1.4.0.0 dev

gradle/libs.versions.toml

@@ -1,7 +1,7 @@
 [versions]
 guava_version = "33.5.0-jre"
-junit_version = "6.0.1"
-log4j_version = "2.25.2"
+junit_version = "6.0.2"
+log4j_version = "2.25.3"
 
 [libraries]
 guava = { module = "com.google.guava:guava", version.ref = "guava_version" }

src/main/java/com/wildermods/masshash/Blob.java

@@ -4,6 +4,7 @@
 import java.io.InputStream;
 import java.nio.file.Files;
 import java.nio.file.Path;
+import java.util.Objects;
 
 import com.wildermods.masshash.exception.IntegrityException;
 import com.wildermods.masshash.utils.ByteUtil;
@@ -15,6 +16,11 @@
  */
 public record Blob(byte[] data, String hash) implements IBlob {
 
+	public Blob {
+		Objects.requireNonNull(data);
+		Objects.requireNonNull(hash);
+	}
+
     /**
      * Constructs a Blob from the given data and computes its hash.
      *
@@ -118,23 +124,7 @@ public Blob(InputStream stream, Hash hash) throws IOException, IntegrityExceptio
      * @return A new {@link Hash} object that represents this blob, but  with no associated data.
      */
 	public Hash dropData() {
-		if(isTransient()) {
-			throw new UnsupportedOperationException("Data already dropped!");
-		}
-		return new Blob((byte[])null, hash);
-	}
-
-	/**
-	 * Returns the data associated with this Blob
-	 * 
-	 * @return a byte array that contains the data stored in this blob
-	 */
-	@Override
-	public byte[] data() {
-		if(data == null) {
-			throw new UnsupportedOperationException("Null data! Was the data dropped?");
-		}
-		return data;
+		return Hash.of(hash());
 	}
 
 	@Override
@@ -143,7 +133,7 @@ public int hashCode() {
 	}
 
 	/**
-	 * Compares this object with another Hash object for equality. All {@link Blob} objects are also instances of {@link Hash}.
+	 * Compares this object with another Hash object for equality. All {@link IBlob} objects are also instances of {@link Hash}.
 	 * <p>
 	 * Two {@link Hash} objects are considered equal if their hashes are the same. This method specifically compares
 	 * the hash of the other object with the hash of this object. If the other object is not an instance of {@link Hash},

src/main/java/com/wildermods/masshash/Data.java

@@ -1,5 +1,8 @@
 package com.wildermods.masshash;
 
+import java.io.ByteArrayInputStream;
+import java.io.IOException;
+import java.io.InputStream;
 import java.util.Arrays;
 
 /**
@@ -15,6 +18,21 @@ public interface Data {
 	 */
 	public byte[] data();
 
+	/**
+	 * Returns an {@link InputStream} for reading the data.
+	 * 
+	 * Default implementation wraps {@link data()} in a {@link ByteArrayInputStream}
+	 * Classes that can stream data without holding it in memory should override this.
+	 * 
+	 * @return an {@link InputStream} for the data
+	 */
+	public default InputStream dataStream() throws IOException {
+		if(isTransient()) {
+			throw new IllegalStateException("No data to stream!");
+		}
+		return new ByteArrayInputStream(data());
+	}
+
 	/**
 	 * Checks if the data is transient, meaning the data is null or otherwise unavailable.
 	 * 

src/main/java/com/wildermods/masshash/Hash.java

@@ -41,6 +41,51 @@ public default boolean hashEquals(String hash) {
 	 * @return a new {@link Hash} instance.
 	 */
 	public static Hash of(String hash) {
-		return new Blob((byte[])null, hash);
+		return new Hash() {
+
+			@Override
+			public String hash() {
+				return hash;
+			}
+
+			@Override
+			public int hashCode() {
+				return hash.hashCode();
+			}
+
+			/**
+			 * Compares this object with another Hash object for equality. All {@link IBlob} objects are also instances of {@link Hash}.
+			 * <p>
+			 * Two {@link Hash} objects are considered equal if their hashes are the same. This method specifically compares
+			 * the hash of the other object with the hash of this object. If the other object is not an instance of {@link Hash},
+			 * the method returns {@code false}.
+			 * </p>
+			 * 
+			 * @param o the object to compare with this Hash object.
+			 * @return {@code true} if the other object is a {@link Hash} and has the same hash; {@code false} otherwise.
+			 */
+			@Override
+			public boolean equals(Object o) {
+				if(o instanceof Hash) {
+					return hash().equals(((Hash) o).hash());
+				}
+				return false;
+			}
+
+			/**
+			 * Returns a string representation of this Blob, which is its hash value.
+			 * <p>
+			 * This method overrides the default {@link Object#toString()} method to provide a more meaningful
+			 * string representation of the Blob.
+			 * </p>
+			 * 
+			 * @return the hash of the Blob as a string.
+			 */
+			@Override
+			public String toString() {
+				return hash();
+			}
+
+		};
 	}
 }

src/main/java/com/wildermods/masshash/Hasher.java

@@ -93,7 +93,7 @@ public Hasher(final Stream<Path> files) throws IOException {
 	 *        before being added to the result map. The updated reference value will be associated with the computed hash.
 	 * @throws IOException if an I/O error occurs during hashing
 	 */
-	public Hasher(final Stream<Path> files, final BiConsumer<Reference<Path>, Blob> forEachBlob) throws IOException {
+	public Hasher(final Stream<Path> files, final BiConsumer<Reference<Path>, IBlob> forEachBlob) throws IOException {
 		this(files, (p) -> true, forEachBlob);
 	}
 
@@ -112,7 +112,7 @@ public Hasher(final Stream<Path> files, final BiConsumer<Reference<Path>, Blob>
 	 * @throws IOException if an I/O error occurs during hashing
 	 * @throws IllegalArgumentException if no files match the predicate
 	 */
-	public Hasher(final Stream<Path> files, final Predicate<Path> predicate, final BiConsumer<Reference<Path>, Blob> forEachBlob) throws IOException {
+	public Hasher(final Stream<Path> files, final Predicate<Path> predicate, final BiConsumer<Reference<Path>, IBlob> forEachBlob) throws IOException {
 		this(files, Runtime.getRuntime().availableProcessors(), predicate, forEachBlob);
 	}
 
@@ -147,7 +147,7 @@ public Hasher(final Stream<Path> files, final Predicate<Path> predicate, final B
 	 * @throws IOException if an error occurs while reading files or during thread execution
 	 * @throws IllegalArgumentException if no files matched the provided predicate
 	 */
-	public Hasher(final Stream<Path> files, int threads, final Predicate<Path> predicate, final BiConsumer<Reference<Path>,Blob> forEachBlob) throws IOException {
+	public Hasher(final Stream<Path> files, int threads, final Predicate<Path> predicate, final BiConsumer<Reference<Path>,IBlob> forEachBlob) throws IOException {
 		final int processors = Runtime.getRuntime().availableProcessors();
 		Objects.requireNonNull(files);
 		Objects.requireNonNull(predicate);
@@ -214,12 +214,12 @@ public Hasher(final Stream<Path> files, int threads, final Predicate<Path> predi
 				for (Path file : sublist) {
 					Reference<Path> newFile = new Reference<>(file);
 					//Read and hash the file into a Blob, then discard the Blob’s data to conserve memory
-					Hash blob = new Blob(file);
-					forEachBlob.accept(newFile, (Blob) blob);
-					((Blob) blob).dropData();
+					IBlob blob = LightBlob.from(file);
+					forEachBlob.accept(newFile, (IBlob) blob);
+					Hash hash = blob.dropData();
 
 					//Group files by their content hash. Files with the same hash will share the same key
-					local.computeIfAbsent(blob, k -> new HashSet<>()).add(newFile.get());
+					local.computeIfAbsent(hash, k -> new HashSet<>()).add(newFile.get());
 				}
 				return local;
 			}));

src/main/java/com/wildermods/masshash/IBlob.java

@@ -17,4 +17,13 @@ public interface IBlob extends Data, Hash {
 	 * indicating data corruption or alteration.
 	 */
 	public void verify() throws IntegrityException;
+
+	/**
+	 * Drops the data from the current object and returns a new Hash that represents the hash of this Blob.
+	 * The original blob still holds the data for as long as you keep it referenced.
+	 *
+	 * @return A new {@link Hash} object that represents this blob, but  with no associated data.
+	 */
+	public Hash dropData();
+
 }