Creates a wide format for matches

csv-wide.R

# How to run the script
# rscript csv-wide.R "~/path/to/in.csv" "~/path/to/out.csv" column_name_for_matches
args = commandArgs(trailingOnly=TRUE)
match_column <- args[3]
max_width <- 0
csv <- read.csv(args[1], header = T, sep = ",", quote = "\"",
				stringsAsFactors = F, encoding="UTF-8", fill=T)
csv[[match_column]] <- as.character(csv[[match_column]])
#grabs the max length of columns that will exist
for(needle in unique(csv[[match_column]])) {
	if(as.numeric(length(unlist(csv[which(csv[[match_column]]==needle),])))>max_width) {
		max_width <- as.numeric(length(unlist(csv[which(csv[[match_column]]==needle),names(csv)!=match_column])))
	}
}
#creates empty matrix
formatted <- matrix("", nrow = length(unique(csv[[match_column]])), ncol = max_width)
#names rows based on the search match
row.names(formatted) <- unique(csv[[match_column]])
#populates the matrix
for(needle in unique(csv[[match_column]])) {
	row <- as.vector(apply(csv[which(csv[[match_column]]==needle),names(csv)!=match_column],1,unlist))
	row <- append(row,rep(NA,max_width-length(row)))
	formatted[needle,] <- row
}
write.csv(formatted, file=args[2], row.names = TRUE, na="")