Use org.jsoup instead of net.sourceforge.htmlcleaner

.gitignore

@@ -2,3 +2,4 @@
 bin/*
 testreport/*
 examples/backtothefuture/build/*
+target/*

pom.xml

@@ -0,0 +1,41 @@
+<project xmlns="http://maven.apache.org/POM/4.0.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
+	xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
+	<modelVersion>4.0.0</modelVersion>
+	<groupId>OpenGraph</groupId>
+	<artifactId>OpenGraph</artifactId>
+	<version>0.0.1-SNAPSHOT</version>
+	<name>A Facebook OpenGraph implementation for Java</name>
+	<description>OpenGraph for Java</description>
+
+	<dependencies>
+		<dependency>
+			<groupId>org.hamcrest</groupId>
+			<artifactId>hamcrest-core</artifactId>
+			<version>1.3</version>
+		</dependency>
+		<dependency>
+			<groupId>org.jsoup</groupId>
+			<artifactId>jsoup</artifactId>
+			<version>1.9.2</version>
+		</dependency>
+		<dependency>
+			<groupId>junit</groupId>
+			<artifactId>junit</artifactId>
+			<version>4.12</version>
+		</dependency>
+	</dependencies>
+
+	<build>
+		<sourceDirectory>src</sourceDirectory>
+		<plugins>
+			<plugin>
+				<artifactId>maven-compiler-plugin</artifactId>
+				<version>3.3</version>
+				<configuration>
+					<source />
+					<target />
+				</configuration>
+			</plugin>
+		</plugins>
+	</build>
+</project>

src/org/opengraph/OpenGraph.java → src/main/java/org/opengraph/OpenGraph.java

@@ -1,7 +1,5 @@
 package org.opengraph;
 
-import org.htmlcleaner.HtmlCleaner;
-import org.htmlcleaner.TagNode;
 
 import java.io.BufferedReader;
 import java.io.InputStreamReader;
@@ -13,6 +11,13 @@
 import java.util.regex.Matcher;
 import java.util.regex.Pattern;
 
+import org.jsoup.Jsoup;
+import org.jsoup.nodes.Document;
+import org.jsoup.nodes.Element;
+import org.jsoup.select.Elements;
+
+
+
 /**
  * A Java object representation of an Open Graph enabled webpage.
  * A simplified layer over a Hastable.
@@ -88,51 +93,46 @@ public OpenGraph(String url, boolean ignoreSpecErrors) throws java.io.IOExceptio
         }
 
         String headContentsStr = headContents.toString();
-        HtmlCleaner cleaner = new HtmlCleaner();
-        // parse the string HTML
-        TagNode pageData = cleaner.clean(headContentsStr);
 
-		// read in the declared namespaces
-		boolean hasOGspec = false;
-		TagNode headElement = pageData.findElementByName("head", true);
-		if (headElement.hasAttribute("prefix"))
-		{
-			String namespaceData = headElement.getAttributeByName("prefix");
-			Pattern pattern = Pattern.compile("(([A-Za-z0-9_]+):\\s+(http:\\/\\/ogp.me\\/ns(\\/\\w+)*#))\\s*");
-			Matcher matcher = pattern.matcher(namespaceData);
-			while (matcher.find())
-			{
+        // parse the string HTML
+        Document parsedDocument = Jsoup.parse(headContentsStr);
+        // read in the declared namespaces
+        Elements headElement = parsedDocument.getElementsByTag("head");
+
+        boolean hasOGspec = false;
+        if (headElement.hasAttr("prefix")) {
+            String namespaceData = headElement.attr("prefix");
+            Pattern pattern = Pattern.compile("(([A-Za-z0-9_]+):\\s+(http:\\/\\/ogp.me\\/ns(\\/\\w+)*#))\\s*");
+            Matcher matcher = pattern.matcher(namespaceData);
+            while (matcher.find()) {
                 String prefix = matcher.group(2);
-				String documentURI = matcher.group(3);
-				pageNamespaces.add(new OpenGraphNamespace(prefix, documentURI));
-				if (prefix.equals("og"))
-					hasOGspec = true;
+                String documentURI = matcher.group(3);
+                pageNamespaces.add(new OpenGraphNamespace(prefix, documentURI));
+                if (prefix.equals("og"))
+                    hasOGspec = true;
             }
-		}
+        }
 
-		// some pages do not include the new OG spec
-		// this fixes compatibility
-		if (!hasOGspec)
-			pageNamespaces.add(new OpenGraphNamespace("og", "http:// ogp.me/ns#"));
+        // some pages do not include the new OG spec
+        // this fixes compatibility
+        if (!hasOGspec)
+            pageNamespaces.add(new OpenGraphNamespace("og", "http:// ogp.me/ns#"));
 
         // open only the meta tags
-        TagNode[] metaData = pageData.getElementsByName("meta", true);
-        for (TagNode metaElement : metaData)
-		{
-			for (OpenGraphNamespace namespace : pageNamespaces)
-			{
-				String target = null;
-	            if (metaElement.hasAttribute("property"))
-	                target = "property";
-	            else if (metaElement.hasAttribute("name"))
-	                target = "name";
-
-				if (target != null && metaElement.getAttributeByName(target).startsWith(namespace.getPrefix() + ":"))
-				{
-					setProperty(namespace, metaElement.getAttributeByName(target), metaElement.getAttributeByName("content"));
-					break;
-				}
-			}
+        Elements metaData = parsedDocument.getElementsByTag("meta");
+        for (Element metaElement : metaData) {
+            for (OpenGraphNamespace namespace : pageNamespaces) {
+                String target = null;
+                if (metaElement.hasAttr("property"))
+                    target = "property";
+                else if (metaElement.hasAttr("name"))
+                    target = "name";
+
+                if (target != null && metaElement.attr(target).startsWith(namespace.getPrefix() + ":")) {
+                    setProperty(namespace, metaElement.attr(target), metaElement.attr("content"));
+                    break;
+                }
+            }
         }
 
         /**

src/org/opengraphtests/OpenGraphTest.java → src/test/main/java/org/opengraph/OpenGraphTest.java

@@ -1,4 +1,4 @@
-package org.opengraphtests;
+package org.opengraph;
 
 import org.junit.Test;
 import org.opengraph.OpenGraph;
@@ -9,7 +9,7 @@ public class OpenGraphTest {
   @Test
   public void shouldHandleMissingContentType() throws java.lang.Exception {
     OpenGraph site = new OpenGraph("http://www.bbc.com/future/story/20140428-the-myth-of-tech-revolutions", true);
-    assertEquals("Why it’s time to ditch the word ‘revolution’ in tech", site.getContent("title"));
+    assertEquals("Why it�s time to ditch the word �revolution� in tech", site.getContent("title"));
     assertEquals("624", site.getContent("image:width"));
   }