0

I have certain files (VI3g raster data) that I want to read in. The names of the files that I want to read in, are saved in a vector called 'filename' (where the first element of the vector is the name of the first file that I want to read in), and the variable names(where the first element of the vector is the name of the first variable that I want to read in) that I want to assign them to is called 'varname'.

filename      varname
file1.VI3g    variable_xy
file2.VI3G    variable_z
...           ...

My approach (that works) to this was the following:

for (i in 1:12) {
assign(varname[i], ReadVI3g(filename[i]))
}

But the data are Rasterlayer. What I would need for my further calculations, are vectors, as some calculations do not work with Rasterlayers. So my approach was to transform the data via getValues in order to get vectors instead of Rasterlayer.

for (i in 1:12) {
  assign(varname[i], ReadVI3g(filename[i]))
  varname[i]<- getValues(varname[i])
}

This however does not work. I get the message:

Error in (function (classes, fdef, mtable)  : unable to find an 
inherited method for function ‘getValues’ for signature ‘"character",
"missing", "missing"’

I understand the problem, but I am not able to find a solution. I tried to get around the problem with assign as well, but it did not work either.

And follow up question: As I am new to R, I have the tendency to solve quite a lot of things with loops. I know that it is inefficient and that there are often smarter solutions. Is there an easy way to avoid the loop here?

Thank you very much in advance.

SpanishTrain
  • 17
  • 2
  • try `getValues(get(varname[i]))` – Tensibai Sep 04 '15 at 08:10
  • Or simply `assign(varname[i], getValues(ReadVI3g(filename[i])))` – Tensibai Sep 04 '15 at 08:12
  • A better approach might be using lapply, it's generally not practical to have all these vectors in your environment. Something like `myvars <- lapply(varnames,function(x){........}` – Heroka Sep 04 '15 at 08:16
  • It seems to me that your approach is fundamentally flawed. Whenever you think you need to use `assign` you are wrong. But as you do not explain what you want to achieve, provide no reproducible example, and use unknown function `ReadVI3g` it is impossible to provide good advice. You should probably create a list of RasterLayer objects (`s = lapply(filename, raster)` ) or a RasterStack `s = stack(filename)`. You also say that 'some calculations do not work with RasterLayers' -- perhaps, but you could be wrong. Do tell us what you want to do to allow us to guide you in the right direction. – Robert Hijmans Sep 05 '15 at 12:43

1 Answers1

0

I would go with the following idea:

apply(df,1,function(x) { assign(x['varname'], getValues(ReadVI3g(x['filename'])), env=.GlobalEnv) })

The env=.GlobalEnv is needed so the variable_xy is set up in the GlobalEnv and not only in the inner function scope.

side note: loops are not inherently inefficient, but they have to be used wisely, prefer the *apply family when possible.

Tensibai
  • 15,557
  • 1
  • 37
  • 57