Was the filtrate function removed from ncdump?

[RMORSEcode]

The changes you made corrected the problems I had with raster (THANK YOU!), but now I can't seem to find a function and am getting an error on activate Did filtrate become hyper_filter in ncdump?

https://github.com/hypertidy/angstroms/blob/master/inst/couple-Atlantis/cpolar_coupling_2.R from line 216:

library(ncdump) library(dplyr) ncfile <- "~/Git/shinyrAtlantis/egfiles/SOtempdummy.nc" con <- ncdf4::nc_open(ncfile) trans <- ncdump::NetCDF(ncfile) %>% activate("temperature") %>% filtrate() hslab <- bind_rows(lapply(trans, function(x) tibble(name = x$name[1], start = min(x$step), count =length(x$step)))) slab <- ncdf4::ncvar_get(con, "temperature", start = hslab$start, count = hslab$count) temp <- as_tib_slab(trans, slab, activ = "temperature")

Here is the error: trans <- ncdump::NetCDF(ncfile) %>% activate("temperature") %>% filtrate() Error in filtrate(.) : could not find function "filtrate" In addition: Warning message: In active<-.default(*tmp*, value = what) : activating as a default, this object not recognized as activatable

[mdsumner]

Ah yes, this has completely changed - filtrate was a very early stab at having axis-transforms, and that became hyper_filter in tidync - and recently that changed again so that there's really on one class.

I'll update that example script! - this has languished and just isn't properly fleshed out.

[RMORSEcode]

Thanks!

On Nov 3, 2017 7:02 PM, "Michael Sumner" notifications@github.com wrote:

Ah yes, this has completely changed - filtrate was a very early stab at having axis-transforms, and that became hyper_filter in tidync - and recently that changed again so that there's really on one class.

I'll update that example script! - this has languished and just isn't properly fleshed out.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/hypertidy/angstroms/issues/7#issuecomment-341848003, or mute the thread https://github.com/notifications/unsubscribe-auth/AKIPqJnjuwJZ4w02hJGzE2LXXYGEHRJPks5sy5uLgaJpZM4QRqtz .

[mdsumner]

Ok, please ignore cpolar_coupling_2.R - that was a premature attempt to move to tidync/sf combination. cpolar_coupling_z_index.R is more straightforward and should work, but still with the caveat that it's an ongoing work-in-progress for me.