author | Christian Urban <christian dot urban at kcl dot ac dot uk> |
Fri, 15 Jul 2016 10:54:09 +0100 | |
changeset 406 | 0a42d73e795b |
parent 330 | 0806e45d873c |
permissions | -rw-r--r-- |
val http_pattern = """"https?://[^"]*"""".r val email_pattern = """([a-z0-9_\.-]+)@([\da-z\.-]+)\.([a-z\.]{2,6})""".r def print_str(s: String) = if (s == "") () else println(s) def crawl(url: String, n: Int) : Unit = { if (n == 0) () else { println(s"Visiting: $n $url") val page = get_page(url) print_str(email_pattern.findAllIn(page).mkString("\n")) for (u <- get_all_URLs(page).par) crawl(u, n - 1) } }