01: /* StripWWWRuleTest
02: *
03: * Created on Oct 6, 2004
04: *
05: * Copyright (C) 2004 Internet Archive.
06: *
07: * This file is part of the Heritrix web crawler (crawler.archive.org).
08: *
09: * Heritrix is free software; you can redistribute it and/or modify
10: * it under the terms of the GNU Lesser Public License as published by
11: * the Free Software Foundation; either version 2.1 of the License, or
12: * any later version.
13: *
14: * Heritrix is distributed in the hope that it will be useful,
15: * but WITHOUT ANY WARRANTY; without even the implied warranty of
16: * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
17: * GNU Lesser Public License for more details.
18: *
19: * You should have received a copy of the GNU Lesser Public License
20: * along with Heritrix; if not, write to the Free Software
21: * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
22: */
23: package org.archive.crawler.url.canonicalize;
24:
25: import org.apache.commons.httpclient.URIException;
26: import org.archive.net.UURIFactory;
27:
28: import junit.framework.TestCase;
29:
30: /**
31: * Test stripping 'www' if present.
32: * @author stack
33: * @version $Date: 2006-06-12 18:19:23 +0000 (Mon, 12 Jun 2006) $, $Revision: 4280 $
34: */
35: public class StripWWWRuleTest extends TestCase {
36:
37: public void testCanonicalize() throws URIException {
38: String url = "http://WWW.aRchive.Org/index.html";
39: String expectedResult = "http://aRchive.Org/index.html";
40: String result = (new StripWWWRule("test")).canonicalize(url,
41: UURIFactory.getInstance(url));
42: assertTrue("Failed " + result, expectedResult.equals(result));
43: url = "http://wWWW.aRchive.Org/index.html";
44: expectedResult = "http://wWWW.aRchive.Org/index.html";
45: result = (new StripWWWRule("test")).canonicalize(url,
46: UURIFactory.getInstance(url));
47: assertTrue("Failed " + result, expectedResult.equals(result));
48: url = "http://ww.aRchive.Org/index.html";
49: expectedResult = "http://ww.aRchive.Org/index.html";
50: result = (new StripWWWRule("test")).canonicalize(url,
51: UURIFactory.getInstance(url));
52: assertTrue("Failed " + result, expectedResult.equals(result));
53: url = "http://www001.aRchive.Org/index.html";
54: expectedResult = "http://www001.aRchive.Org/index.html";
55: result = (new StripWWWRule("test")).canonicalize(url,
56: UURIFactory.getInstance(url));
57: assertTrue("Failed " + result, expectedResult.equals(result));
58: }
59: }
|