[R-sig-Geo] Ncdf with missing variable columns

[matteo lattuada]

Hi all,

I'm trying to get some variables iteratively from a directory with a large
number of .nc files and create a csv.file.

I wrote this script:

filenames <- list.files("directory", pattern="*.nc", full.names=TRUE)

res<-lapply(filenames,function(nc){
  t<-nc_open(filename=nc)
  cast<-ncvar_get(t,"wod_unique_cast")
  temp<-ncvar_get(t,"Temperature")
  sal<-ncvar_get(t,"Salinity")
  x<-ncvar_get(t,"lon")
  y<-ncvar_get(t,"lat")
  date<-ncvar_get(t,"date")
  depth<-ncvar_get(t,"z")
  df<-data.frame(cast,x,y,date,depth,temp,sal)
})

df.new<-ldply(res,data.frame)

This script is working well. The problem is that some of the .nc file miss
one of the variables. For example some don't have measurements of
"salinity". When a variable is not found, the iteration stops. Do you have
any hint to get all the variables present in a .nc file and add a NA column
for the absent variables?

Thank you
Kindest Regards
Matteo

	[[alternative HTML version deleted]]