Hi All,
I am currently working on an analysis which requires a call to an
external FORTRAN routine contained within a file called MCDS.EXE. This
file is usually called from within a WINDOWS program called DISTANCE. I
have some R script from the developers of the original software which
apparently makes a call to this .EXE file which i need to re-engineer
slightly, but i am having difficulty interpreting how the script makes
this call and passes to the EXE file the relevant input command file and
data file for processing. I am also unsure if making such a call will
work within the OSX environment for a (windows?) executable file. Below
is an example from the script which is supposed to make the call. Any
suggestions would be much appreciated.
Cheers
A
cds <- function (key, adj, L, w, A=NA, xi, zi,
file.base,ext.files=F,bootstrap=F) {
#Purpose: Driver function to run the mcds.exe engine from R
#Updated by Tiago on 19/12/2004 so that it can take an external data and
command input files
#Usefull if you want to do a non-standard bootstrap analysis of some
data analised in distance
#and for wich distance is not capable of producing variance estimates
#This way, you can call this function inside a bootstrap procedure, as
long as you update the
#data file at each resample
#Inputs:
# key - vector string containing key functions
# adj - vector string containing adjustment terms
# L - total survey effort
# A - survey area
# xi - vector of perpendicular distances
# zi - vector of cluster sizes
# file.base - if specified and engine is cds, then the cds input and
# output files are written into the current directory, with the
# cds.file.base as a prefix and '.txt' as a suffix. E.g., setting
# cds.file.base to 'cds' produces 'cds.cmd.txt', 'cds.data.txt',
# 'cds.log.txt','cds.stat.txt','cds.plot.txt' and 'cds.boot.txt'
# If not specified, these files are created in a temp location and
# are deleted at the end.
# ext.files - If true, then the funtion uses the external files
'file.base'+'.cms.'+txt' as command file
# and 'file.base'+'.data.'+.txt' as data file. Typicaly these
files would be the result of
# runing Distance in debug mode, and should be placed in the
working directory for R.
# By default ext.files is false, so the function looks for the
files produced by functions
# 'create.data.file' and 'create.command.file'
# You need to change the input comand file in order for the
mcds engine to produce the files
# that the function 'read.stats.file' expects, and that means
that inside the command file you should define
# file names with prefix = file.base
# bootstrap - If true, procedure is being called inside a bootstrap
routine, and intermediate files
# are deleted at each loop step, except the command file
#Outputs: list, containing
# Nhat.grp, Nhat.ind, mu, nL, Es, LnL, AIC, status
# Note - status is an integer:
# 1=OK, 2=warnings, 3=errors, 4=file errors, 5=some other problem
(e.g., program crash)
run.cds<-function(cmd.file.name) {
#Purpose: runs the MCDS.exe engine and waits for it to finish
# *Note* that mcds.exe needs to be in the working directory, or in the
# PATH windows environment
# variable for this to work, as it makes no attempt to find the
# location of the file
#Inputs:
# cmd.file.name - name of the command file to run
#Returns:
# A status integer - 1=OK, 2=warnings, 3=errors, 4=file errors, 5=some
other problem (e.g., program crash)
______________________________________________
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
