Calculate threshold before putting data in deserialization to avoid memory issue

bug 34265, reported by Marcos Cesar de Oliveira

git-svn-id: https://svn.apache.org/repos/asf/jakarta/commons/proper/collections/trunk@169102 13f79535-47bb-0310-9956-ffa450edef68

RELEASE-NOTES.html

@@ -64,6 +64,7 @@ If this causes major headaches to anyone please contact commons-dev at jakarta.a
 <li>AbstractLinkedMap.init() - Now calls createEntry() to create the map entry object [33706]</li>
 <li>BeanMap.initialize() - Internal variable now correctly initialised with only write methods that actually exist [15895]</li>
 <li>TransformedMap.putAll - Now allows putAll of an empty map [34686]</li>
+<li>AbstractHashedMap deserialization - Fix to prevent doubling of internal data array [34265]</li>
 </ul>
 
 <center><h3>JAVADOC</h3></center>

src/java/org/apache/commons/collections/map/AbstractHashedMap.java

@@ -1204,13 +1204,13 @@ public class AbstractHashedMap extends AbstractMap implements IterableMap {
         int capacity = in.readInt();
         int size = in.readInt();
         init();
+        threshold = calculateThreshold(capacity, loadFactor);
         data = new HashEntry[capacity];
         for (int i = 0; i < size; i++) {
             Object key = in.readObject();
             Object value = in.readObject();
             put(key, value);
         }
-        threshold = calculateThreshold(data.length, loadFactor);
     }
     
     //-----------------------------------------------------------------------