Selective variable reads in open_dataset

[jhamman]

One of the beautiful things about the netCDF data model is that the variables can be read individually. I'm suggesting adding a variables keyword (or something along those lines) to the open_dataset function to support selecting one or more or all variables in a file. This will allow for faster reads and smaller memory usage when the full set of variables is not needed.

[shoyer]

xray already only loads array data into memory when accessed. You can call open_dataset on a 100GB NetCDF file and none of the array data will be loaded into memory until try to do something with a variable's values. Do you work with NetCDF files with so many variables that just loading the metadata for all variables is prohibitively slow?

[jhamman]

Alright, my mistake. That's a nice feature.

I do work with some large netCDF files but the loading of the metadata is probably a miniscule fraction of the total read.