Merge pull request #16 from Nuix/continued_memory_refinements

More tweaks aimed at curbing RegexScanner memory usage

Author: JuicyDragon

Java/src/main/java/com/nuix/superutilities/regex/ItemRegexMatchCollection.java

@@ -1,9 +1,11 @@
 package com.nuix.superutilities.regex;
 
+import java.io.IOException;
 import java.util.ArrayList;
 import java.util.List;
 import java.util.stream.Collectors;
 
+import nuix.Case;
 import nuix.Item;
 
 /***
@@ -12,23 +14,28 @@
  *
  */
 public class ItemRegexMatchCollection {
-	private Item item = null;
+	private String itemGuid = null;
 	private List<RegexMatch> matchData = new ArrayList<RegexMatch>();
 
 	/***
 	 * Creates a new empty instance against the specified item.
 	 * @param item The item to associated.
 	 */
 	public ItemRegexMatchCollection(Item item){
-		this.item = item;
+		this.itemGuid = item.getGuid();
 	}
 
 	/***
 	 * Gets the associated item.
 	 * @return The associated item.
 	 */
-	public Item getItem(){
-		return item;
+	public Item getItem(Case nuixCase){
+		try {
+			return nuixCase.search(String.format("guid:%s", itemGuid)).get(0);
+		} catch (IOException e) {
+			e.printStackTrace();
+			return null;
+		}
 	}
 
 	/***
@@ -66,7 +73,8 @@ public List<RegexMatch> getPropertyMatches(){
 	 * @param matchEnd Offset in source text where this match ends
 	 */
 	public void addMatch(PatternInfo patternInfo, String location, boolean isContentMatch, String value, String valueContext, int matchStart, int matchEnd){
-		matchData.add(new RegexMatch(patternInfo,location,isContentMatch,value,valueContext,matchStart,matchEnd));
+		// Intern location in case there is a large amount of duplication of a small set of actual values
+		matchData.add(new RegexMatch(patternInfo,location.intern(),isContentMatch,value,valueContext,matchStart,matchEnd));
 	}
 
 	/***

Java/src/main/java/com/nuix/superutilities/regex/RegexScanError.java

@@ -1,5 +1,8 @@
 package com.nuix.superutilities.regex;
 
+import java.io.IOException;
+
+import nuix.Case;
 import nuix.Item;
 
 /***
@@ -8,7 +11,7 @@
  *
  */
 public class RegexScanError {
-	private Item item = null;
+	private String itemGuid = null;
 	private PatternInfo patternInfo = null;
 	private String location = null;
 	private Exception exception = null;
@@ -21,7 +24,7 @@ public class RegexScanError {
 	 * @param exception The exception which was thrown
 	 */
 	public RegexScanError(Item item, PatternInfo patternInfo, String location, Exception exception){
-		this.item = item;
+		this.itemGuid = item.getGuid();
 		this.patternInfo = patternInfo;
 		this.location = location;
 		this.exception = exception;
@@ -31,8 +34,17 @@ public RegexScanError(Item item, PatternInfo patternInfo, String location, Excep
 	 * Gets the associated item
 	 * @return The associated item
 	 */
-	public Item getItem() {
-		return item;
+	public Item getItem(Case nuixCase) {
+		try {
+			return nuixCase.search(String.format("guid:%s", itemGuid)).get(0);
+		} catch (IOException e) {
+			e.printStackTrace();
+			return null;
+		}
+	}
+	
+	public String getItemGuid() {
+		return this.itemGuid;
 	}
 
 	/***

Java/src/main/java/com/nuix/superutilities/regex/RegexScanner.java

@@ -21,6 +21,7 @@
 import com.nuix.superutilities.misc.FormatUtility;
 
 import nuix.Item;
+import nuix.ItemCustomMetadataMap;
 
 /***
  * Class for scanning a series of items with a series of regular expressions.
@@ -107,7 +108,7 @@ protected void fireScanError(RegexScanError error){
 			if(error.getLocation() != null){
 				errorMessage.add("\tLocation: "+error.getLocation());	
 			}
-			errorMessage.add("\tItem GUID: "+error.getItem().getGuid());
+			errorMessage.add("\tItem GUID: "+error.getItemGuid());
 			logger.error(errorMessage.toString());
 			logger.error(error.getException());
 		}
@@ -444,12 +445,25 @@ protected ItemRegexMatchCollection scanItem(Item item) {
 	 * @return Map of "stringified" metadata properties for the specified item
 	 */
 	public static Map<String,String> getStringProperties(Item item, Set<String> specificProperties){
+		// Note below String.intern use on property names which likely is highly repetitive
+		
 		Map<String,String> result = new HashMap<String,String>();
-		for (Entry<String, Object> entry : item.getProperties().entrySet()) {
-			if(specificProperties == null || specificProperties.contains(entry.getKey())){
-				result.put(entry.getKey(), FormatUtility.getInstance().convertToString(entry.getValue()));
+		
+		if(specificProperties == null | specificProperties.size() == 0) {
+			// We're scanning all the properties
+			for (Entry<String, Object> entry : item.getProperties().entrySet()) {
+				result.put(entry.getKey().intern(), FormatUtility.getInstance().convertToString(entry.getValue()));
+			}
+		} else {
+			// We're just scanning specific properties
+			Map<String,Object> itemProperties =  item.getProperties();
+			for(String specificProperty : specificProperties) {
+				if(itemProperties.containsKey(specificProperty)) {
+					result.put(specificProperty.intern(), FormatUtility.getInstance().convertToString(itemProperties.get(specificProperty)));
+				}
 			}
 		}
+		
 		return result;
 	}
 
@@ -462,11 +476,22 @@ public static Map<String,String> getStringProperties(Item item, Set<String> spec
 	 */
 	public static Map<String,String> getStringCustomMetadata(Item item, Set<String> specificFields){
 		Map<String,String> result = new HashMap<String,String>();
-		for (Entry<String, Object> entry : item.getCustomMetadata().entrySet()) {
-			if(specificFields == null || specificFields.contains(entry.getKey())){
-				result.put(entry.getKey(), FormatUtility.getInstance().convertToString(entry.getValue()));
+		
+		if(specificFields == null || specificFields.size() == 0) {
+			// We're scanning all the custom metadata fields
+			for (Entry<String, Object> entry : item.getCustomMetadata().entrySet()) {
+				result.put(entry.getKey().intern(), FormatUtility.getInstance().convertToString(entry.getValue()));
+			}
+		} else {
+			ItemCustomMetadataMap itemCustomMetadata = item.getCustomMetadata();
+			// We're scanning specific custom metadata fields
+			for(String specificField : specificFields) {
+				if(itemCustomMetadata.containsKey(specificField)) {
+					result.put(specificField.intern(),FormatUtility.getInstance().convertToString(itemCustomMetadata.get(specificField)));
+				}
 			}
 		}
+		
 		return result;
 	}