Merge pull request #29 from futures/DublinCoreIndexer

Create initial Dublin Core indexer framework and REST API endpoint.
fcrepo4-archive · Feb 7, 2013 · bae91c7 · bae91c7
2 parents 245b8d7 + 96662a0
commit bae91c7
Show file tree

Hide file tree

Showing 14 changed files with 348 additions and 4 deletions.
diff --git a/src/main/java/org/fcrepo/modeshape/FedoraDatastreams.java b/src/main/java/org/fcrepo/modeshape/FedoraDatastreams.java
@@ -232,6 +232,10 @@ private URI addDatastreamNode(final String dsPath,
 			ds.setProperty("fedora:created", Calendar.getInstance());
 		}
 		ds.setProperty("jcr:lastModified", Calendar.getInstance());
+
+        // TODO: I guess we should also have the PID + DSID..
+        ds.setProperty("dc:identifier", new String[] { ds.getIdentifier() });
+
 		session.save();
 		session.logout();
 		logger.debug("Finished adding datastream node at path: " + dsPath);

diff --git a/src/main/java/org/fcrepo/modeshape/FedoraObjects.java b/src/main/java/org/fcrepo/modeshape/FedoraObjects.java
@@ -11,10 +11,7 @@
 import java.io.IOException;
 import java.util.Calendar;
 
-import javax.jcr.Node;
-import javax.jcr.NodeIterator;
-import javax.jcr.RepositoryException;
-import javax.jcr.Session;
+import javax.jcr.*;
 import javax.jcr.nodetype.NodeType;
 import javax.ws.rs.DELETE;
 import javax.ws.rs.GET;
@@ -74,6 +71,7 @@ public Response ingest(@PathParam("pid") final String pid)
 			obj.addMixin("fedora:owned");
 			obj.setProperty("fedora:ownerId", "Fedo Radmin");
 			obj.setProperty("jcr:lastModified", Calendar.getInstance());
+            obj.setProperty("dc:identifier", new String[] { obj.getIdentifier(), pid });
 			session.save();
 			session.logout();
 			logger.debug("Finished ingest with pid: " + pid);
@@ -146,4 +144,5 @@ public Long getNumObjects() throws RepositoryException {
 		session.logout();
 		return numObjects;
 	}
+
 }
diff --git a/src/main/java/org/fcrepo/modeshape/indexer/DublinCore.java b/src/main/java/org/fcrepo/modeshape/indexer/DublinCore.java
@@ -0,0 +1,58 @@
+package org.fcrepo.modeshape.indexer;
+
+import org.fcrepo.modeshape.AbstractResource;
+import org.fcrepo.modeshape.indexer.dublincore.AbstractIndexer;
+import org.fcrepo.modeshape.indexer.dublincore.IndexFromWellKnownPath;
+
+import javax.inject.Inject;
+import javax.jcr.Node;
+import javax.jcr.RepositoryException;
+import javax.jcr.Session;
+import javax.ws.rs.GET;
+import javax.ws.rs.Path;
+import javax.ws.rs.PathParam;
+import javax.ws.rs.Produces;
+import javax.ws.rs.core.Response;
+
+import java.io.InputStream;
+import java.util.List;
+
+import static javax.ws.rs.core.MediaType.TEXT_XML;
+import static javax.ws.rs.core.Response.ok;
+
+
+@Path("/objects/{pid}/oai_dc")
+public class DublinCore extends AbstractResource {
+
+    @Inject
+    List<AbstractIndexer> indexers;
+
+    @GET
+    @Produces({ TEXT_XML })
+    public Response getObjectAsDublinCore(@PathParam("pid") final String pid) throws RepositoryException {
+        final Session session = repo.login();
+
+        try {
+            if (session.nodeExists("/objects/" + pid)) {
+                final Node obj = session.getNode("/objects/" + pid);
+
+                for(AbstractIndexer indexer : indexers) {
+                    InputStream inputStream = indexer.getStream(obj);
+
+                    if(inputStream != null) {
+                        return ok(inputStream, TEXT_XML).build();
+                    }
+                }
+
+                return four04;
+            } else {
+                return four04;
+            }
+
+        } finally {
+            session.logout();
+        }
+
+    }
+
+}
diff --git a/src/main/java/org/fcrepo/modeshape/indexer/dublincore/AbstractIndexer.java b/src/main/java/org/fcrepo/modeshape/indexer/dublincore/AbstractIndexer.java
@@ -0,0 +1,16 @@
+package org.fcrepo.modeshape.indexer.dublincore;
+
+
+import javax.jcr.Node;
+import java.io.ByteArrayInputStream;
+import java.io.InputStream;
+import java.io.UnsupportedEncodingException;
+
+public abstract class AbstractIndexer {
+
+    public AbstractIndexer() {
+    }
+
+    public abstract InputStream getStream(Node node);
+
+}
diff --git a/src/main/java/org/fcrepo/modeshape/indexer/dublincore/IndexFromJcrProperties.java b/src/main/java/org/fcrepo/modeshape/indexer/dublincore/IndexFromJcrProperties.java
@@ -0,0 +1,46 @@
+package org.fcrepo.modeshape.indexer.dublincore;
+
+import javax.jcr.*;
+import java.io.ByteArrayInputStream;
+import java.io.InputStream;
+import java.io.UnsupportedEncodingException;
+
+public class IndexFromJcrProperties extends AbstractIndexer {
+
+    @Override
+    public InputStream getStream(Node node) {
+
+
+        String str = "<oai_dc:dc xmlns:oai_dc=\"http://www.openarchives.org/OAI/2.0/oai_dc/\" xmlns:dc=\"http://purl.org/dc/elements/1.1/\">";
+
+        try {
+            final String[] nameGlobs = new String[] { "dc:*"};
+            PropertyIterator iter = node.getProperties(nameGlobs);
+
+            while(iter.hasNext()) {
+                Property property = iter.nextProperty();
+                if(property.isMultiple()) {
+                    for(final Value v : property.getValues()) {
+                        str += "\t<" + property.getName() + ">" + v.getString() + "</" + property.getName() + ">\n";
+                    }
+                }
+                else {
+                    str += "\t<" + property.getName() + ">" + property.getValue().getString() + "</" + property.getName() + ">\n";
+
+                }
+            }
+
+            str += "</oai_dc:dc>";
+
+            try {
+                return new ByteArrayInputStream(str.getBytes("UTF-8"));
+            } catch (UnsupportedEncodingException e) {
+                e.printStackTrace();
+                return null;
+            }
+        } catch (RepositoryException e) {
+            e.printStackTrace();
+            return null;
+        }
+    }
+}
diff --git a/src/main/java/org/fcrepo/modeshape/indexer/dublincore/IndexFromWellKnownPath.java b/src/main/java/org/fcrepo/modeshape/indexer/dublincore/IndexFromWellKnownPath.java
@@ -0,0 +1,44 @@
+package org.fcrepo.modeshape.indexer.dublincore;
+
+import javax.inject.Inject;
+import javax.jcr.Binary;
+import javax.jcr.Node;
+import javax.jcr.RepositoryException;
+
+import java.io.InputStream;
+
+import static org.modeshape.jcr.api.JcrConstants.JCR_CONTENT;
+import static org.modeshape.jcr.api.JcrConstants.JCR_DATA;
+
+public class IndexFromWellKnownPath extends AbstractIndexer {
+
+    private String wellKnownPath;
+
+    @Override
+    public InputStream getStream(Node node) {
+
+        try {
+            return getContentInputStream(node);
+        } catch (RepositoryException e) {
+            e.printStackTrace();
+
+            return null;
+        }
+    }
+
+    private InputStream getContentInputStream(Node node) throws RepositoryException {
+        if(node.hasNode(this.wellKnownPath)) {
+            final Node dc = node.getNode(this.wellKnownPath);
+
+            Binary binary = dc.getNode(JCR_CONTENT).getProperty(JCR_DATA).getBinary();
+
+            return binary.getStream();
+        } else {
+            return null;
+        }
+    }
+
+    public void setWellKnownPath(String wellKnownPath) {
+        this.wellKnownPath = wellKnownPath;
+    }
+}
diff --git a/src/main/java/org/fcrepo/modeshape/indexer/dublincore/WorstCaseIndexer.java b/src/main/java/org/fcrepo/modeshape/indexer/dublincore/WorstCaseIndexer.java
@@ -0,0 +1,21 @@
+package org.fcrepo.modeshape.indexer.dublincore;
+
+import javax.jcr.Node;
+import java.io.ByteArrayInputStream;
+import java.io.InputStream;
+import java.io.UnsupportedEncodingException;
+
+public class WorstCaseIndexer extends AbstractIndexer {
+
+    @Override
+    public InputStream getStream(Node node) {
+        String str = "<oai_dc:dc xmlns:oai_dc=\"http://www.openarchives.org/OAI/2.0/oai_dc/\" ></oai_dc:dc>";
+
+        try {
+            return new ByteArrayInputStream(str.getBytes("UTF-8"));
+        } catch (UnsupportedEncodingException e) {
+            e.printStackTrace();
+            return null;
+        }
+    }
+}
diff --git a/src/main/resources/fedora-node-types.cnd b/src/main/resources/fedora-node-types.cnd
@@ -16,6 +16,13 @@
  */
 <fedora = 'info:fedora'>
 
+/*
+ * Dublin Core. See:
+ * 
+ * http://dublincore.org/documents/dcmi-namespace/
+ */
+<dc = 'http://purl.org/dc/terms/'>
+
 
 /*
  * Any Fedora resource.
@@ -26,6 +33,11 @@
 	 * Temporary for us until we better understand the use of jcr:created.
 	 */
 	- fedora:created (STRING) COPY
+
+	/*
+	 * See: http://dublincore.org/documents/dcmi-terms/#elements-identifier
+	 */
+	- dc:identifier (STRING) multiple COPY
 
 
 /*

diff --git a/src/main/resources/spring/indexer.xml b/src/main/resources/spring/indexer.xml
@@ -0,0 +1,32 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<beans xmlns="http://www.springframework.org/schema/beans"
+       xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
+       xmlns:context="http://www.springframework.org/schema/context"
+       xsi:schemaLocation="
+       http://www.springframework.org/schema/beans http://www.springframework.org/schema/beans/spring-beans.xsd
+       http://www.springframework.org/schema/context http://www.springframework.org/schema/context/spring-context-3.0.xsd">
+
+
+    <context:annotation-config/>
+
+    <bean class="org.springframework.beans.factory.config.ListFactoryBean">
+        <property name="targetListClass">
+            <value>java.util.LinkedList</value>
+        </property>
+        <property name="sourceList">
+            <list>
+                <ref local="DCfromJcrPropertiesIndexer"/>
+                <ref local="wkpIndexer"/>
+            </list>
+        </property>
+    </bean>
+
+    <bean id="wkpIndexer" class="org.fcrepo.modeshape.indexer.dublincore.IndexFromWellKnownPath">
+         <property name="wellKnownPath" value="DC" />
+     </bean>
+
+
+    <bean id="DCfromJcrPropertiesIndexer" class="org.fcrepo.modeshape.indexer.dublincore.IndexFromJcrProperties" />
+
+
+</beans>
diff --git a/src/main/resources/spring/master.xml b/src/main/resources/spring/master.xml
@@ -9,5 +9,6 @@
     <import resource="classpath:/spring/repo.xml"/>
     <import resource="classpath:/spring/rest.xml"/>
     <import resource="classpath:/spring/eventing.xml"/>
+    <import resource="classpath:/spring/indexer.xml"/>
 
 </beans>
diff --git a/src/main/resources/spring/rest.xml b/src/main/resources/spring/rest.xml
@@ -32,6 +32,7 @@
             <bean class="org.fcrepo.modeshape.FedoraNamespaces"/>
             <bean class="org.fcrepo.modeshape.FedoraObjects"/>
             <bean class="org.fcrepo.modeshape.foxml.FedoraOXML"/>
+            <bean class="org.fcrepo.modeshape.indexer.DublinCore" />
         </jaxrs:serviceBeans>
         <jaxrs:extensionMappings>
             <entry key="json" value="application/json"/>

diff --git a/src/test/java/org/fcrepo/modeshape/indexer/DublinCoreTest.java b/src/test/java/org/fcrepo/modeshape/indexer/DublinCoreTest.java
@@ -0,0 +1,76 @@
+package org.fcrepo.modeshape.indexer;
+
+
+import org.apache.commons.httpclient.HttpClient;
+import org.apache.commons.httpclient.methods.GetMethod;
+import org.apache.commons.httpclient.methods.PostMethod;
+import org.apache.commons.httpclient.methods.StringRequestEntity;
+import org.junit.Test;
+import org.junit.runner.RunWith;
+import org.slf4j.Logger;
+import org.slf4j.LoggerFactory;
+import org.springframework.test.context.ContextConfiguration;
+import org.springframework.test.context.junit4.SpringJUnit4ClassRunner;
+
+import static java.util.regex.Pattern.DOTALL;
+import static java.util.regex.Pattern.compile;
+import static org.junit.Assert.assertEquals;
+import static org.junit.Assert.assertTrue;
+
+@RunWith(SpringJUnit4ClassRunner.class)
+@ContextConfiguration({ "/spring-test/rest.xml", "/spring-test/repo.xml", "/spring-test/indexer.xml" })
+public class DublinCoreTest {
+    private static final int SERVER_PORT = 8080;
+    private static final String HOSTNAME = "localhost";
+    private static final String serverAddress = "http://" + HOSTNAME + ":"
+            + SERVER_PORT + "/";
+
+    final private Logger logger = LoggerFactory
+            .getLogger(DublinCoreTest.class);
+
+    final private HttpClient client = new HttpClient();
+
+    @Test
+    public void testJcrPropertiesBasedOaiDc() throws Exception {
+        PostMethod createObjMethod = new PostMethod(serverAddress
+                + "rest/objects/fdsa");
+        client.executeMethod(createObjMethod);
+
+        GetMethod getWorstCaseOaiMethod = new GetMethod(serverAddress
+                + "rest/objects/fdsa/oai_dc");
+        int status = client.executeMethod(getWorstCaseOaiMethod);
+        assertEquals(200, status);
+
+        final String response = getWorstCaseOaiMethod.getResponseBodyAsString();
+        assertTrue("Didn't find oai_dc!",
+                compile("oai_dc", DOTALL).matcher(response).find());
+
+        assertTrue("Didn't find dc:identifier!",
+                compile("dc:identifier", DOTALL).matcher(response).find());
+    }
+
+    @Test
+    public void testWellKnownPathOaiDc() throws Exception {
+        PostMethod createObjMethod = new PostMethod(serverAddress
+                + "rest/objects/lkjh");
+        client.executeMethod(createObjMethod);
+
+
+        PostMethod createDSMethod = new PostMethod(serverAddress
+                + "rest/objects/lkjh/datastreams/DC");
+
+        createDSMethod.setRequestEntity(new StringRequestEntity(
+                "marbles for everyone", null, null));
+
+        client.executeMethod(createDSMethod);
+
+        GetMethod getWorstCaseOaiMethod = new GetMethod(serverAddress
+                + "rest/objects/lkjh/oai_dc");
+        int status = client.executeMethod(getWorstCaseOaiMethod);
+        assertEquals(200, status);
+
+        final String response = getWorstCaseOaiMethod.getResponseBodyAsString();
+        assertTrue("Didn't find our datastream!",
+                compile("marbles for everyone", DOTALL).matcher(response).find());
+    }
+}