to finish #52.
The logical thing to do next is to create a min/max/ttl patch
to ncwa. The current patch allows user's to find the min/max/ttl
in a timeseries, or between files, for a given gridpoint.
What is often scientifically more interesting is the min/max/ttl
for a given region. This is what ncwa is setup to do.
User's specify the dimensions and hyperslabs over which to
average. As you did with ncra/ncea, the goal is to generalize
so that now averaging is the default, but not the only, operation.
A word of caution. The code that does the averaging for
ncwa is much more complex than the ncra averaging because
now there are arbitrary numbers of dimensions involved rather
than just a single dimension. So hacking it may require
more patience. If you choose to do this then I recommend
adding torture tests to nco_tst.sh before submitting a patch.
Thanks,
Charlie
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hi Charlie,
what do you recommend I tackle now on you TODO list ?
Henry
Henry,
We need to finish #52.
The logical thing to do next is to create a min/max/ttl patch
to ncwa. The current patch allows user's to find the min/max/ttl
in a timeseries, or between files, for a given gridpoint.
What is often scientifically more interesting is the min/max/ttl
for a given region. This is what ncwa is setup to do.
User's specify the dimensions and hyperslabs over which to
average. As you did with ncra/ncea, the goal is to generalize
so that now averaging is the default, but not the only, operation.
A word of caution. The code that does the averaging for
ncwa is much more complex than the ncra averaging because
now there are arbitrary numbers of dimensions involved rather
than just a single dimension. So hacking it may require
more patience. If you choose to do this then I recommend
adding torture tests to nco_tst.sh before submitting a patch.
Thanks,
Charlie