require(ggplot2);
require(plyr);
require(stringr);
require(maps);
options(stringsAsFactors = FALSE)

##Map plotting
statesline <- data.frame(map("state", plot=FALSE)[c("x","y")])
usamap <- qplot(x, y, data=statesline, geom="path")

#County Maps
countfr<-map_data("county")
names(countfr)<-c("long","lat","group","order","State","County")
countfr$State <- state.abb[match(countfr$State, tolower(state.name))]
state_df <- map_data("state")
WImap<-map_data("state",region="Wisconsin")
##Import and clean main data.
clues<-read.csv(file="~/Downloads/clues.csv",encoding="UTF-8")
clues<-subset(clues,select=c(Letterbox.Name,Placer,City,St,County,Place.Date,Boxes,Lat,Lon))
names(clues)<-c("Name","Placer","City","State","County","Date","Boxes","Blat","Blong")
clues$Boxes[which(clues$Boxes<0)]<-NA
clues$Blat[which(clues$Blat<10)]<-NA
clues$Blong[which(clues$Blong>-10)]<-NA
#Match with lowercase names in maps package
clues$County<-tolower(clues$County)
#Crude removal of some mystery boxes
clues$City[which(nchar(clues$City)<4)]<-NA
#Remove some malformed dates
clues$Date[which(nchar(clues$Date)<6)]<-NA
clues$Date<-as.Date(clues$Date,format="%m/%d/%Y");
#Restrict to the US and to properly formed counties
clues<-clues[which(clues$State %in% state.abb),];
clues<-clues[which(clues$County %in% countfr$County),];
#Create a subset for later.  
#WIboxes<-subset(clues,State=="WI",select=c("Blat","Blong"));

#Data frame of factor counts.  
nboxsum<-ddply(clues, c("County","State"),"nrow")
names(nboxsum)<-c("County","State","Boxes")
nboxsum$Letterboxes<-cut(nboxsum$Boxes,breaks=c(1,2,5,10,25,50,100,1000),labels=c("0-1","2-5","5-10","10-25","25-50","50-100","100-1000"),right=FALSE)

choropleth <- merge(countfr, nboxsum, by = c("County","State"))
choropleth$State<-as.factor(choropleth$State)
choropleth$County<-as.factor(choropleth$County)
choropleth <- choropleth[order(choropleth$order), ]
#WIboxcty<-subset(choropleth,State=="WI")
#WIboxcty<-rbind.fill(WIboxcty,WIboxes)

lbplot<-ggplot(choropleth, aes(long, lat, group = group)) + geom_polygon(aes(fill = Letterboxes,name="Boxes Per County"), colour = alpha("white", 1/2), size = 0.2) + geom_polygon(data = state_df, colour = "grey", fill = NA) + scale_fill_brewer(pal = "Greens")+theme_bw()+coord_map(project="lagrange")+scale_x_continuous(name=NULL,breaks = NA)+scale_y_continuous(name=NULL,breaks = NA)+ opts(axis.title.y = theme_blank(), axis.text.y =  theme_blank(),axis.title.x = theme_blank(), axis.text.x =  theme_blank());
#Wisconsin Plot

#wiplot<-ggplot(WIboxcty, aes(long, lat, group = group)) + geom_polygon(aes(fill = Letterboxes), colour = alpha("white", 1/2), size = 0.2) + geom_polygon(data = WImap, colour = "grey", fill = NA) + scale_fill_brewer(pal = "Greens")+theme_bw()+coord_map(project="lagrange")+scale_x_continuous(breaks = NA)+scale_y_continuous(breaks = NA);

##Output code.  uncomment to run.
#svg(filename="USALetterboxes.svg",onefile=TRUE,width=7,height=4.5,bg="transparent")
#lbplot
#dev.off()