Start() doesn't interpolate if lat='all' , lon='all'

Hi @nperez and @aho,

I'm trying to interpolate with Start() using lat='all' and lon='all', but it returns the original dimensions. It happens for both retrieve=TRUE and retrieve=FALSE. The interpolation is done (I can see some warnings that arise during the interpolation), but the returned array has the original dimensions.

not_working <- startR::Start(dataset = '/esarchive/exp/ecearth/a1ua/cmorfiles/DCPP/EC-Earth-Consortium/EC-Earth3/dcppA-hindcast/$member$/Amon/$var$/gr/v20190713/$var$_Amon_*_s$sdate$-$member$_gr_$fyear$.nc',
                             var = 'tos',
                             sdate = paste0(1960:1961),
                             fmonth = 'all',
                             lat = 'all',
                             lon = 'all',
                             fyear = indices(1),
                             member = 'all',
                             fyear_depends = 'sdate',
                             fmonth_across = 'fyear',
                             merge_across_dims = TRUE,
                             transform = CDORemapper, transform_extra_cells = 2, transform_params = list(grid = 'r120x60', method = 'conservative', crop = FALSE), transform_vars = c('lat','lon'),
                             synonims = list(fmonth = c('fmonth','time'), lon = c('lon', 'longitude'), lat = c('lat', 'latitude')),
                             return_vars = list(lat = 'dataset', lon = 'dataset'),
                             lat_reorder = Sort(decreasing = F),
                             lon_reorder = CircularSort(0, 360),
                             num_procs = 1 , retrieve = FALSE)

It works if you select the latitudes and longitudes with values():

working <- startR::Start(dataset = '/esarchive/exp/ecearth/a1ua/cmorfiles/DCPP/EC-Earth-Consortium/EC-Earth3/dcppA-hindcast/$member$/Amon/$var$/gr/v20190713/$var$_Amon_*_s$sdate$-$member$_gr_$fyear$.nc',
                             var = 'tos',
                             sdate = paste0(1960:1961),
                             fmonth = 'all',
                             lat = values(list(-90,90)),
                             lon = values(list(0,359.9)),
                             fyear = indices(1),
                             member = 'all',
                             fyear_depends = 'sdate',
                             fmonth_across = 'fyear',
                             merge_across_dims = TRUE,
                             transform = CDORemapper, transform_extra_cells = 2, transform_params = list(grid = 'r120x60', method = 'conservative', crop = FALSE), transform_vars = c('lat','lon'),
                             synonims = list(fmonth = c('fmonth','time'), lon = c('lon', 'longitude'), lat = c('lat', 'latitude')),
                             return_vars = list(lat = 'dataset', lon = 'dataset'),
                             lat_reorder = Sort(decreasing = F),
                             lon_reorder = CircularSort(0, 360),
                             num_procs = 1 , retrieve = FALSE)

I can work with the second option without any problem, so I'm opening this issue just for your information.

Thank you,

Carlos