Alexander Barth

If you want to propose an interface for using a different value for _FillValue, this would be helpful. Currently one need to do this: ```julia ds = NCDataset("file.nc"); float_var =...

And reading a subset, should this be like `read(ds["float_var"],:,:,1; fillvalue=NaN)` which is a bit similar to `view(array,:,:,1)` ? For information, there is also the experimental/un-exported in-place `NCDatasets.load!` function: ```julia #...

Maybe this is helpful to get the home directory on Windows: https://stackoverflow.com/a/42700844 I am wondering if NetCDF can also look for `USERPROFILE` if `HOME` is undefined (on Windows)? https://github.com/Unidata/netcdf-c/blob/93742f8e7b77425400ceeb1791766bcd5b5ecb59/libdispatch/ddispatch.c#L79 (I...

Support for parallel NetCDF I/O would indeed be nice. Do you have any complete known-working C code example? I tried the Ubuntu 20.04 package libnetcdf-mpi-dev with [nc4_pnc_put.c](http://cucis.ece.northwestern.edu/projects/PnetCDF/Examples/nc4_pnc_put.c) But this fails...

Concerning ` v(:, :, 1:5)` as an alias of `view(v,:,:,1:5)`: I like follow the approach [There should be one-- and preferably only one --obvious way to do it.](https://www.python.org/dev/peps/pep-0020/). To me...

Yes, ~~DickArrays.jl~~DiskArrays.jl is exactly what I head in mind. It is great to see that it already exists! Thanks @visr ! One of the reason that I am reluctant to...

> Nice typo in your first sentence ;) Haha, what a wired thing to have in one's mind! ;-) > To be honest, I have not done an extensive evaluation...

@meggart, thanks for pointing out the type-stability which is indeed a concern. I am currently working on a [branch](https://github.com/Alexander-Barth/NCDatasets.jl/tree/return-type) where I requite that all attributes affecting the type of the...

I think I have now a solution to make it type stable. Essentially, all variables affecting the element-type are now part of the `CFVariable` via a named tuple (called `_storage_attrib`)....

@Balinus, you can keep the singleton dimension if you read the full variable, like `Array(v)` or `v[:,:,:]` (or even `v[:,:,1:1]` for a single time slice). Here the behavior is identical...