// This version of the crawler only// checks links in the "domain" urbancimport io.Sourceimport scala.util.matching.Regeximport scala.util._// gets the first 10K of a web-pagedef get_page(url: String) : String = { Try(Source.fromURL(url).take(10000).mkString) getOrElse { println(s" Problem with: $url"); ""}}// regexes for URLs and "my" domainval http_pattern = """\"https?://[^\"]*\"""".rval my_urls = """urbanc""".rdef unquote(s: String) = s.drop(1).dropRight(1)def get_all_URLs(page: String) : Set[String] = { http_pattern.findAllIn(page).map(unquote).toSet}def crawl(url: String, n: Int) : Unit = { if (n == 0) () else if (my_urls.findFirstIn(url) == None) { println(s"Visiting: $n $url") get_page(url); () } else { println(s"Visiting: $n $url") for (u <- get_all_URLs(get_page(url))) crawl(u, n - 1) }}// staring URL for the crawlerval startURL = """http://www.inf.kcl.ac.uk/staff/urbanc/"""// can now deal with depth 3 and beyondcrawl(startURL, 3)