search

dankelley / ocencdf

Interface from oce to netcdf
https://dankelley.github.io/ocencdf/
2 stars 1 forks source link

include force_v4 argument with default TRUE in all functions #16

Closed richardsc closed 9 months ago

richardsc commented 9 months ago

This PR arose from a recent issue noted by @kurtisanstey where the netCDF file creation fails if the variables are "too big" (I put this in quotes because I don't actually know what the limit is). This appears to be a limitation of the netCDF v3 file, which is what ncdf4::nc_create() defaults to, by way of the force_v4=FALSE argument.

All I've done here is to add the force_v4 argument to oce2ncdf() (and each of the sub-functions, e.g. adp2ncdf()), but with a default value of TRUE. Considering netCDF v4 was released in 2008, I think we're safe to assume that most people creating new nc files are fine with using version 4.

richardsc commented 9 months ago

I just noticed that I messed up my fork/edit/PR and did it on the main branch instead of develop. @dankelley let me know if you want me to do it again.

dankelley commented 9 months ago

Thanks. I'll propagate it into the develop etc.

PS. no, I didn't get a notification. I can't for the life of me figure out what setting controls that. I do pay for GH, and wish there were some way to get help on such things.

kurtisanstey commented 8 months ago

Thanks for your help on this, folks. Your package has been very useful in getting Sentinel V data into a usable .nc state!