(This article was first published on

**Bart Rogiers - Sreigor**, and kindly contributed to R-bloggers)

The *.gslib file format originates from the Geostatistical Software Library, but is also used in the follow-up of that software, i.e. the Stanford Geostatistical Modelling Software (SGeMS). Since not all geostatistical algorithms are implemented in R yet (especially the multiple-point ones), it is useful to have these functions for interacting with SGeMS. I also used it for 3D visualisation of geological datasets that I produced with R.

I extended some code that can be found here for writing *.gslib files from an R dataframe, with an added variable to enable a numeric no data flag. A reading function is also included to be able to reload the files back into R. Not so hard to program, but it might be useful to somebody..

{

cat('GSLIB file created in R\n', file=file)

cat(length(names(dat)), file=file, append=TRUE)

cat('\n', file=file, append=TRUE)

write(cbind(names(dat)), file=file, append=TRUE)

for(i in 1:ncol(dat)) dat[is.na(dat[,i]),i] <- nodatavalue

for(i in 1:ncol(dat)) dat[dat[,i]=='NA',i] <- nodatavalue

for(i in 1:ncol(dat)) dat[is.infinite(dat[,i]),i] <- nodatavalue

write.table(dat,file=file, append=TRUE, sep='\t', col.names=FALSE, row.names=FALSE)

}

read.gslib <- function(filename)

{

char <- scan(filename, what=character())

nums <- which(as.numeric(char)!='NaN')

nrcol <- as.numeric(char[nums[1]])

begin <- nums[2]

eind <- nums[length(nums)]

nrrows <- (eind-begin+1)/nrcol

indices <- seq(1,nrrows*nrcol-1,nrcol)

mat <- matrix(ncol=nrcol, nrow=nrrows)

i <- 1

while(i <= nrcol)

{

mat[,i] <- as.numeric(char[(indices+i+begin-2)])

i <- i+1

}

mat <- as.data.frame(mat)

names(mat) <- c(char[(nums[1]+c(1:nrcol))])

return(mat)

}

To

**leave a comment**for the author, please follow the link and comment on his blog:**Bart Rogiers - Sreigor**.R-bloggers.com offers

**daily e-mail updates**about R news and tutorials on topics such as: visualization (ggplot2, Boxplots, maps, animation), programming (RStudio, Sweave, LaTeX, SQL, Eclipse, git, hadoop, Web Scraping) statistics (regression, PCA, time series, trading) and more...