just the facts ma'am

scrapeweed.R

library(XML)

#There are probably smarter ways to do this, but this works well enough
url.prefix<-"http://www.priceofweed.com/prices/United%20States/"
url.suffix<-".html"
state.uri<- paste(url.prefix,state.name,url.suffix,sep="");


#This function is quick and dirty (emphasis on dirty).  It works as well
#as it does because priceofweed.com has very structured web pages.  The first
#table on each page is the summary (trimmed mean) of reported prices.
#No attempt is made to check for consistency but as you can see we don't encounter any problems
weed.grab<- function(sleep=1) {
	weed.name<- c("HQ.price", "MQ.price","LQ.price","HQ.n","MQ.n","LQ.n")
	weed.mat<- matrix(0,50,6,dimnames=list(state.name,weed.name))
	for (i in 1:50) {
		state.int<- unlist(readHTMLTable(doc=state.uri[i] , as.data.frame=FALSE)[[1]])
		weed.mat[i,]<- c(as.numeric(substring(state.int[6:8],2)),as.numeric(state.int[10:12]))
		rm(state.int)
		Sys.sleep(sleep)
		}
	weed.prices.df<- as.data.frame(weed.mat)
	names(weed.prices.df)<- c("High Quality","Med. Quality","Low Quality","HQ.n","MQ.n","LQ.n")
	return(weed.prices.df)
        };