This project has retired. For details please refer to its Attic page.
DefaultValidatorTest xref
View Javadoc
1   /*
2    * Licensed to the Apache Software Foundation (ASF) under one or more
3    * contributor license agreements.  See the NOTICE file distributed with
4    * this work for additional information regarding copyright ownership.
5    * The ASF licenses this file to You under the Apache License, Version 2.0
6    * (the "License"); you may not use this file except in compliance with
7    * the License.  You may obtain a copy of the License at
8    *
9    *  http://www.apache.org/licenses/LICENSE-2.0
10   *
11   * Unless required by applicable law or agreed to in writing, software
12   * distributed under the License is distributed on an "AS IS" BASIS,
13   * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
14   * See the License for the specific language governing permissions and
15   * limitations under the License.
16   */
17  
18  package org.apache.any23.validator;
19  
20  import org.apache.any23.extractor.html.DomUtils;
21  import org.apache.any23.extractor.html.TagSoupParser;
22  import org.junit.After;
23  import org.junit.Assert;
24  import org.junit.Before;
25  import org.junit.Test;
26  import org.slf4j.Logger;
27  import org.slf4j.LoggerFactory;
28  import org.w3c.dom.Node;
29  
30  import java.io.IOException;
31  import java.io.InputStream;
32  import java.net.URI;
33  import java.net.URISyntaxException;
34  import java.util.List;
35  
36  /**
37   * Test case for {@link DefaultValidator}.
38   *
39   * @author Michele Mostarda (mostarda@fbk.eu)
40   */
41  public class DefaultValidatorTest {
42  
43      private static final Logger logger = LoggerFactory.getLogger(DefaultValidatorTest.class);
44  
45      private DefaultValidator validator;
46  
47      @Before
48      public void setUp() {
49          validator = new DefaultValidator();
50      }
51  
52      @After
53      public void tearDown() {
54          validator = null;
55      }
56  
57      @Test
58      public void testRegisterRule() {
59          validator.addRule(FakeRule.class, FakeFix.class);
60          List<Class<? extends Fix>> fixes = validator.getFixes(FakeRule.class);
61          Assert.assertEquals("Unexpected fixes size.", 1, fixes.size());
62          Assert.assertEquals("Unexpected fix.", FakeFix.class, fixes.get(0));
63          validator.removeRule(FakeRule.class);
64          Assert.assertEquals("Unexpected fixes size.", 0, validator.getFixes(FakeRule.class).size());
65      }
66  
67      @Test
68      public void testMissingOGNamespace() throws IOException, ValidatorException, URISyntaxException {
69          DOMDocument document = loadDocument("missing-og-namespace.html");
70          Assert.assertNull(document.getNode("/HTML").getAttributes().getNamedItem("xmlns:og"));
71          ValidationReport validationReport = validator.validate(document, true);
72          Assert.assertNotNull(document.getNode("/HTML").getAttributes().getNamedItem("xmlns:og"));
73          if (logger.isDebugEnabled()) {
74              logger.debug(validationReport.toString());
75          }
76      }
77  
78      @Test
79      public void testMissingItemscopeAttributeValue() throws IOException, URISyntaxException, ValidatorException {
80          DOMDocument document = loadDocument("microdata-basic.html");
81          List<Node> nullItemScopeNodes = document.getNodesWithAttribute("itemscope");
82          for (Node node : nullItemScopeNodes) {
83              // all nodes with itemscope have an empty string value
84              Assert.assertEquals("", node.getAttributes().getNamedItem("itemscope").getNodeValue());
85          }
86          ValidationReport validationReport = validator.validate(document, true);
87          List<Node> fixedItemScopeNodes = document.getNodesWithAttribute("itemscope");
88          for (Node node : fixedItemScopeNodes) {
89              // all nodes with itemscope now have a default value of "itemscope"
90              Assert.assertNotNull(node.getAttributes().getNamedItem("itemscope").getNodeValue());
91              Assert.assertEquals("itemscope", node.getAttributes().getNamedItem("itemscope").getNodeValue());
92          }
93          if (logger.isDebugEnabled()) {
94              logger.debug(validationReport.toString());
95          }
96      }
97  
98      @Test
99      public void testMetaNameMisuse() throws Exception {
100         DOMDocument document = loadDocument("meta-name-misuse.html");
101         ValidationReport validationReport = validator.validate(document, true);
102         if (logger.isDebugEnabled()) {
103             logger.debug(validationReport.toString());
104             logger.debug(DomUtils.serializeToXML(document.getOriginalDocument(), true));
105         }
106 
107         List<Node> metas = document.getNodes("/HTML/HEAD/META");
108         for (Node meta : metas) {
109             Node name = meta.getAttributes().getNamedItem("name");
110             if (name != null) {
111                 Assert.assertFalse(name.getTextContent().contains(":"));
112             }
113         }
114     }
115 
116     @Test
117     public void testAboutNotIRIRule() throws Exception {
118         DOMDocument document = loadDocument("invalid-rdfa-about.html");
119         ValidationReport validationReport = validator.validate(document, true);
120         if (logger.isDebugEnabled()) {
121             logger.debug(validationReport.toString());
122         }
123         Assert.assertEquals("Unexpected number of issues.", 1, validationReport.getIssues().size());
124     }
125 
126     public static DOMDocument loadDocument(String document) throws IOException, URISyntaxException {
127         InputStream is = DefaultValidatorTest.class.getResourceAsStream(document);
128         final String documentIRI = "http://test.com";
129         TagSoupParser tsp = new TagSoupParser(is, documentIRI);
130         return new DefaultDOMDocument(new URI(documentIRI), tsp.getDOM());
131     }
132 
133     static class FakeRule implements Rule {
134         public String getHRName() {
135             return "fake-rule";
136         }
137 
138         public boolean applyOn(DOMDocument document, @SuppressWarnings("rawtypes") RuleContext context,
139                 ValidationReportBuilder validationReportBuilder) {
140             throw new UnsupportedOperationException();
141         }
142     }
143 
144     static class FakeFix implements Fix {
145         public String getHRName() {
146             return "fake-fix";
147         }
148 
149         public void execute(Rule rule, @SuppressWarnings("rawtypes") RuleContext context, DOMDocument document) {
150             throw new UnsupportedOperationException();
151         }
152     }
153 
154 }