1 import io.Source

     2 import scala.util.matching.Regex

     3 

     4 // gets the first ~10K of a page

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

     6   try {

     7     Source.fromURL(url).take(10000).mkString  

     8   }

     9   catch {

    10     case e => {

    11       println("  Problem with: " + url)

    12       ""

    13     }

    14   }

    15 }

    16 

    17 // staring URL for the crawler

    18 val startURL = """http://www.inf.kcl.ac.uk/staff/urbanc/"""

    19 

    20 // regex for URLs

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

    22 val my_urls = """urbanc""".r

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

    24 

    25 // http://net.tutsplus.com/tutorials/other/8-regular-expressions-you-should-know/

    26 

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

    28 

    29 def get_all_URLs(page: String) : Set[String] = {

    30   (http_pattern.findAllIn(page)).map { unquote(_) }.toSet

    31 }

    32 

    33 // naive version - seraches until a given depth

    34 // visits pages potentially more than once

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

    36   if (n == 0) ()

    37   //else if (my_urls.findFirstIn(url) == None) ()

    38   else {

    39     println("Visiting: " + n + " " + url)

    40     val page = get_page(url)

    41     println(email_pattern.findAllIn(page).mkString("\n"))

    42     for (u <- get_all_URLs(page)) crawl(u, n - 1)

    43   }

    44 }

    45 

    46 // can now deal with depth 3

    47 // start on command line

    48 crawl(startURL, 3)

    49