Merge pull request #14 from Nuix/profile_digester

Update ProfileDigester.java

Java/src/main/java/com/nuix/superutilities/misc/ProfileDigester.java

@@ -2,6 +2,7 @@
 
 import java.nio.charset.Charset;
 import java.security.MessageDigest;
+import java.security.NoSuchAlgorithmException;
 import java.util.Collection;
 import java.util.HashMap;
 import java.util.Map;
@@ -34,6 +35,8 @@ public class ProfileDigester {
 	private Consumer<String> infoMessageCallback = null;
 	private BiConsumer<String,Item> errorCallback = null;
 
+	private MessageDigest md5Digester = null;
+
 	private void fireProgressUpdate(int current, int total) {
 		if(progressCallback != null) {
 			progressCallback.accept(current, total);
@@ -81,8 +84,17 @@ public void whenErrorLogged(BiConsumer<String,Item> callback) {
 		errorCallback = callback;
 	}
 
-	public ProfileDigester() {}
-	public ProfileDigester(MetadataProfile metadataProfile) { profile = metadataProfile; }
+	public ProfileDigester() {
+		try {
+			md5Digester = MessageDigest.getInstance("md5");
+		} catch (NoSuchAlgorithmException e) {
+			e.printStackTrace();
+		}
+	}
+	public ProfileDigester(MetadataProfile metadataProfile) {
+		this();
+		profile = metadataProfile;
+	}
 
 	/***
 	 * Adds items to an item set using "Scripted" deduplication, providing a MD5 hash generated from the concatenation of the values yielded
@@ -131,7 +143,7 @@ public ItemSet addItemsToItemSet(Case nuixCase, String itemSetName, String dedup
 			if(includeItemText) {
 				description = String.format("Generated using MD5 of profile '%s' field values concatenation and Item Text", profile.getName());
 			} else {
-				description = String.format("Generated using MD5 of profile '%s' field values concatenationt", profile.getName());
+				description = String.format("Generated using MD5 of profile '%s' field values concatenation", profile.getName());
 			}
 
 			Map<String,Object> itemSetSettings = new HashMap<String,Object>();
@@ -180,28 +192,34 @@ public void itemProcessed(ItemEventInfo info) {
 	/***
 	 * Generates MD5 digest byte array for a given item.  Digest is generated by digesting concatenation of values yielded by the
 	 * metadata profile associated with this instance for the given item and optionally including the item's content text.
+	 * Note that method is synchronized due to:
+	 * - Reuse of MD5 digester
+	 * - Some metadata profile fields don't seem to play nice when called concurrently
 	 * @param item The item to generate a custom MD5 digest for.
 	 * @return Byte array representation of the MD5 digest
 	 * @throws Exception Most likely if metadata profile has not yet been set for this instance.
 	 */
-	public byte[] generateMd5Bytes(Item item) throws Exception {
+	public synchronized byte[] generateMd5Bytes(Item item) throws Exception {
 		if(profile == null) {
 			throw new IllegalArgumentException("profile cannot be null, please provide a profile by calling setProfile(MetadataProfile profile) before calling this method");
 		}
 
-		MessageDigest md = MessageDigest.getInstance("MD5");
 		for(MetadataItem field : profile.getMetadata()) {
 			String fieldValue = field.evaluate(item);
 			if(fieldValue != null) {
-				md.update(fieldValue.getBytes());
+				md5Digester.update(fieldValue.getBytes());
 			}
 		}
 
 		if(includeItemText) {
-			md.update(item.getTextObject().toString().getBytes(Charset.forName("utf8")));
+			md5Digester.update(item.getTextObject().toString().getBytes(Charset.forName("utf8")));
 		}
 
-		return md.digest();
+		// Capture our result and then cleanup for the next call
+		byte[] result = md5Digester.digest();
+		md5Digester.reset();
+
+		return result;
 	}
 
 	/***