OAI-PMH error handling: meaningful errors in XML for verb processing

doc/release-notes/9275-harvest-invalid-query-params.md

@@ -0,0 +1,4 @@
+OAI-PMH error handling has been improved to display a machine-readable error in XML rather than a 500 error with no further information.
+
+- /oai?foo=bar will show "No argument 'verb' found"
+- /oai?verb=foo&verb=bar will show "Verb must be singular, given: '[foo, bar]'"

docker-compose-dev.yml

@@ -19,6 +19,9 @@ services:
       DATAVERSE_AUTH_OIDC_CLIENT_SECRET: 94XHrfNRwXsjqTqApRrwWmhDLDHpIYV8
       DATAVERSE_AUTH_OIDC_AUTH_SERVER_URL: http://keycloak.mydomain.com:8090/realms/test
       DATAVERSE_JSF_REFRESH_PERIOD: "1"
+      # to get HarvestingServerIT to pass
+      dataverse_oai_server_maxidentifiers: "2"
+      dataverse_oai_server_maxrecords: "2"
       JVM_ARGS: -Ddataverse.files.storage-driver-id=file1
         -Ddataverse.files.file1.type=file
         -Ddataverse.files.file1.label=Filesystem

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

@@ -31,8 +31,11 @@
 import edu.harvard.iq.dataverse.settings.SettingsServiceBean;
 import edu.harvard.iq.dataverse.util.MailUtil;
 import edu.harvard.iq.dataverse.util.SystemConfig;
+import io.gdcc.xoai.exceptions.BadArgumentException;
+import io.gdcc.xoai.exceptions.BadVerbException;
 import io.gdcc.xoai.exceptions.OAIException;
 import io.gdcc.xoai.model.oaipmh.Granularity;
+import io.gdcc.xoai.model.oaipmh.verbs.Verb;
 import io.gdcc.xoai.services.impl.SimpleResumptionTokenFormat;
 import org.apache.commons.lang3.StringUtils;
 
@@ -256,9 +259,18 @@ private void processRequest(HttpServletRequest httpServletRequest, HttpServletRe
                         "Sorry. OAI Service is disabled on this Dataverse node.");
                 return;
             }
-
-            RawRequest rawRequest = RequestBuilder.buildRawRequest(httpServletRequest.getParameterMap());
-
+
+            RawRequest rawRequest = null;
+            try {
+                rawRequest = RequestBuilder.buildRawRequest(httpServletRequest.getParameterMap());
+            } catch (BadVerbException bve) {
+                // Verb.Type is required. Hard-code one.
+                rawRequest = new RawRequest(Verb.Type.Identify);
+                // Ideally, withError would accept a BadVerbException.
+                BadArgumentException bae = new BadArgumentException(bve.getLocalizedMessage());
+                rawRequest.withError(bae);
+            }
+
             OAIPMH handle = dataProvider.handle(rawRequest);
             response.setContentType("text/xml;charset=UTF-8");
 

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

@@ -860,7 +860,40 @@ public void testMultiRecordOaiSet() throws InterruptedException {
         logger.info("deleteResponse.getStatusCode(): " + deleteResponse.getStatusCode());
         assertEquals(200, deleteResponse.getStatusCode(), "Failed to delete the control multi-record set");
     }
-
+
+    @Test
+    public void testInvalidQueryParams() {
+
+        // The query parameter "verb" must appear.
+        Response noVerbArg = given().get("/oai?foo=bar");
+        noVerbArg.prettyPrint();
+        noVerbArg.then().assertThat()
+                .statusCode(OK.getStatusCode())
+                // This should be "badVerb"
+                .body("oai.error.@code", equalTo("badArgument"))
+                .body("oai.error", equalTo("No argument 'verb' found"));
+
+        // The query parameter "verb" cannot appear more than once.
+        Response repeated = given().get( "/oai?verb=foo&verb=bar");
+        repeated.prettyPrint();
+        repeated.then().assertThat()
+                .statusCode(OK.getStatusCode())
+                // This should be "badVerb"
+                .body("oai.error.@code", equalTo("badArgument"))
+                .body("oai.error", equalTo("Verb must be singular, given: '[foo, bar]'"));
+
+    }
+
+    @Test
+    public void testNoSuchSetError() {
+        Response noSuchSet = given().get("/oai?verb=ListIdentifiers&set=census&metadataPrefix=dc");
+        noSuchSet.prettyPrint();
+        noSuchSet.then().assertThat()
+                .statusCode(OK.getStatusCode())
+                .body("oai.error.@code", equalTo("noRecordsMatch"))
+                .body("oai.error", equalTo("Requested set 'census' does not exist"));
+    }
+
     // TODO: 
     // What else can we test? 
     // Some ideas: