progs/crawler2.scala
changeset 722 14914b57e207
parent 550 71fc4a7a7039
--- a/progs/crawler2.scala	Thu Apr 16 19:15:46 2020 +0100
+++ b/progs/crawler2.scala	Wed May 06 15:37:31 2020 +0100
@@ -1,44 +1,38 @@
-// This version of the crawler only
-// checks links in the "domain" urbanc
+// This version of the crawler that also
+// "harvests" email addresses from webpages
 
 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)("ISO-8859-1").take(10000).mkString). 
+  Try(Source.fromURL(url)("ISO-8859-1").take(10000).mkString).
     getOrElse { println(s"  Problem with: $url"); ""}
 }
 
-// regexes for URLs and "my" domain
+// regexes for URLs, for "my" domain and for email addresses
 val http_pattern = """"https?://[^"]*"""".r
-val my_urls = """urban""".r       /*@\label{myurlline}@*/
-//val my_urls = """kcl.ac.uk""".r 
+val email_pattern = """([a-z0-9_\.-]+)@([\da-z\.-]+)\.([a-z\.]{2,6})""".r /*@\label{emailline}@*/
 
 def unquote(s: String) = s.drop(1).dropRight(1)
 
 def get_all_URLs(page: String) : Set[String] = 
   http_pattern.findAllIn(page).map(unquote).toSet
 
+def print_str(s: String) = 
+  if (s == "") () else println(s)
+
 def crawl(url: String, n: Int) : Unit = {
-  if (n == 0) ()                   /*@\label{changestartline}@*/
-  else if (my_urls.findFirstIn(url) == None) { 
-    println(s"Visiting: $n $url")
-    get_page(url); () 
-  }                                /*@\label{changeendline}@*/
+  if (n == 0) ()
   else {
-    println(s"Visiting: $n $url")
-    for (u <- get_all_URLs(get_page(url)).par) crawl(u, n - 1)
+    println(s"  Visiting: $n $url")
+    val page = get_page(url)
+    print_str(email_pattern.findAllIn(page).mkString("\n")) /*@\label{mainline}@*/
+    for (u <- get_all_URLs(page).par) crawl(u, n - 1)
   }
 }
 
-// starting URL for the crawler
+// staring URL for the crawler
 val startURL = """https://nms.kcl.ac.uk/christian.urban/"""
-//val startURL = """https://nms.kcl.ac.uk/christian.urban/bsc-projects-17.html"""
-
 
-// can now deal with depth 3 and beyond
 crawl(startURL, 3)
-
-