1 // This version of the crawler that also

     2 // "harvests" email addresses from webpages

     3 

     4 import io.Source

     5 import scala.util.matching.Regex

     6 import scala.util._

     7 

     8 def get_page(url: String) : String = {

     9   Try(Source.fromURL(url)("ISO-8859-1").take(10000).mkString).

    10     getOrElse { println(s"  Problem with: $url"); ""}

    11 }

    12 

    13 // regexes for URLs, for "my" domain and for email addresses

    14 val http_pattern = """"https?://[^"]*"""".r

    15 val email_pattern = """([a-z0-9_\.-]+)@([\da-z\.-]+)\.([a-z\.]{2,6})""".r /*@\label{emailline}@*/

    16 

    17 def unquote(s: String) = s.drop(1).dropRight(1)

    18 

    19 def get_all_URLs(page: String) : Set[String] = 

    20   http_pattern.findAllIn(page).map(unquote).toSet

    21 

    22 def print_str(s: String) = 

    23   if (s == "") () else println(s)

    24 

    25 def crawl(url: String, n: Int) : Unit = {

    26   if (n == 0) ()

    27   else {

    28     println(s"  Visiting: $n $url")

    29     val page = get_page(url)

    30     print_str(email_pattern.findAllIn(page).mkString("\n")) /*@\label{mainline}@*/

    31     for (u <- get_all_URLs(page).par) crawl(u, n - 1)

    32   }

    33 }

    34 

    35 // staring URL for the crawler

    36 val startURL = """https://nms.kcl.ac.uk/christian.urban/"""

    37 

    38 

    39 crawl(startURL, 3)