prov schema validation #4378

pom.xml

@@ -144,6 +144,13 @@
             <version>1.1</version>
         </dependency>
         <!-- END Data Deposit API v1 (SWORD v2) -->
+        <dependency>
+            <!-- There are later versions of this lib available at jitpack.io,
+                 but it seemed better to not add another repo. -->
+            <groupId>org.everit.json</groupId>
+            <artifactId>org.everit.json.schema</artifactId>
+            <version>1.5.1</version>
+        </dependency>
         <dependency>
             <groupId>org.mindrot</groupId>
             <artifactId>jbcrypt</artifactId>
@@ -452,6 +459,22 @@
     </dependencies>
 
     <build>
+<!--        <testResources>
+            <testResource>
+                <directory>${project.basedir}/src/main/resources</directory>
+            </testResource>
+        </testResources>-->
+<!--        <testResources>
+            <testResource>
+              <directory>${project.basedir}/src/test/java</directory>
+              <excludes>
+                  <exclude>**/*.java</exclude>
+              </excludes>
+            </testResource>
+            <testResource>
+              <directory>${project.basedir}/src/test/resources</directory>
+            </testResource>
+          </testResources>-->
         <resources>
             <resource>
                 <directory>src/main/java</directory>

src/main/java/Bundle.properties

@@ -1796,8 +1796,9 @@ file.editProvenanceDialog.tip=Provenance is a record of the origin of your data
 file.editProvenanceDialog.uploadSuccess=Upload complete
 file.editProvenanceDialog.uploadError=An error occurred during upload and parsing of your provenance file.
 file.editProvenanceDialog.noEntitiesError=The uploaded provenance file does not contain any entities that can be related to your Data File. 
+file.editProvenanceDialog.invalidSchemaError=The uploaded provenance file does not comply with the W3C Provenance standard.
 file.editProvenanceDialog.bundleFile=Provenance File
-file.editProvenanceDialog.bundleFile.instructions=File must be JSON format and follow W3C standards.
+file.editProvenanceDialog.bundleFile.instructions=File must be JSON format and follow the <a href="https://www.w3.org/Submission/2013/SUBM-prov-json-20130424/" target="_blank">W3C standard</a>.
 file.editProvenanceDialog.bundleFile.alreadyPublished=This Provenance File has been published and cannot be replaced or removed.
 file.editProvenanceDialog.bundleEntity=Data File Entity
 file.editProvenanceDialog.bundleEntity.placeholder=Connect entity...

src/main/java/edu/harvard/iq/dataverse/DatasetPage.java

@@ -1,5 +1,6 @@
 package edu.harvard.iq.dataverse;
 
+import edu.harvard.iq.dataverse.provenance.ProvPopupFragmentBean;
 import edu.harvard.iq.dataverse.api.AbstractApiBean;
 import edu.harvard.iq.dataverse.authorization.AuthenticationServiceBean;
 import edu.harvard.iq.dataverse.authorization.Permission;

src/main/java/edu/harvard/iq/dataverse/EditDatafilesPage.java

@@ -1,6 +1,6 @@
 package edu.harvard.iq.dataverse;
 
-import edu.harvard.iq.dataverse.ProvPopupFragmentBean.UpdatesEntry;
+import edu.harvard.iq.dataverse.provenance.ProvPopupFragmentBean;
 import edu.harvard.iq.dataverse.api.AbstractApiBean;
 import edu.harvard.iq.dataverse.authorization.AuthenticationServiceBean;
 import edu.harvard.iq.dataverse.authorization.Permission;
@@ -1162,7 +1162,7 @@ public String save() {
                 if(systemConfig.isProvCollectionEnabled() && provJsonChanges) {
                     HashMap<String,ProvPopupFragmentBean.UpdatesEntry> provenanceUpdates = provPopupFragmentBean.getProvenanceUpdates();
                     for (int i = 0; i < dataset.getFiles().size(); i++) {
-                        for (UpdatesEntry ue : provenanceUpdates.values()) { 
+                        for (ProvPopupFragmentBean.UpdatesEntry ue : provenanceUpdates.values()) { 
                             if (ue.dataFile.getStorageIdentifier() != null ) {
                                 if (ue.dataFile.getStorageIdentifier().equals(dataset.getFiles().get(i).getStorageIdentifier())) {
                                     dataset.getFiles().set(i, ue.dataFile);

src/main/java/edu/harvard/iq/dataverse/api/Prov.java

@@ -1,8 +1,8 @@
 package edu.harvard.iq.dataverse.api;
 
 import edu.harvard.iq.dataverse.DataFile;
-import edu.harvard.iq.dataverse.ProvEntityFileData;
-import edu.harvard.iq.dataverse.ProvUtilFragmentBean;
+import edu.harvard.iq.dataverse.provenance.ProvEntityFileData;
+import edu.harvard.iq.dataverse.provenance.ProvInvestigator;
 import edu.harvard.iq.dataverse.engine.command.DataverseRequest;
 import edu.harvard.iq.dataverse.engine.command.impl.DeleteProvJsonCommand;
 import edu.harvard.iq.dataverse.engine.command.impl.GetProvFreeFormCommand;
@@ -35,14 +35,12 @@ public class Prov extends AbstractApiBean {
 
     private static final Logger logger = Logger.getLogger(Prov.class.getCanonicalName());
 
-    @Inject
-    ProvUtilFragmentBean provUtil;
+    ProvInvestigator provUtil = ProvInvestigator.getInstance();
 
     /** Provenance JSON methods **/
     @POST
     @Path("{id}/prov-json")
     @Consumes("application/json")
-//MAD: SHOULD NOT WORK ON PUBLISHED
     public Response addProvJson(String body, @PathParam("id") String idSupplied, @QueryParam("entityName") String entityName) {
         if(!systemConfig.isProvCollectionEnabled()) {
             return error(FORBIDDEN, BundleUtil.getStringFromBundle("api.prov.error.provDisabled"));
@@ -56,6 +54,10 @@ public Response addProvJson(String body, @PathParam("id") String idSupplied, @Qu
                 return error(FORBIDDEN, BundleUtil.getStringFromBundle("api.prov.error.jsonUpdateNotAllowed"));
             }
 
+            if(!provUtil.isProvValid(body)) {
+                return error(BAD_REQUEST, BundleUtil.getStringFromBundle("file.editProvenanceDialog.invalidSchemaError"));
+            }
+
             /*Add when we actually integrate provCpl*/
             //else if (dataFile.getProvCplId() != 0) {
             //    return error(METHOD_NOT_ALLOWED, "File provenance has already exists in the CPL system and cannot be uploaded.");

src/main/java/edu/harvard/iq/dataverse/engine/command/impl/PersistProvJsonCommand.java

@@ -1,7 +1,6 @@
 package edu.harvard.iq.dataverse.engine.command.impl;
 
 import edu.harvard.iq.dataverse.DataFile;
-import edu.harvard.iq.dataverse.FileMetadata;
 import edu.harvard.iq.dataverse.authorization.Permission;
 import edu.harvard.iq.dataverse.dataaccess.StorageIO;
 import edu.harvard.iq.dataverse.engine.command.AbstractCommand;
@@ -13,12 +12,8 @@
 import java.io.ByteArrayInputStream;
 import java.io.IOException;
 import java.io.InputStream;
-import java.io.StringReader;
 import java.nio.charset.StandardCharsets;
 import java.util.logging.Logger;
-import javax.json.Json;
-import javax.json.JsonException;
-import javax.json.JsonObject;
 
 
 @RequiredPermissions(Permission.EditDataset)
@@ -46,32 +41,9 @@ public DataFile execute(CommandContext ctxt) throws CommandException {
             String error = "A valid entityName must be provided to connect the DataFile to the provenance data.";
             throw new IllegalCommandException(error, this);
         }
-
-       /**
-         * TODO: We are not yet validating the JSON received as PROV-JSON, but
-         * we could some day.
-         *
-         * Section "2.3 Validating PROV-JSON Documents" of the spec says, "A
-         * schema for PROV-JSON is provided, which defines all the valid
-         * PROV-JSON constructs described in this document. The schema was
-         * written using the schema language specified in [JSON-SCHEMA] (Version
-         * 4). It can be used for the purpose of validating PROV-JSON documents.
-         * A number of libraries for JSON schema validation are available at
-         * json-schema.org/implementations.html." It links to
-         * https://www.w3.org/Submission/prov-json/schema
-         * 
-         * The below chunk just validates that the input is basic json.
-         */
-        StringReader rdr = new StringReader(jsonInput);
-        try {
-            JsonObject jsonObj = Json.createReader(rdr).readObject();
-        } catch (JsonException ex) {
-            String error = "A valid JSON object could not be found. PROV-JSON format is expected.";
-            throw new IllegalCommandException(error, this);
-        }
 
         // Write to StorageIO.
-        // "json.json" looks a little redundand but the standard is called PROV-JSON and it's nice to see .json on disk.
+        // "prov-json.json" looks a little redundand but the standard is called PROV-JSON and it's nice to see .json on disk.
         final String provJsonExtension = "prov-json.json";
         try {
             StorageIO<DataFile> storageIO = dataFile.getStorageIO();
@@ -82,7 +54,6 @@ public DataFile execute(CommandContext ctxt) throws CommandException {
             throw new IllegalCommandException(error, this);
         }
 
-
         dataFile.setProvEntityName(entityName);
         if(saveContext) {
             dataFile = ctxt.files().save(dataFile);

src/main/java/edu/harvard/iq/dataverse/ProvEntityFileData.java → src/main/java/edu/harvard/iq/dataverse/provenance/ProvEntityFileData.java

@@ -1,4 +1,4 @@
-package edu.harvard.iq.dataverse;
+package edu.harvard.iq.dataverse.provenance;
 /**
  *
  * @author madunlap

src/main/java/edu/harvard/iq/dataverse/ProvEntityFileDataConverter.java → src/main/java/edu/harvard/iq/dataverse/provenance/ProvEntityFileDataConverter.java

@@ -1,4 +1,4 @@
-package edu.harvard.iq.dataverse;
+package edu.harvard.iq.dataverse.provenance;
 
 import javax.faces.component.UIComponent;
 import javax.faces.context.FacesContext;