statshome <- paste(readLines("http://www.stat.cmu.edu/",warn=FALSE),
  collapse="")

slurp <- function(url) {
  return(paste(readLines(url,warn=FALSE), collapse=" "))
}

raise.anchors <- function(text) {
  match.locations <- gregexpr('href=".+?"',text,useBytes=TRUE,ignore.case=TRUE)
  return(regmatches(text, match.locations)[[1]])
}

anchors.with.absolute.urls <- function(urls) {
  return(urls[grepl('"http://',urls)])
}

extract.links <- function(text) {
  anchors <- raise.anchors(text)
  good.anchors <- anchors.with.absolute.urls(anchors)
  url.locations <- gregexpr('http://[^"]+?', good.anchors,ignore.case=TRUE)
  absolute.urls <- regmatches(good.anchors,url.locations)
  absolute.urls <- do.call(c,absolute.urls)
  return(absolute.urls)
}

surf.from <- function(some.urls,home="http://crookedtimber.org/") {
  if (length(some.urls) == 0) {
    return(home)
  } else {
    return(sample(some.urls,size=1))
  }
}

surf.to <- function(origin,target,max.iter=20,verbose=TRUE,...) {
  url <- origin
  count <- 0
  while ( !(grepl(target, url, ignore.case=TRUE)) &&(count < max.iter)) {
    text <- slurp(url)
    outgoing.urls <- extract.links(text)
    url <- surf.from(outgoing.urls,...)
    if(verbose) { cat(paste("surfing to",url,"\n")) }
    count <- count+1
  }
  return(count)
}

surf.to.mk2 <- function(origin,target,max.iter=20,verbose=TRUE,...) {
  url <- origin
  last_url <- url
  count <- 0
  while ( !(grepl(target, url, ignore.case=TRUE)) &&(count < max.iter)) {
    text <- try(slurp(url))                   # Error handling
    if (class(text) == "try-error") {         # If slurp() yields an error
      url <- last_url
    } else {                                  # If slurp() worked OK
      last_url <- url
      outgoing.urls <- extract.links(text)
      url <- surf.from(outgoing.urls,...)
      if(verbose) { cat(paste("surfing to",url,"\n")) }
      count <- count+1
    }
  }
  return(count)
}