How to open netcdf in reading mode from memory?

[ianliu]

I'm trying to open a NetCDF file from memory with

buffer: bytes = ...
ds = xr.open_dataset("dummy.nc", engine="netcdf4", memory=buffer)

but this fails with

TypeError: NetCDF4BackendEntrypoint.open_dataset() got an unexpected keyword argument 'memory'

According to netCDF4 API docs for Dataset, I can pass a memory keyword argument to create a dataset from memory like so:

from netCDF4 import Dataset
buffer: bytes = ...
ds = Dataset("dummy.nc", memory=buffer)

Is there another way to create a netcdf dataset from memory?

[ianliu]

I have created a pull request to expose the memory argument here: #6956