Hi there,

Thank you for creating the set of tools, it's great to have python tools for post-processing CROCO output. I have run into a few issues with the vslice function.

Here is the code snippet:

Setting the correct paths (where the data is stored)

import numpy as np
import xarray as xr
from glob import glob
import xcroco as xco

#Loading CROCO files
model_path = glob('/media/matthew/Cerberus1/Run_False_Bay_2008_2018_SANHO/croco_avg_Y2008M1.nc.1')
grid_path = '/home/matthew/confs/Run_False_Bay/CROCO_FILES/croco_grd.nc.1'
D = xco.croco_dataset(model_path, grid=grid_path)
#Compute depths
z = D.z_rho.compute()
z = D.z_w.compute()
z = D.z_u.compute()
z = D.z_v.compute()

Extract section

Npoints = 40
lat_ends = [-33.53,-34.55]
lon_ends = [19.0,19.5]
lats = np.linspace(lat_ends[0], lat_ends[1], Npoints)
lons = np.linspace(lon_ends[0], lon_ends[1], Npoints)
sec = xr.Dataset()
sec['temp'] = xco.vslice(D, D.temp, lat_rho=lats, lon_rho=lons).compute().load()
sec['temp_interp'] = xco.vslice(D, D.salt, lat_rho=lats, lon_rho=lons, zlevels=np.arange(-500,0.1)).compute().load()
sec

#Error
I get the following error:
TypeError: Using a DataArray object to construct a variable is ambiguous, please extract the data using the .data property.

I have managed to get the section by adding .data at various points in the source code but this is obviously not ideal. Can you suggest any fixes?