121978: Add ORCID IDs to OAI records

atmire · Dec 31, 2024 · 048b38f · 048b38f
1 parent 5a43e6b
commit 048b38f
Show file tree

Hide file tree

Showing 4 changed files with 245 additions and 0 deletions.
diff --git a/dspace-oai/src/main/java/org/dspace/xoai/util/ItemUtils.java b/dspace-oai/src/main/java/org/dspace/xoai/util/ItemUtils.java
@@ -18,15 +18,21 @@
 import com.lyncode.xoai.util.Base64Utils;
 import org.apache.logging.log4j.LogManager;
 import org.apache.logging.log4j.Logger;
+import org.apache.solr.client.solrj.SolrQuery;
+import org.apache.solr.common.SolrDocumentList;
 import org.dspace.app.util.factory.UtilServiceFactory;
 import org.dspace.app.util.service.MetadataExposureService;
+import org.dspace.authority.AuthoritySearchService;
+import org.dspace.authority.factory.AuthorityServiceFactory;
 import org.dspace.authorize.AuthorizeException;
 import org.dspace.content.Bitstream;
 import org.dspace.content.Bundle;
 import org.dspace.content.Item;
 import org.dspace.content.MetadataField;
 import org.dspace.content.MetadataValue;
 import org.dspace.content.authority.Choices;
+import org.dspace.content.authority.factory.ContentAuthorityServiceFactory;
+import org.dspace.content.authority.service.MetadataAuthorityService;
 import org.dspace.content.factory.ContentServiceFactory;
 import org.dspace.content.service.BitstreamService;
 import org.dspace.content.service.ItemService;
@@ -59,6 +65,13 @@ public class ItemUtils {
 
     private static final ConfigurationService configurationService
             = DSpaceServicesFactory.getInstance().getConfigurationService();
+
+    private static final AuthoritySearchService authoritySearchService
+            = AuthorityServiceFactory.getInstance().getAuthoritySearchService();
+
+    private static final MetadataAuthorityService metadataAuthorityService
+            = ContentAuthorityServiceFactory.getInstance().getMetadataAuthorityService();
+
     /**
      * Default constructor
      */
@@ -223,9 +236,41 @@ private static void fillSchemaElement(Element schema, MetadataValue val) throws
             if (val.getConfidence() != Choices.CF_NOVALUE) {
                 valueElem.getField().add(createValue("confidence", val.getConfidence() + ""));
             }
+            addOrcidIdentifier(valueElem, val);
+        }
+    }
+
+    /**
+     * Adds a ORCID identifier field to the provided Element, if applicable.
+     */
+    private static void addOrcidIdentifier(Element valueElem, MetadataValue mdValue) {
+        // Stop if the metadata value's field isn't enabled for ORCID authorities.
+        if (!isOrcidAuthorityControlled(mdValue.getMetadataField())) {
+            return;
+        }
+
+        try {
+            // Query the authority core using the authority key on the metadata value.
+            SolrQuery query = new SolrQuery();
+            query.setQuery("id:" + mdValue.getAuthority());
+            SolrDocumentList results = authoritySearchService.search(query).getResults();
+
+            if (results.getNumFound() > 0 && results.get(0).get("orcid_id") != null) {
+                // If there is one, add the ORCID identifier to the Element.
+                valueElem.getField().add(createValue("orcid_id", (String) results.get(0).get("orcid_id")));
+            }
+        } catch (Exception e) {
+            throw new RuntimeException(e);
         }
     }
 
+    private static boolean isOrcidAuthorityControlled(MetadataField field) {
+        String solrAuthorAuthority = configurationService.getProperty("choices.plugin." + field.toString('.'));
+        return metadataAuthorityService.isAuthorityControlled(metadataAuthorityService.makeFieldKey(field))
+                && solrAuthorAuthority != null
+                && solrAuthorAuthority.equals("SolrAuthorAuthority");
+    }
+
     /**
      * Utility method to retrieve a structured XML in XOAI format
      * @param context

diff --git a/dspace-oai/src/test/java/org/dspace/xoai/tests/stylesheets/OpenAireXslTest.java b/dspace-oai/src/test/java/org/dspace/xoai/tests/stylesheets/OpenAireXslTest.java
@@ -0,0 +1,69 @@
+/**
+ * The contents of this file are subject to the license and copyright
+ * detailed in the LICENSE and NOTICE files at the root of the source
+ * tree and available online at
+ *
+ * http://www.dspace.org/license/
+ */
+package org.dspace.xoai.tests.stylesheets;
+
+import static org.dspace.xoai.tests.support.XmlMatcherBuilder.xml;
+import static org.hamcrest.MatcherAssert.assertThat;
+import static org.hamcrest.Matchers.containsString;
+
+import org.dspace.xoai.tests.support.XmlMatcherBuilder;
+import org.junit.Test;
+
+public class OpenAireXslTest extends AbstractXSLTest {
+
+    @Test
+    public void openAireCanTransformOrcidAuthors() throws Exception {
+        String result = apply("oai_openaire.xsl").to(resource("xoai-test2.xml"));
+
+        // match title
+        assertThat(result, openAire().withXPath(
+                "/oaire:resource/datacite:titles/datacite:title",
+                containsString("Test Item")));
+
+        // match authors
+        assertThat(result, openAire().withXPath(
+                "/oaire:resource/datacite:creators/datacite:creator[1]/datacite:creatorName",
+                containsString("PlainText, One")));
+        assertThat(result, openAire().withXPath(
+                "/oaire:resource/datacite:creators/datacite:creator[2]/datacite:creatorName",
+                containsString("Smith, John")));
+        assertThat(result, openAire().withXPath(
+                "/oaire:resource/datacite:creators/datacite:creator[3]/datacite:creatorName",
+                containsString("Doe, Jane")));
+        assertThat(result, openAire().withXPath(
+                "/oaire:resource/datacite:creators/datacite:creator[4]/datacite:creatorName",
+                containsString("PlainText, Two")));
+
+        // match ORCID IDs
+        // first author shouldn't have an ORCID ID
+        assertThat(result, openAire().withXPath(
+                "/oaire:resource/datacite:creators/datacite:creator[1]/datacite:nameIdentifier" +
+                        "[@schemeURI='http://orcid.org']", containsString("")));
+        assertThat(result, openAire().withXPath(
+                "/oaire:resource/datacite:creators/datacite:creator[2]/datacite:nameIdentifier" +
+                        "[@schemeURI='http://orcid.org']", containsString("0000-0000-0000-1234")));
+        assertThat(result, openAire().withXPath(
+                "/oaire:resource/datacite:creators/datacite:creator[3]/datacite:nameIdentifier" +
+                        "[@schemeURI='http://orcid.org']", containsString("0000-0000-0000-5678")));
+        // final author shouldn't have an ORCID ID
+        assertThat(result, openAire().withXPath(
+                "/oaire:resource/datacite:creators/datacite:creator[1]/datacite:nameIdentifier" +
+                        "[@schemeURI='http://orcid.org']", containsString("")));
+
+    }
+
+    private XmlMatcherBuilder openAire() {
+        return xml().withNamespace("oaire", "http://namespace.openaire.eu/schema/oaire/")
+                    .withNamespace("datacite", "http://datacite.org/schema/kernel-4")
+                    .withNamespace("dc", "http://purl.org/dc/elements/1.1/")
+                    .withNamespace("doc", "http://www.lyncode.com/xoai")
+                    .withNamespace("rdf", "http://www.w3.org/TR/rdf-concepts/")
+                    .withNamespace("vc", "http://www.w3.org/2007/XMLSchema-versioning")
+                    .withNamespace("xsi", "http://www.w3.org/2001/XMLSchema-instance");
+    }
+}
diff --git a/dspace-oai/src/test/resources/xoai-test2.xml b/dspace-oai/src/test/resources/xoai-test2.xml
@@ -0,0 +1,123 @@
+<?xml version="1.0" encoding="UTF-8" ?>
+<!--
+
+    The contents of this file are subject to the license and copyright
+    detailed in the LICENSE and NOTICE files at the root of the source
+    tree and available online at
+
+    http://www.dspace.org/license/
+
+-->
+<metadata xmlns="http://www.lyncode.com/xoai" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
+          xsi:schemaLocation="http://www.lyncode.com/xoai http://www.lyncode.com/xsd/xoai.xsd">
+    <element name="dc">
+        <element name="contributor">
+            <element name="author">
+                <element name="none">
+                    <field name="value">PlainText, One</field>
+                    <field name="authority">e069e116-6ed5-40dd-85da-8a7d3e30d663</field>
+                    <field name="confidence">-1</field>
+                    <field name="value">Smith, John</field>
+                    <field name="authority">376bbd13-dac5-4a32-9d35-70742e3f84c5</field>
+                    <field name="confidence">600</field>
+                    <field name="orcid_id">0000-0000-0000-1234</field>
+                    <field name="value">Doe, Jane</field>
+                    <field name="authority">102616ab-95dd-4974-af1a-8f587c61eed1</field>
+                    <field name="confidence">600</field>
+                    <field name="orcid_id">0000-0000-0000-5678</field>
+                    <field name="value">PlainText, Two</field>
+                    <field name="authority">a485dbe7-c5de-470f-ae33-ca166e6cdbae</field>
+                    <field name="confidence">-1</field>
+                </element>
+            </element>
+        </element>
+        <element name="date">
+            <element name="accessioned">
+                <element name="none">
+                    <field name="value">2024-12-29</field>
+                </element>
+            </element>
+            <element name="available">
+                <element name="none">
+                    <field name="value">2024-12-31</field>
+                </element>
+            </element>
+            <element name="issued">
+                <element name="none">
+                    <field name="value">2024-12-30</field>
+                </element>
+            </element>
+        </element>
+        <element name="identifier">
+            <element name="uri">
+                <element name="none">
+                    <field name="value">http://localhost:4000/handle/123456789/5</field>
+                </element>
+            </element>
+        </element>
+        <element name="description">
+            <element name="abstract">
+                <element name="none">
+                    <field name="value">This is a test item.</field>
+                </element>
+            </element>
+        </element>
+        <element name="title">
+            <element name="none">
+                <field name="value">Test Item</field>
+            </element>
+        </element>
+    </element>
+    <element name="bundles">
+        <element name="bundle">
+            <field name="name">ORIGINAL</field>
+            <element name="bitstreams">
+                <element name="bitstream">
+                    <field name="name">test.pdf</field>
+                    <field name="originalName">test.pdf</field>
+                    <field name="format">application/pdf</field>
+                    <field name="size">28425</field>
+                    <field name="url">http://localhost:4000/bitstreams/0e26b0db-1714-48e5-9ac0-77499ebe5cd8/download
+                    </field>
+                    <field name="checksum">0cf16e811f0070cd46ba2e43ca956a3d</field>
+                    <field name="checksumAlgorithm">MD5</field>
+                    <field name="sid">1</field>
+                </element>
+            </element>
+        </element>
+        <element name="bundle">
+            <field name="name">LICENSE</field>
+            <element name="bitstreams">
+                <element name="bitstream">
+                    <field name="name">license.txt</field>
+                    <field name="originalName">license.txt</field>
+                    <field name="format">text/plain; charset=utf-8</field>
+                    <field name="size">1748</field>
+                    <field name="url">http://localhost:4000/bitstreams/fc78f4c5-350f-434a-8aa5-a89477994e73/download
+                    </field>
+                    <field name="checksum">8a4605be74aa9ea9d79846c1fba20a33</field>
+                    <field name="checksumAlgorithm">MD5</field>
+                    <field name="sid">1</field>
+                </element>
+            </element>
+        </element>
+    </element>
+    <element name="others">
+        <field name="handle">123456789/5</field>
+        <field name="identifier">oai:localhost:123456789/5</field>
+        <field name="lastModifyDate">2024-12-31 14:35:41.465</field>
+        <element name="access-status">
+            <field name="value">open.access</field>
+        </element>
+    </element>
+    <element name="repository">
+        <field name="url">http://localhost:4000</field>
+        <field name="name">DSpace at My University</field>
+        <field name="mail">[email protected]</field>
+    </element>
+    <element name="license">
+        <field name="bin">
+            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
+        </field>
+    </element>
+</metadata>
diff --git a/dspace/config/crosswalks/oai/metadataFormats/oai_openaire.xsl b/dspace/config/crosswalks/oai/metadataFormats/oai_openaire.xsl
@@ -143,6 +143,8 @@
         <datacite:creators>
             <!-- datacite.creator -->
             <xsl:for-each select="./doc:element/doc:field[@name='value']">
+                <!-- set 'orcid' variable if there is an 'orcid_id' field and if the first preceding 'value' field matches the text of the current node -->
+                <xsl:variable name="orcid" select="../doc:field[@name='orcid_id' and preceding-sibling::doc:field[@name='value'][ 1]/text() = current()/text()]" />
                 <xsl:variable name="isRelatedEntity">
                     <xsl:call-template name="isRelatedEntity">
                         <xsl:with-param name="element" select="."/>
@@ -164,6 +166,12 @@
                             <datacite:creatorName>
                                 <xsl:value-of select="./text()"/>
                             </datacite:creatorName>
+                            <xsl:if test="$orcid">
+                                <!-- https://guidelines.openaire.eu/en/latest/data/field_creator.html#nameidentifier-r -->
+                                <datacite:nameIdentifier nameIdentifierScheme="ORCID" schemeURI="http://orcid.org">
+                                    <xsl:value-of select="$orcid"/>
+                                </datacite:nameIdentifier>
+                            </xsl:if>
                         </datacite:creator>
                     </xsl:otherwise>
                 </xsl:choose>