Merge pull request #7974 from pkiraly/7431-fixing-invalid-OAI-PMH-response

7431 fixing invalid oai pmh response

Author: kcondon

src/main/java/edu/harvard/iq/dataverse/harvest/server/web/servlet/OAIServlet.java

@@ -46,13 +46,9 @@
 
 import java.io.IOException;
 import java.io.OutputStream;
-import java.io.OutputStreamWriter;
-import java.io.Writer;
 import java.util.Date;
 import java.util.HashMap;
 import java.util.logging.Logger;
-import java.util.zip.DeflaterOutputStream;
-import java.util.zip.GZIPOutputStream;
 import javax.ejb.EJB;
 import javax.mail.internet.InternetAddress;
 import javax.servlet.ServletConfig;

src/main/java/edu/harvard/iq/dataverse/harvest/server/xoai/XlistRecords.java

@@ -1,4 +1,3 @@
-
 package edu.harvard.iq.dataverse.harvest.server.xoai;
 
 import com.lyncode.xml.exceptions.XmlWriteException;

src/main/java/edu/harvard/iq/dataverse/harvest/server/xoai/Xrecord.java

@@ -1,4 +1,3 @@
-
 package edu.harvard.iq.dataverse.harvest.server.xoai;
 
 import com.lyncode.xoai.model.oaipmh.Header;
@@ -17,6 +16,7 @@
 import java.io.OutputStream;
 import java.net.InetAddress;
 import java.net.UnknownHostException;
+import org.apache.poi.util.ReplacingInputStream;
 
 /**
  *
@@ -85,7 +85,11 @@ public void writeToStream(OutputStream outputStream) throws IOException {
                 if (dataset != null && formatName != null) {
                     InputStream inputStream = null;
                     try {
-                        inputStream = ExportService.getInstance().getExport(dataset, formatName);
+                        inputStream = new ReplacingInputStream(
+                            ExportService.getInstance().getExport(dataset, formatName),
+                            "<?xml version=\"1.0\" encoding=\"UTF-8\"?>",
+                            ""
+                        );
                     } catch (ExportException ex) {
                         inputStream = null;
                     }
@@ -101,7 +105,6 @@ public void writeToStream(OutputStream outputStream) throws IOException {
             }
         }
         outputStream.flush();
-
     }
 
     private String itemHeaderToString(Header header) {

src/test/java/edu/harvard/iq/dataverse/api/HarvestingServerIT.java

@@ -18,6 +18,7 @@
 import static com.jayway.restassured.RestAssured.given;
 import java.util.List;
 import static junit.framework.Assert.assertEquals;
+import static org.junit.Assert.assertFalse;
 import static org.junit.Assert.assertNotNull;
 
 /**
@@ -163,6 +164,20 @@ public void testOaiFunctionality() throws InterruptedException {
         // And the record should be the dataset we have just created:
         assertEquals(datasetPersistentId, listIdentifiersResponse.getBody().xmlPath().getString("OAI-PMH.ListIdentifiers.header.identifier"));
 
+        Response listRecordsResponse = UtilIT.getOaiListRecords(setName, "oai_dc");
+        assertEquals(OK.getStatusCode(), listRecordsResponse.getStatusCode());
+        List listRecords = listRecordsResponse.getBody().xmlPath().getList("OAI-PMH.ListRecords.record");
+
+        assertNotNull(listRecords);
+        assertEquals(1, listRecords.size());
+        assertEquals(datasetPersistentId, listRecordsResponse.getBody().xmlPath().getString("OAI-PMH.ListRecords.record[0].header.identifier"));
+
+        // assert that Datacite format does not contain the XML prolog
+        Response listRecordsResponseDatacite = UtilIT.getOaiListRecords(setName, "Datacite");
+        assertEquals(OK.getStatusCode(), listRecordsResponseDatacite.getStatusCode());
+        String body = listRecordsResponseDatacite.getBody().asString();
+        assertFalse(body.contains("<?xml version=\"1.0\" encoding=\"UTF-8\"?>"));
+
         // And now run GetRecord on the OAI record for the dataset:
         Response getRecordResponse = UtilIT.getOaiRecord(datasetPersistentId, "oai_dc");
 

src/test/java/edu/harvard/iq/dataverse/api/UtilIT.java

@@ -2359,7 +2359,12 @@ static Response getOaiListIdentifiers(String setName, String metadataFormat) {
         String apiPath = String.format("/oai?verb=ListIdentifiers&set=%s&metadataPrefix=%s", setName, metadataFormat);
         return given().get(apiPath);
     }
-    
+
+    static Response getOaiListRecords(String setName, String metadataFormat) {
+        String apiPath = String.format("/oai?verb=ListRecords&set=%s&metadataPrefix=%s", setName, metadataFormat);
+        return given().get(apiPath);
+    }
+
     static Response changeAuthenticatedUserIdentifier(String oldIdentifier, String newIdentifier, String apiToken) {
         Response response;
         String path = String.format("/api/users/%s/changeIdentifier/%s", oldIdentifier, newIdentifier );