Merge pull request #268 from kit-data-manager/265-store-provenance-in-crates-about-the-tool-which-created-it

265 store provenance in crates about the tool which created it

Author: Pfeil

CITATION.cff

@@ -10,6 +10,6 @@ authors:
   given-names: "Sabrine"
   orcid: "https://orcid.org/0000-0002-4480-6116"
 title: "ro-crate-java"
-version: 1.0.3
+version: 2.1.0-rc3
 date-released: 2022-07-19
 url: "https://github.com/kit-data-manager/ro-crate-java"

build.gradle

@@ -111,6 +111,22 @@ configurations {
     performanceTestImplementation.extendsFrom implementation
 }
 
+// Task for creating a resource file with the version info
+tasks.register("generateVersionProps", WriteProperties) { t ->
+    def generatedResourcesDir = project.layout.buildDirectory.dir(["resources", "main"].join(File.separator))
+    def outputFile = generatedResourcesDir.map { it.file("version.properties") }
+
+    t.destinationFile = outputFile.get().asFile
+    t.property("version", version)
+}
+
+tasks.register("generateVersionPropsTest", WriteProperties) { t ->
+    def generatedResourcesDir = project.layout.buildDirectory.dir(["resources", "test"].join(File.separator))
+    def outputFile = generatedResourcesDir.map { it.file("version.properties") }
+
+    t.destinationFile = outputFile.get().asFile
+    t.property("version", version)
+}
 
 tasks.register('performanceContextEntitiesBenchmark', JavaExec) {
     description = "Run the context entities benchmarks."
@@ -154,8 +170,13 @@ tasks.register('performanceReadWriteMultipleCratesBenchmark', JavaExec) {
     mainClass = 'edu.kit.datamanager.ro_crate.multiplecrates.MultipleCratesWriteAndRead'
 }
 
+compileJava {
+    dependsOn generateVersionProps
+}
+
 test {
     useJUnitPlatform()
+    dependsOn generateVersionPropsTest
     finalizedBy jacocoTestReport
 }
 

src/main/java/edu/kit/datamanager/ro_crate/Crate.java

@@ -22,6 +22,28 @@
  */
 public interface Crate {
 
+  /**
+   * Mark the crate as imported, i.e. it has been read from a file
+   * or is for other reasons not considered a new crate.
+   * <p>
+   * This is useful mostly for readers to indicate this in case
+   * the crate may not have any provenance information yet and
+   * should still be recognized as an imported crate.
+   *
+   * @return this crate, for convenience.
+   */
+  Crate markAsImported();
+
+  /**
+   * Check if the crate is marked as imported.
+   * <p>
+   * If true, it indicates that the crate has been read from a file
+   * or is for other reasons not considered a new crate.
+   *
+   * @return true if the crate is marked as imported, false otherwise.
+   */
+  boolean isImported();
+
   /**
    * Read version from the crate descriptor and return it as a class
    * representation.

src/main/java/edu/kit/datamanager/ro_crate/RoCrate.java

@@ -50,6 +50,24 @@ public class RoCrate implements Crate {
 
     protected Collection<File> untrackedFiles;
 
+    /**
+     * Indicates whether this crate has been imported from an external source.
+     * This is used to determine if ro-crate-java should add a CreateAction
+     * or an UpdateAction in the provenance on export.
+     */
+    protected boolean isImported = false;
+
+    @Override
+    public RoCrate markAsImported() {
+        this.isImported = true;
+        return this;
+    }
+
+    @Override
+    public boolean isImported() {
+        return this.isImported;
+    }
+
     @Override
     public CratePreview getPreview() {
         return this.roCratePreview;

src/main/java/edu/kit/datamanager/ro_crate/entities/AbstractEntity.java

@@ -112,6 +112,19 @@ public JsonNode getProperty(String propertyKey) {
         return this.properties.get(propertyKey);
     }
 
+    /**
+     * Returns the value of the property with the given key as a String.
+     * If the property is not found, it returns null.
+     *
+     * @param propertyKey the key of the property.
+     * @return the value of the property as a String or null if not found.
+     */
+    public String getIdProperty(String propertyKey) {
+        return Optional.ofNullable(this.properties.get(propertyKey))
+                .map(jsonNode -> jsonNode.path("@id").asText(null))
+                .orElse(null);
+    }
+
     @JsonIgnore
     public String getId() {
         JsonNode id = this.properties.get("@id");
@@ -241,9 +254,7 @@ private static boolean addProperty(ObjectNode whereToAdd, String key, JsonNode v
     public void addIdProperty(String name, String id) {
         if (id == null || id.isBlank()) { return; }
         mergeIdIntoValue(id, this.properties.get(name))
-                .ifPresent(newValue -> {
-                    this.properties.set(name, newValue);
-                });
+                .ifPresent(newValue -> this.properties.set(name, newValue));
         this.linkedTo.add(id);
         this.notifyObservers();
     }
@@ -356,7 +367,7 @@ private static void checkFormatISO8601(String date) throws IllegalArgumentExcept
     /**
      * Adds a property with date time format. The property should match the ISO 8601
      * date format.
-     * 
+     * <p>
      * Same as {@link #addProperty(String, String)} but with internal check.
      *
      * @param key   key of the property (e.g. datePublished)
@@ -411,7 +422,7 @@ public T setId(String id) {
                 if (IdentifierUtils.isValidUri(id)) {
                     this.id = id;
                 } else {
-                    this.id = IdentifierUtils.encode(id).get();
+                    this.id = IdentifierUtils.encode(id).orElse(this.id);
                 }
             }
             return self();
@@ -448,7 +459,7 @@ public T addTypes(Collection<String> types) {
         /**
          * Adds a property with date time format. The property should match the ISO 8601
          * date format.
-         * 
+         * <p>
          * Same as {@link #addProperty(String, String)} but with internal check.
          *
          * @param key   key of the property (e.g. datePublished)
@@ -508,7 +519,7 @@ public T addProperty(String key, boolean value) {
         /**
          * ID properties are often used when referencing other entities within
          * the ROCrate. This method adds automatically such one.
-         * 
+         * <p>
          * Instead of {@code "name": "id" }
          * this will add {@code "name" : {"@id": "id"} }
          * 

src/main/java/edu/kit/datamanager/ro_crate/preview/CratePreview.java

@@ -40,6 +40,8 @@ default void generate(Crate crate, File targetDir) throws IOException {
         // as this is usually called in the process of writing a crate
         // (including preview)
         new CrateWriter<>(new WriteFolderStrategy().disablePreview())
+                // We assume the caller (e.g. a writer) already stored the provenance.
+                .withAutomaticProvenance(null)
                 .save(crate, targetDir.getAbsolutePath());
         this.saveAllToFolder(targetDir);
         try (var stream = Files.list(targetDir.toPath())) {

src/main/java/edu/kit/datamanager/ro_crate/reader/CrateReader.java

@@ -97,7 +97,7 @@ public RoCrate readCrate(T location) throws IOException {
         usedFiles.add(files.toPath().resolve(FILE_METADATA_JSON).toFile().getPath());
         usedFiles.add(files.toPath().resolve(FILE_PREVIEW_HTML).toFile().getPath());
         usedFiles.add(files.toPath().resolve(FILE_PREVIEW_FILES).toFile().getPath());
-        return rebuildCrate(metadataJson, files, usedFiles);
+        return rebuildCrate(metadataJson, files, usedFiles).markAsImported();
     }
 
     private RoCrate rebuildCrate(ObjectNode metadataJson, File files, HashSet<String> usedFiles) {

src/main/java/edu/kit/datamanager/ro_crate/util/ClasspathPropertiesVersionProvider.java

@@ -0,0 +1,42 @@
+package edu.kit.datamanager.ro_crate.util;
+
+import java.io.IOException;
+import java.io.InputStream;
+import java.net.URL;
+import java.util.Properties;
+
+public class ClasspathPropertiesVersionProvider implements VersionProvider {
+    public static final String VERSION_PROPERTIES = "version.properties";
+
+    /**
+     * Cached version to avoid repeated file/resource reads.
+     */
+    private String cachedVersion = null;
+
+    /**
+     * Constructs a ClasspathPropertiesVersionProvider that reads the version from a properties file in the classpath.
+     */
+    public ClasspathPropertiesVersionProvider() {
+        this.cachedVersion = getVersion();
+    }
+
+    @Override
+    public String getVersion() {
+        if (cachedVersion != null) {
+            return cachedVersion;
+        }
+
+        URL resource = this.getClass().getResource("/" + VERSION_PROPERTIES);
+        assert resource != null : VERSION_PROPERTIES + " not found in classpath";
+
+        try (InputStream input = resource.openStream()) {
+            Properties properties = new Properties();
+            properties.load(input);
+            String version = properties.getProperty("version");
+            assert version != null : "Version property not found in " + VERSION_PROPERTIES;
+            return version.trim();
+        } catch (IOException e) {
+            throw new IllegalStateException("Failed to read version from properties file", e);
+        }
+    }
+}

src/main/java/edu/kit/datamanager/ro_crate/util/Graph.java

@@ -0,0 +1,61 @@
+package edu.kit.datamanager.ro_crate.util;
+
+import com.fasterxml.jackson.databind.JsonNode;
+
+import java.util.stream.StreamSupport;
+
+/**
+ * Utility class for handling operations on RO-Crate graphs.
+ * Provides methods to find entities by ID or type within a graph.
+ * <p>
+ * {@see JsonUtilFunctions}.
+ */
+public class Graph {
+
+    private Graph() {
+        // Private constructor to prevent instantiation
+    }
+
+    /**
+     * Finds an entity in the graph by its ID.
+     *
+     * @param graph The JSON node representing the graph.
+     * @param id    The ID of the entity to find.
+     * @return The entity as a JsonNode if found, null otherwise.
+     */
+    public static JsonNode findEntityById(JsonNode graph, String id) {
+        for (JsonNode entity : graph) {
+            if (entity.has("@id") && entity.get("@id").asText().equals(id)) {
+                return entity;
+            }
+        }
+        return null;
+    }
+
+    /**
+     * Finds an entity in the graph by its type.
+     *
+     * @param graph The JSON node representing the graph.
+     * @param type  The type of the entity to find.
+     * @return The entity as a JsonNode if found, null otherwise.
+     */
+    public static JsonNode findEntityByType(JsonNode graph, String type) {
+        return StreamSupport.stream(graph.spliterator(), false)
+                .filter(entity -> entity.path("@type").asText().equals(type))
+                .findFirst()
+                .orElse(null);
+    }
+
+    /**
+     * Finds all entities in the graph by their type.
+     *
+     * @param graph The JSON node representing the graph.
+     * @param type  The type of the entities to find.
+     * @return An array of JsonNode containing all entities of the specified type.
+     */
+    public static JsonNode[] findEntitiesByType(JsonNode graph, String type) {
+        return StreamSupport.stream(graph.spliterator(), false)
+                .filter(entity -> entity.path("@type").asText().equals(type))
+                .toArray(JsonNode[]::new);
+    }
+}

src/main/java/edu/kit/datamanager/ro_crate/util/VersionProvider.java

@@ -0,0 +1,10 @@
+package edu.kit.datamanager.ro_crate.util;
+
+public interface VersionProvider {
+    /**
+     * Returns the version of the ro-crate-java library.
+     *
+     * @return The version string.
+     */
+    String getVersion();
+}