7
|
1 |
val my_urls = """urbanc""".r
|
|
2 |
|
|
3 |
def crawl(url: String, n: Int) : Unit = {
|
|
4 |
if (n == 0) ()
|
254
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
5 |
else if (my_urls.findFirstIn(url) == None) {
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
6 |
println(s"Visiting: $n $url")
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
7 |
get_page(url); ()
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
8 |
}
|
7
|
9 |
else {
|
96
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
10 |
println(s"Visiting: $n $url")
|
254
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
11 |
for (u <- get_all_URLs(get_page(url))) crawl(u, n - 1)
|
7
|
12 |
}
|
|
13 |
}
|