##############
# string to upper case function
###############

CapLeading <- function (string){
   fn <- function(x){
     v <- unlist(strsplit(x, split = " "))
     u <- sapply(v, function(x){
            x <- tolower(x)
            substring(x, 1, 1) <- toupper(substring(x, 1, 1))
            x})
     paste(u, collapse = " ")
     }
   sapply(string, fn)
   }

### outputfiles
fileName<-"euraocean"
fileType<-"Q"

#### sets the number of populations to through
#lowest K
Start_K<-12
#highest K
End_K<-12

#### loop does all the work
while (Start_K <= End_K)
{
	KNum<-Start_K
	
	# get output file data

	QFile<-read.csv(paste(fileName,KNum,fileType,sep="."),sep=" ",header=FALSE)
	QFam <-read.csv(paste(fileName,".fam",sep=""),sep=" ",header=FALSE)

	# remove a bunch of extraneous columns
	QFam$V6<-NULL
	QFam$V5<-NULL
	QFam$V4<-NULL
	QFam$V3<-NULL
	QFam$V1<-NULL
	QFam$Id<-QFam$V2
	QFam$V2<-NULL
	QFileFam<-cbind(QFam,QFile)

	#get more info from master data frame for the entries you have in the fam file
	QFileFamMaster<-merge(QFileFam,HGDPMaster)
	
	KPlot<-data.frame()
	# get unique population names
	allUnique<-unique(QFileFamMaster$PopulationSpecific)

	for (levels in allUnique)
	{

	# uncomment these lines if you want to limit to specific populations
	#if(levels =="Bengali" | 
	#levels == "hema" |
	#levels == "Algeria" |
	#levels == "samoan" |
	#levels == "krygzstani" |
	#levels == "yhai"|
	#levels == "punjabi arain" |
	#levels == "mepalese" |
	#levels == "spaniards" |
	#levels == "yemenese" |
	#levels == "lebanese" |
	#levels == "saudis" |
	#levels == "egyptians" |
	#levels == "romanians" |
	#levels == "cypriots" |
	#levels == "lithuanians" |
	#levels == "belorussian" |
	#levels == "syrians" |
	#levels == "iranians" |
	#levels == "north Kannadi" |
	#levels == "turks" |
	#levels == "armenians" |
	#levels == "georgians" |
	#levels == "moroccans" |
	#levels == "evenkis" |
	#levels == "nganassans" |
	#levels == "tn dalit" |
	#levels == "Brahui" |
	#levels == "Sindhi" |
	#levels == "French" |
	#levels == "Papuan" |
	#levels == "Palestinian" |
	#levels == "Cambodian" |
	#levels == "Japanese" |
	#levels == "Orcadian" |
	#levels == "Han" |
	#levels == "Pima" |
	#levels == "Dai" |
	#levels == "FrenBasque" |
	#levels == "Adygei" |
	#levels == "Irula" |
	#levels == "Gujarati_A" |
	#levels == "Gujarati_B" |
	#levels == "Iban" |
	#levels == "tn brahmin" |
	#levels == "vietnamese" 
	#)

	### this brace only matters if the previous line is uncommented
	{
		slice <- (QFileFamMaster[QFileFamMaster$PopulationSpecific==levels ,])
		a = 1
		colSlice = 2
		rowVec<-NULL
		evalString<-NULL
		while (a <= KNum)
		{
			theMean<-mean(slice[,colSlice])
			rowVec[a]<-theMean
			assign(paste("K",a,sep=""),theMean )
			theKassign<-paste("K",a,sep="")
			theKassign<-paste(theKassign,"=rowVec[",sep="")
			theKassign<-paste(theKassign,a,sep="")
			theKassign<-paste(theKassign,"]",sep="")
				
			if (a == 1)
			{evalString <-	paste(evalString,theKassign,sep="")}
			else
			{
			{evalString <-	paste(evalString,theKassign,sep=",")}
			}
			a<-a+1	
			colSlice<-colSlice+1
		}
	
	### this renames some outputs
		if (levels == "French Basque")
		{
			levels = "Basque"	
		}
		levelString<-paste("'",levels,sep="")
		levelString<-paste(levelString,"'",sep="")
		evalLevels<-paste("PopulationSpecific=",levelString,sep="")
		evalString<-paste(evalString,evalLevels,sep=",")
		evalString<-paste("df<-data.frame(",evalString,sep="")
		evalString<-paste(evalString,")",sep="")
		df<-data.frame()
		eval(parse(text=evalString))

		KPlot<-rbind(KPlot,df) 

		}
	}

### outputs spread sheet for unique populations
write.table(KPlot,file="K.csv")
KPlot<-t(KPlot)

PopNames<-KPlot[KNum+1,]
KPlot<-KPlot[-c(KNum+1),]
x11()
par(mai=c(1.5,0,.1,.25),font=2,font.lab=4)
### the plot. cex is the font size
barplot(KPlot,col=rainbow(KNum),names.arg=CapLeading(PopNames),cex.names=.75,las=2)
mtext(paste("K =",KNum,sep=" "),4,cex=3)

rm(KPlot)

## increment up for the next K
Start_K<-Start_K+1

}