progs/crawler1.scala
author Christian Urban <christian dot urban at kcl dot ac dot uk>
Thu, 14 Nov 2013 20:08:39 +0000
changeset 190 0b63c0edfb09
parent 112 95ee5cc5c05d
child 242 35104ee14f87
permissions -rw-r--r--
added new version

// A crawler which checks whether there
// are problems with links in web-pages

import io.Source
import scala.util.matching.Regex
import scala.util._

// gets the first 10K of a web-page
def get_page(url: String) : String = {
  Try(Source.fromURL(url).take(10000).mkString) getOrElse 
    { println(s"  Problem with: $url"); ""}
}

// regex for URLs
val http_pattern = """\"https?://[^\"]*\"""".r

// drops the first and last character from a string
def unquote(s: String) = s.drop(1).dropRight(1)

def get_all_URLs(page: String) : Set[String] = {
  http_pattern.findAllIn(page).map(unquote).toSet
}

// naive version - seraches until a given depth
// visits pages potentially more than once
def crawl(url: String, n: Int) : Unit = {
  if (n == 0) ()
  else {
    println(s"Visiting: $n $url")
    for (u <- get_all_URLs(get_page(url))) crawl(u, n - 1)
  }
}

// staring URL for the crawler
val startURL = """http://www.inf.kcl.ac.uk/staff/urbanc/"""
//val startURL = """http://www.inf.kcl.ac.uk/staff/mml/"""

crawl(startURL, 2)