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 

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

    25 

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

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

    28 }

    29 

    30 // naive version - seraches until a given depth

    31 // visits pages potentially more than once

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

    33   if (n == 0) ()

    34   else if (my_urls.findFirstIn(url) == None) ()

    35   else {

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

    37     for (u <- get_all_URLs(get_page(url))) crawl(u, n - 1)

    38   }

    39 }

    40 

    41 // can now deal with depth 3

    42 // start on command line

    43 crawl(startURL, 4)

    44