Update get_pfafstetter_basins.R

[domisch]

Added a pre-selection of the 100 longest streams from which the most contributing (main) stream will be identified based on the area

[merretbuurman]

Hi! Thanks for the contribution!
Unfortunately I could not get the example to run. This is what happens:

> pfafstetter <- get_pfafstetter_basins(g = g_subset, subc_raster = subc_raster,  out_dir = my_directory, file_name = "pfafstetter_raster.tif",  data_table = FALSE,  n_cores = 4)
Setting up parallel backend...
Using 4 CPUs... 
Preparing the graph...
Calculating the area of single sub-catchments...
Finding the most contributing stream... 
Error in setnames(x, value) : 
  Can't assign 1 names to a 0-column data.table
> 

Some hint:
During testing, I found out that after the line cat("Finding the most contributing stream..."), the object headwater is an object of type character with a length of hundred, but it has no names - length(names(headwater)) is zero, so maybe that is what causes it to fail? names(headwater) is used repeatedly in the lines after that. But I didn't debug any further yet.