You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@mahout.apache.org by sr...@apache.org on 2010/01/10 21:30:25 UTC
svn commit: r897700 - in /lucene/mahout/trunk/math/src:
main/java-templates/org/apache/mahout/math/map/
test/java-templates/org/apache/mahout/math/map/
Author: srowen
Date: Sun Jan 10 20:30:25 2010
New Revision: 897700
URL: http://svn.apache.org/viewvc?rev=897700&view=rev
Log:
MAHOUT-239 -- more files
Added:
lucene/mahout/trunk/math/src/main/java-templates/org/apache/mahout/math/map/OpenKeyTypeValueTypeHashMap.java.t
lucene/mahout/trunk/math/src/test/java-templates/org/apache/mahout/math/map/
lucene/mahout/trunk/math/src/test/java-templates/org/apache/mahout/math/map/OpenKeyTypeValueTypeHashMapTest.java.t
Added: lucene/mahout/trunk/math/src/main/java-templates/org/apache/mahout/math/map/OpenKeyTypeValueTypeHashMap.java.t
URL: http://svn.apache.org/viewvc/lucene/mahout/trunk/math/src/main/java-templates/org/apache/mahout/math/map/OpenKeyTypeValueTypeHashMap.java.t?rev=897700&view=auto
==============================================================================
--- lucene/mahout/trunk/math/src/main/java-templates/org/apache/mahout/math/map/OpenKeyTypeValueTypeHashMap.java.t (added)
+++ lucene/mahout/trunk/math/src/main/java-templates/org/apache/mahout/math/map/OpenKeyTypeValueTypeHashMap.java.t Sun Jan 10 20:30:25 2010
@@ -0,0 +1,607 @@
+/**
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements. See the NOTICE file
+ * distributed with this work for additional information
+ * regarding copyright ownership. The ASF licenses this file
+ * to you under the Apache License, Version 2.0 (the
+ * "License"); you may not use this file except in compliance
+ * with the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing,
+ * software distributed under the License is distributed on an
+ * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
+ * KIND, either express or implied. See the License for the
+ * specific language governing permissions and limitations
+ * under the License.
+ */
+
+/*
+Copyright � 1999 CERN - European Organization for Nuclear Research.
+Permission to use, copy, modify, distribute and sell this software and its documentation for any purpose
+is hereby granted without fee, provided that the above copyright notice appear in all copies and
+that both that copyright notice and this permission notice appear in supporting documentation.
+CERN makes no representations about the suitability of this software for any purpose.
+It is provided "as is" without expressed or implied warranty.
+*/
+package org.apache.mahout.math.map;
+
+import java.util.Arrays;
+
+import org.apache.mahout.math.function.${keyTypeCap}${valueTypeCap}Procedure;
+import org.apache.mahout.math.function.${keyTypeCap}Procedure;
+import org.apache.mahout.math.list.${keyTypeCap}ArrayList;
+
+#if (${keyType} != ${valueType})
+import org.apache.mahout.math.list.${valueTypeCap}ArrayList;
+#end
+
+/**
+ * Open hash map from ${keyType} keys to ${valueType} values.
+ **/
+public class Open${keyTypeCap}${valueTypeCap}HashMap extends Abstract${keyTypeCap}${valueTypeCap}Map {
+ protected static final byte FREE = 0;
+ protected static final byte FULL = 1;
+ protected static final byte REMOVED = 2;
+#if (${keyTypeFloating} == 'true')
+#set ($noKeyComment = "${keyTypeCap}.NaN")
+ protected static final ${keyType} NO_KEY_VALUE = ${keyTypeCap}.NaN;
+#else
+#set ($noKeyComment = "0")
+ protected static final ${keyType} NO_KEY_VALUE = 0;
+#end
+
+ /** The hash table keys. */
+ protected ${keyType}[] table;
+
+ /** The hash table values. */
+ protected ${valueType}[] values;
+
+ /** The state of each hash table entry (FREE, FULL, REMOVED). */
+ protected byte[] state;
+
+ /** The number of table entries in state==FREE. */
+ protected int freeEntries;
+
+
+ /** Constructs an empty map with default capacity and default load factors. */
+ public Open${keyTypeCap}${valueTypeCap}HashMap() {
+ this(defaultCapacity);
+ }
+
+ /**
+ * Constructs an empty map with the specified initial capacity and default load factors.
+ *
+ * @param initialCapacity the initial capacity of the map.
+ * @throws IllegalArgumentException if the initial capacity is less than zero.
+ */
+ public Open${keyTypeCap}${valueTypeCap}HashMap(int initialCapacity) {
+ this(initialCapacity, defaultMinLoadFactor, defaultMaxLoadFactor);
+ }
+
+ /**
+ * Constructs an empty map with the specified initial capacity and the specified minimum and maximum load factor.
+ *
+ * @param initialCapacity the initial capacity.
+ * @param minLoadFactor the minimum load factor.
+ * @param maxLoadFactor the maximum load factor.
+ * @throws IllegalArgumentException if <tt>initialCapacity < 0 || (minLoadFactor < 0.0 || minLoadFactor >= 1.0) ||
+ * (maxLoadFactor <= 0.0 || maxLoadFactor >= 1.0) || (minLoadFactor >=
+ * maxLoadFactor)</tt>.
+ */
+ public Open${keyTypeCap}${valueTypeCap}HashMap(int initialCapacity, double minLoadFactor, double maxLoadFactor) {
+ setUp(initialCapacity, minLoadFactor, maxLoadFactor);
+ }
+
+ /** Removes all (key,value) associations from the receiver. Implicitly calls <tt>trimToSize()</tt>. */
+ @Override
+ public void clear() {
+ Arrays.fill(this.state, 0, state.length - 1, FREE);
+ distinct = 0;
+ freeEntries = table.length; // delta
+ trimToSize();
+ }
+
+ /**
+ * Returns a deep copy of the receiver.
+ *
+ * @return a deep copy of the receiver.
+ */
+ @Override
+ public Object clone() {
+ Open${keyTypeCap}${valueTypeCap}HashMap copy = (Open${keyTypeCap}${valueTypeCap}HashMap) super.clone();
+ copy.table = copy.table.clone();
+ copy.values = copy.values.clone();
+ copy.state = copy.state.clone();
+ return copy;
+ }
+
+ /**
+ * Returns <tt>true</tt> if the receiver contains the specified key.
+ *
+ * @return <tt>true</tt> if the receiver contains the specified key.
+ */
+ @Override
+ public boolean containsKey(${keyType} key) {
+ return indexOfKey(key) >= 0;
+ }
+
+ /**
+ * Returns <tt>true</tt> if the receiver contains the specified value.
+ *
+ * @return <tt>true</tt> if the receiver contains the specified value.
+ */
+ @Override
+ public boolean containsValue(${valueType} value) {
+ return indexOfValue(value) >= 0;
+ }
+
+ /**
+ * Ensures that the receiver can hold at least the specified number of associations without needing to allocate new
+ * internal memory. If necessary, allocates new internal memory and increases the capacity of the receiver. <p> This
+ * method never need be called; it is for performance tuning only. Calling this method before <tt>put()</tt>ing a
+ * large number of associations boosts performance, because the receiver will grow only once instead of potentially
+ * many times and hash collisions get less probable.
+ *
+ * @param minCapacity the desired minimum capacity.
+ */
+ @Override
+ public void ensureCapacity(int minCapacity) {
+ if (table.length < minCapacity) {
+ int newCapacity = nextPrime(minCapacity);
+ rehash(newCapacity);
+ }
+ }
+
+ /**
+ * Applies a procedure to each key of the receiver, if any. Note: Iterates over the keys in no particular order.
+ * Subclasses can define a particular order, for example, "sorted by key". All methods which <i>can</i> be expressed
+ * in terms of this method (most methods can) <i>must guarantee</i> to use the <i>same</i> order defined by this
+ * method, even if it is no particular order. This is necessary so that, for example, methods <tt>keys</tt> and
+ * <tt>values</tt> will yield association pairs, not two uncorrelated lists.
+ *
+ * @param procedure the procedure to be applied. Stops iteration if the procedure returns <tt>false</tt>, otherwise
+ * continues.
+ * @return <tt>false</tt> if the procedure stopped before all keys where iterated over, <tt>true</tt> otherwise.
+ */
+ @Override
+ public boolean forEachKey(${keyTypeCap}Procedure procedure) {
+ for (int i = table.length; i-- > 0;) {
+ if (state[i] == FULL) {
+ if (!procedure.apply(table[i])) {
+ return false;
+ }
+ }
+ }
+ return true;
+ }
+
+ /**
+ * Applies a procedure to each (key,value) pair of the receiver, if any. Iteration order is guaranteed to be
+ * <i>identical</i> to the order used by method {@link #forEachKey(${keyTypeCap}Procedure)}.
+ *
+ * @param procedure the procedure to be applied. Stops iteration if the procedure returns <tt>false</tt>, otherwise
+ * continues.
+ * @return <tt>false</tt> if the procedure stopped before all keys where iterated over, <tt>true</tt> otherwise.
+ */
+ @Override
+ public boolean forEachPair(${keyTypeCap}${valueTypeCap}Procedure procedure) {
+ for (int i = table.length; i-- > 0;) {
+ if (state[i] == FULL) {
+ if (!procedure.apply(table[i], values[i])) {
+ return false;
+ }
+ }
+ }
+ return true;
+ }
+
+ /**
+ * Returns the value associated with the specified key. It is often a good idea to first check with {@link
+ * #containsKey(double)} whether the given key has a value associated or not, i.e. whether there exists an association
+ * for the given key or not.
+ *
+ * @param key the key to be searched for.
+ * @return the value associated with the specified key; <tt>0</tt> if no such key is present.
+ */
+ @Override
+ public ${valueType} get(${keyType} key) {
+ int i = indexOfKey(key);
+ if (i < 0) {
+ return 0;
+ } //not contained
+ return values[i];
+ }
+
+ /**
+ * @param key the key to be added to the receiver.
+ * @return the index where the key would need to be inserted, if it is not already contained. Returns -index-1 if the
+ * key is already contained at slot index. Therefore, if the returned index < 0, then it is already contained
+ * at slot -index-1. If the returned index >= 0, then it is NOT already contained and should be inserted at
+ * slot index.
+ */
+ protected int indexOfInsertion(${keyType} key) {
+ ${keyType}[] tab = table;
+ byte[] stat = state;
+ int length = tab.length;
+
+ int hash = HashFunctions.hash(key) & 0x7FFFFFFF;
+ int i = hash % length;
+ int decrement = hash % (length - 2); // double hashing, see http://www.eece.unm.edu/faculty/heileman/hash/node4.html
+ //int decrement = (hash / length) % length;
+ if (decrement == 0) {
+ decrement = 1;
+ }
+
+ // stop if we find a removed or free slot, or if we find the key itself
+ // do NOT skip over removed slots (yes, open addressing is like that...)
+ while (stat[i] == FULL && tab[i] != key) {
+ i -= decrement;
+ //hashCollisions++;
+ if (i < 0) {
+ i += length;
+ }
+ }
+
+ if (stat[i] == REMOVED) {
+ // stop if we find a free slot, or if we find the key itself.
+ // do skip over removed slots (yes, open addressing is like that...)
+ // assertion: there is at least one FREE slot.
+ int j = i;
+ while (stat[i] != FREE && (stat[i] == REMOVED || tab[i] != key)) {
+ i -= decrement;
+ //hashCollisions++;
+ if (i < 0) {
+ i += length;
+ }
+ }
+ if (stat[i] == FREE) {
+ i = j;
+ }
+ }
+
+
+ if (stat[i] == FULL) {
+ // key already contained at slot i.
+ // return a negative number identifying the slot.
+ return -i - 1;
+ }
+ // not already contained, should be inserted at slot i.
+ // return a number >= 0 identifying the slot.
+ return i;
+ }
+
+ /**
+ * @param key the key to be searched in the receiver.
+ * @return the index where the key is contained in the receiver, returns -1 if the key was not found.
+ */
+ protected int indexOfKey(${keyType} key) {
+ ${keyType}[] tab = table;
+ byte[] stat = state;
+ int length = tab.length;
+
+ int hash = HashFunctions.hash(key) & 0x7FFFFFFF;
+ int i = hash % length;
+ int decrement = hash % (length - 2); // double hashing, see http://www.eece.unm.edu/faculty/heileman/hash/node4.html
+ //int decrement = (hash / length) % length;
+ if (decrement == 0) {
+ decrement = 1;
+ }
+
+ // stop if we find a free slot, or if we find the key itself.
+ // do skip over removed slots (yes, open addressing is like that...)
+ while (stat[i] != FREE && (stat[i] == REMOVED || tab[i] != key)) {
+ i -= decrement;
+ //hashCollisions++;
+ if (i < 0) {
+ i += length;
+ }
+ }
+
+ if (stat[i] == FREE) {
+ return -1;
+ } // not found
+ return i; //found, return index where key is contained
+ }
+
+ /**
+ * @param value the value to be searched in the receiver.
+ * @return the index where the value is contained in the receiver, returns -1 if the value was not found.
+ */
+ protected int indexOfValue(${valueType} value) {
+ ${valueType}[] val = values;
+ byte[] stat = state;
+
+ for (int i = stat.length; --i >= 0;) {
+ if (stat[i] == FULL && val[i] == value) {
+ return i;
+ }
+ }
+
+ return -1; // not found
+ }
+
+ /**
+ * Returns the first key the given value is associated with. It is often a good idea to first check with {@link
+ * #containsValue(int)} whether there exists an association from a key to this value. Search order is guaranteed to be
+ * <i>identical</i> to the order used by method {@link #forEachKey(${keyTypeCap}Procedure)}.
+ *
+ * @param value the value to search for.
+ * @return the first key for which holds <tt>get(key) == value</tt>;
+ * returns <tt>${noKeyComment}</tt> if no such key
+ * exists.
+ */
+ @Override
+ public ${keyType} keyOf(${valueType} value) {
+ //returns the first key found; there may be more matching keys, however.
+ int i = indexOfValue(value);
+ if (i < 0) {
+ return NO_KEY_VALUE;
+ }
+ return table[i];
+ }
+
+ /**
+ * Fills all keys contained in the receiver into the specified list. Fills the list, starting at index 0. After this
+ * call returns the specified list has a new size that equals <tt>this.size()</tt>. Iteration order is guaranteed to
+ * be <i>identical</i> to the order used by method {@link #forEachKey(${keyTypeCap}Procedure)}.
+ * <p> This method can be used
+ * to iterate over the keys of the receiver.
+ *
+ * @param list the list to be filled, can have any size.
+ */
+ @Override
+ public void keys(${keyTypeCap}ArrayList list) {
+ list.setSize(distinct);
+ ${keyType} [] elements = list.elements();
+
+ ${keyType} [] tab = table;
+ byte[] stat = state;
+
+ int j = 0;
+ for (int i = tab.length; i-- > 0;) {
+ if (stat[i] == FULL) {
+ elements[j++] = tab[i];
+ }
+ }
+ }
+
+ /**
+ * Fills all pairs satisfying a given condition into the specified lists. Fills into the lists, starting at index 0.
+ * After this call returns the specified lists both have a new size, the number of pairs satisfying the condition.
+ * Iteration order is guaranteed to be <i>identical</i> to the order used by method {@link
+ * #forEachKey(${keyTypeCap}Procedure)}. <p> <b>Example:</b> <br>
+ * <pre>
+ * ${keyTypeCap}${valueTypeCap}Procedure condition = new ${keyTypeCap}${valueTypeCap}Procedure() { // match even values only
+ * public boolean apply(${keyType} key, ${valueType} value) { return value%2==0; }
+ * }
+ * keys = (8,7,6), values = (1,2,2) --> keyList = (6,8), valueList = (2,1)</tt>
+ * </pre>
+ *
+ * @param condition the condition to be matched. Takes the current key as first and the current value as second
+ * argument.
+ * @param keyList the list to be filled with keys, can have any size.
+ * @param valueList the list to be filled with values, can have any size.
+ */
+ @Override
+ public void pairsMatching(${keyTypeCap}${valueTypeCap}Procedure condition,
+ ${keyTypeCap}ArrayList keyList,
+ ${valueTypeCap}ArrayList valueList) {
+ keyList.clear();
+ valueList.clear();
+
+ for (int i = table.length; i-- > 0;) {
+ if (state[i] == FULL && condition.apply(table[i], values[i])) {
+ keyList.add(table[i]);
+ valueList.add(values[i]);
+ }
+ }
+ }
+
+ /**
+ * Associates the given key with the given value. Replaces any old <tt>(key,someOtherValue)</tt> association, if
+ * existing.
+ *
+ * @param key the key the value shall be associated with.
+ * @param value the value to be associated.
+ * @return <tt>true</tt> if the receiver did not already contain such a key; <tt>false</tt> if the receiver did
+ * already contain such a key - the new value has now replaced the formerly associated value.
+ */
+ @Override
+ public boolean put(${keyType} key, ${valueType} value) {
+ int i = indexOfInsertion(key);
+ if (i < 0) { //already contained
+ i = -i - 1;
+ this.values[i] = value;
+ return false;
+ }
+
+ if (this.distinct > this.highWaterMark) {
+ int newCapacity = chooseGrowCapacity(this.distinct + 1, this.minLoadFactor, this.maxLoadFactor);
+ /*
+ log.info("grow rehashing ");
+ log.info("at distinct="+distinct+", capacity="+table.length+" to newCapacity="+newCapacity+" ...");
+ */
+ rehash(newCapacity);
+ return put(key, value);
+ }
+
+ this.table[i] = key;
+ this.values[i] = value;
+ if (this.state[i] == FREE) {
+ this.freeEntries--;
+ }
+ this.state[i] = FULL;
+ this.distinct++;
+
+ if (this.freeEntries < 1) { //delta
+ int newCapacity = chooseGrowCapacity(this.distinct + 1, this.minLoadFactor, this.maxLoadFactor);
+ rehash(newCapacity);
+ }
+
+ return true;
+ }
+
+ /**
+ * Rehashes the contents of the receiver into a new table with a smaller or larger capacity. This method is called
+ * automatically when the number of keys in the receiver exceeds the high water mark or falls below the low water
+ * mark.
+ */
+ protected void rehash(int newCapacity) {
+ int oldCapacity = table.length;
+ //if (oldCapacity == newCapacity) return;
+
+ ${keyType}[] oldTable = table;
+ ${valueType}[] oldValues = values;
+ byte[] oldState = state;
+
+ ${keyType}[] newTable = new ${keyType}[newCapacity];
+ ${valueType}[] newValues = new ${valueType}[newCapacity];
+ byte[] newState = new byte[newCapacity];
+
+ this.lowWaterMark = chooseLowWaterMark(newCapacity, this.minLoadFactor);
+ this.highWaterMark = chooseHighWaterMark(newCapacity, this.maxLoadFactor);
+
+ this.table = newTable;
+ this.values = newValues;
+ this.state = newState;
+ this.freeEntries = newCapacity - this.distinct; // delta
+
+ for (int i = oldCapacity; i-- > 0;) {
+ if (oldState[i] == FULL) {
+ ${keyType} element = oldTable[i];
+ int index = indexOfInsertion(element);
+ newTable[index] = element;
+ newValues[index] = oldValues[i];
+ newState[index] = FULL;
+ }
+ }
+ }
+
+ /**
+ * Removes the given key with its associated element from the receiver, if present.
+ *
+ * @param key the key to be removed from the receiver.
+ * @return <tt>true</tt> if the receiver contained the specified key, <tt>false</tt> otherwise.
+ */
+ @Override
+ public boolean removeKey(${keyType} key) {
+ int i = indexOfKey(key);
+ if (i < 0) {
+ return false;
+ } // key not contained
+
+ this.state[i] = REMOVED;
+ //this.values[i]=0; // delta
+ this.distinct--;
+
+ if (this.distinct < this.lowWaterMark) {
+ int newCapacity = chooseShrinkCapacity(this.distinct, this.minLoadFactor, this.maxLoadFactor);
+ /*
+ if (table.length != newCapacity) {
+ log.info("shrink rehashing ");
+ log.info("at distinct="+distinct+", capacity="+table.length+" to newCapacity="+newCapacity+" ...");
+ }
+ */
+ rehash(newCapacity);
+ }
+
+ return true;
+ }
+
+ /**
+ * Initializes the receiver.
+ *
+ * @param initialCapacity the initial capacity of the receiver.
+ * @param minLoadFactor the minLoadFactor of the receiver.
+ * @param maxLoadFactor the maxLoadFactor of the receiver.
+ * @throws IllegalArgumentException if <tt>initialCapacity < 0 || (minLoadFactor < 0.0 || minLoadFactor >= 1.0) ||
+ * (maxLoadFactor <= 0.0 || maxLoadFactor >= 1.0) || (minLoadFactor >=
+ * maxLoadFactor)</tt>.
+ */
+ @Override
+ protected void setUp(int initialCapacity, double minLoadFactor, double maxLoadFactor) {
+ int capacity = initialCapacity;
+ super.setUp(capacity, minLoadFactor, maxLoadFactor);
+ capacity = nextPrime(capacity);
+ if (capacity == 0) {
+ capacity = 1;
+ } // open addressing needs at least one FREE slot at any time.
+
+ this.table = new ${keyType}[capacity];
+ this.values = new ${valueType}[capacity];
+ this.state = new byte[capacity];
+
+ // memory will be exhausted long before this pathological case happens, anyway.
+ this.minLoadFactor = minLoadFactor;
+ if (capacity == PrimeFinder.largestPrime) {
+ this.maxLoadFactor = 1.0;
+ } else {
+ this.maxLoadFactor = maxLoadFactor;
+ }
+
+ this.distinct = 0;
+ this.freeEntries = capacity; // delta
+
+ // lowWaterMark will be established upon first expansion.
+ // establishing it now (upon instance construction) would immediately make the table shrink upon first put(...).
+ // After all the idea of an "initialCapacity" implies violating lowWaterMarks when an object is young.
+ // See ensureCapacity(...)
+ this.lowWaterMark = 0;
+ this.highWaterMark = chooseHighWaterMark(capacity, this.maxLoadFactor);
+ }
+
+ /**
+ * Trims the capacity of the receiver to be the receiver's current size. Releases any superfluous internal memory. An
+ * application can use this operation to minimize the storage of the receiver.
+ */
+ @Override
+ public void trimToSize() {
+ // * 1.2 because open addressing's performance exponentially degrades beyond that point
+ // so that even rehashing the table can take very long
+ int newCapacity = nextPrime((int) (1 + 1.2 * size()));
+ if (table.length > newCapacity) {
+ rehash(newCapacity);
+ }
+ }
+
+ /**
+ * Fills all values contained in the receiver into the specified list. Fills the list, starting at index 0. After this
+ * call returns the specified list has a new size that equals <tt>this.size()</tt>. Iteration order is guaranteed to
+ * be <i>identical</i> to the order used by method {@link #forEachKey(${keyTypeCap}Procedure)}.
+ * <p> This method can be used
+ * to iterate over the values of the receiver.
+ *
+ * @param list the list to be filled, can have any size.
+ */
+ @Override
+ public void values(${valueTypeCap}ArrayList list) {
+ list.setSize(distinct);
+ ${valueType}[] elements = list.elements();
+
+ ${valueType}[] val = values;
+ byte[] stat = state;
+
+ int j = 0;
+ for (int i = stat.length; i-- > 0;) {
+ if (stat[i] == FULL) {
+ elements[j++] = val[i];
+ }
+ }
+ }
+
+ /**
+ * Access for unit tests.
+ * @param capacity
+ * @param minLoadFactor
+ * @param maxLoadFactor
+ */
+ void getInternalFactors(int[] capacity,
+ double[] minLoadFactor,
+ double[] maxLoadFactor) {
+ capacity[0] = table.length;
+ minLoadFactor[0] = this.minLoadFactor;
+ maxLoadFactor[0] = this.maxLoadFactor;
+ }
+}
Added: lucene/mahout/trunk/math/src/test/java-templates/org/apache/mahout/math/map/OpenKeyTypeValueTypeHashMapTest.java.t
URL: http://svn.apache.org/viewvc/lucene/mahout/trunk/math/src/test/java-templates/org/apache/mahout/math/map/OpenKeyTypeValueTypeHashMapTest.java.t?rev=897700&view=auto
==============================================================================
--- lucene/mahout/trunk/math/src/test/java-templates/org/apache/mahout/math/map/OpenKeyTypeValueTypeHashMapTest.java.t (added)
+++ lucene/mahout/trunk/math/src/test/java-templates/org/apache/mahout/math/map/OpenKeyTypeValueTypeHashMapTest.java.t Sun Jan 10 20:30:25 2010
@@ -0,0 +1,383 @@
+/**
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements. See the NOTICE file
+ * distributed with this work for additional information
+ * regarding copyright ownership. The ASF licenses this file
+ * to you under the Apache License, Version 2.0 (the
+ * "License"); you may not use this file except in compliance
+ * with the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing,
+ * software distributed under the License is distributed on an
+ * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
+ * KIND, either express or implied. See the License for the
+ * specific language governing permissions and limitations
+ * under the License.
+ */
+
+#if (${keyTypeFloating} == 'true')
+#set ($keyEpsilon = ", (${keyType})0.000001")
+#else
+#set ($keyEpsilon = "")
+#end
+#if (${valueTypeFloating} == 'true')
+#set ($valueEpsilon = ", (${valueType})0.000001")
+#else
+#set ($valueEpsilon = "")
+#end
+
+ package org.apache.mahout.math.map;
+
+import java.util.ArrayList;
+import java.util.Arrays;
+import java.util.Collections;
+import java.util.List;
+
+import org.apache.mahout.math.function.${keyTypeCap}${valueTypeCap}Procedure;
+import org.apache.mahout.math.function.${keyTypeCap}Procedure;
+import org.apache.mahout.math.list.${keyTypeCap}ArrayList;
+#if (${keyType} != ${valueType})
+import org.apache.mahout.math.list.${valueTypeCap}ArrayList;
+#end
+import org.junit.Assert;
+import org.junit.Test;
+
+public class Open${keyTypeCap}${valueTypeCap}HashMapTest extends Assert {
+
+
+ @Test
+ public void testConstructors() {
+ Open${keyTypeCap}${valueTypeCap}HashMap map = new Open${keyTypeCap}${valueTypeCap}HashMap();
+ int[] capacity = new int[1];
+ double[] minLoadFactor = new double[1];
+ double[] maxLoadFactor = new double[1];
+
+ map.getInternalFactors(capacity, minLoadFactor, maxLoadFactor);
+ assertEquals(AbstractMap.defaultCapacity, capacity[0]);
+ assertEquals(AbstractMap.defaultMaxLoadFactor, maxLoadFactor[0], 0.001);
+ assertEquals(AbstractMap.defaultMinLoadFactor, minLoadFactor[0], 0.001);
+ int prime = PrimeFinder.nextPrime(907);
+ map = new Open${keyTypeCap}${valueTypeCap}HashMap(prime);
+
+ map.getInternalFactors(capacity, minLoadFactor, maxLoadFactor);
+ assertEquals(prime, capacity[0]);
+ assertEquals(AbstractMap.defaultMaxLoadFactor, maxLoadFactor[0], 0.001);
+ assertEquals(AbstractMap.defaultMinLoadFactor, minLoadFactor[0], 0.001);
+
+ map = new Open${keyTypeCap}${valueTypeCap}HashMap(prime, 0.4, 0.8);
+ map.getInternalFactors(capacity, minLoadFactor, maxLoadFactor);
+ assertEquals(prime, capacity[0]);
+ assertEquals(0.4, minLoadFactor[0], 0.001);
+ assertEquals(0.8, maxLoadFactor[0], 0.001);
+ }
+
+ @Test
+ public void testEnsureCapacity() {
+ Open${keyTypeCap}${valueTypeCap}HashMap map = new Open${keyTypeCap}${valueTypeCap}HashMap();
+ int prime = PrimeFinder.nextPrime(907);
+
+ map.ensureCapacity(prime);
+ int[] capacity = new int[1];
+ double[] minLoadFactor = new double[1];
+ double[] maxLoadFactor = new double[1];
+
+ map.getInternalFactors(capacity, minLoadFactor, maxLoadFactor);
+ assertEquals(prime, capacity[0]);
+ }
+
+ @Test
+ public void testClear() {
+ Open${keyTypeCap}${valueTypeCap}HashMap map = new Open${keyTypeCap}${valueTypeCap}HashMap();
+ map.put((${keyType}) 11, (${valueType}) 22);
+ assertEquals(1, map.size());
+ map.clear();
+ assertEquals(0, map.size());
+ assertEquals(0, map.get((${keyType}) 11), 0.0000001);
+ }
+
+ @Test
+ public void testClone() {
+ Open${keyTypeCap}${valueTypeCap}HashMap map = new Open${keyTypeCap}${valueTypeCap}HashMap();
+ map.put((${keyType}) 11, (${valueType}) 22);
+ Open${keyTypeCap}${valueTypeCap}HashMap map2 = (Open${keyTypeCap}${valueTypeCap}HashMap) map.clone();
+ map.clear();
+ assertEquals(1, map2.size());
+ }
+
+ @Test
+ public void testContainsKey() {
+ Open${keyTypeCap}${valueTypeCap}HashMap map = new Open${keyTypeCap}${valueTypeCap}HashMap();
+ map.put(($keyType) 11, (${valueType}) 22);
+ assertTrue(map.containsKey(($keyType) 11));
+ assertFalse(map.containsKey(($keyType) 12));
+ }
+
+ @Test
+ public void testContainValue() {
+ Open${keyTypeCap}${valueTypeCap}HashMap map = new Open${keyTypeCap}${valueTypeCap}HashMap();
+ map.put(($keyType) 11, (${valueType}) 22);
+ assertTrue(map.containsValue((${valueType}) 22));
+ assertFalse(map.containsValue((${valueType}) 23));
+ }
+
+ @Test
+ public void testForEachKey() {
+ final ${keyTypeCap}ArrayList keys = new ${keyTypeCap}ArrayList();
+ Open${keyTypeCap}${valueTypeCap}HashMap map = new Open${keyTypeCap}${valueTypeCap}HashMap();
+ map.put(($keyType) 11, (${valueType}) 22);
+ map.put(($keyType) 12, (${valueType}) 23);
+ map.put(($keyType) 13, (${valueType}) 24);
+ map.put(($keyType) 14, (${valueType}) 25);
+ map.removeKey(($keyType) 13);
+ map.forEachKey(new ${keyTypeCap}Procedure() {
+
+ @Override
+ public boolean apply(${keyType} element) {
+ keys.add(element);
+ return true;
+ }
+ });
+
+ ${keyType}[] keysArray = keys.toArray(new ${keyType}[keys.size()]);
+ Arrays.sort(keysArray);
+
+ assertArrayEquals(new ${keyType}[] {11, 12, 14}, keysArray ${keyEpsilon});
+ }
+
+ private static class Pair implements Comparable<Pair> {
+ ${keyType} k;
+ ${valueType} v;
+
+ Pair(${keyType} k, ${valueType} v) {
+ this.k = k;
+ this.v = v;
+ }
+
+ @Override
+ public int compareTo(Pair o) {
+ if (k < o.k) {
+ return -1;
+ } else if (k == o.k) {
+ return 0;
+ } else {
+ return 1;
+ }
+ }
+ }
+
+ @Test
+ public void testForEachPair() {
+ final List<Pair> pairs = new ArrayList<Pair>();
+ Open${keyTypeCap}${valueTypeCap}HashMap map = new Open${keyTypeCap}${valueTypeCap}HashMap();
+ map.put(($keyType) 11, (${valueType}) 22);
+ map.put(($keyType) 12, (${valueType}) 23);
+ map.put(($keyType) 13, (${valueType}) 24);
+ map.put(($keyType) 14, (${valueType}) 25);
+ map.removeKey(($keyType) 13);
+ map.forEachPair(new ${keyTypeCap}${valueTypeCap}Procedure() {
+
+ @Override
+ public boolean apply(${keyType} first, ${valueType} second) {
+ pairs.add(new Pair(first, second));
+ return true;
+ }
+ });
+
+ Collections.sort(pairs);
+ assertEquals(3, pairs.size());
+ assertEquals(($keyType) 11, pairs.get(0).k ${keyEpsilon});
+ assertEquals((${valueType}) 22, pairs.get(0).v ${valueEpsilon});
+ assertEquals(($keyType) 12, pairs.get(1).k ${keyEpsilon});
+ assertEquals((${valueType}) 23, pairs.get(1).v ${valueEpsilon});
+ assertEquals(($keyType) 14, pairs.get(2).k ${keyEpsilon});
+ assertEquals((${valueType}) 25, pairs.get(2).v ${valueEpsilon});
+
+ pairs.clear();
+ map.forEachPair(new ${keyTypeCap}${valueTypeCap}Procedure() {
+ int count = 0;
+
+ @Override
+ public boolean apply(${keyType} first, ${valueType} second) {
+ pairs.add(new Pair(first, second));
+ count++;
+ return count < 2;
+ }
+ });
+
+ assertEquals(2, pairs.size());
+ }
+
+ @Test
+ public void testGet() {
+ Open${keyTypeCap}${valueTypeCap}HashMap map = new Open${keyTypeCap}${valueTypeCap}HashMap();
+ map.put(($keyType) 11, (${valueType}) 22);
+ map.put(($keyType) 12, (${valueType}) 23);
+ assertEquals(22, map.get(($keyType)11) ${valueEpsilon});
+ assertEquals(0, map.get(($keyType)0) ${valueEpsilon});
+ }
+
+ /*
+ * Note that the javadoc says 'first' but the order
+ * is not defined.
+ */
+ @Test
+ public void testKeyOf() {
+ Open${keyTypeCap}${valueTypeCap}HashMap map = new Open${keyTypeCap}${valueTypeCap}HashMap();
+ map.put(($keyType) 11, (${valueType}) 22);
+ map.put(($keyType) 12, (${valueType}) 22);
+ ${keyType} k = map.keyOf((${valueType})22);
+ assertTrue(k == 11 || k == 12);
+ k = map.keyOf((${valueType})101);
+#if (${keyTypeFloating} == 'true')
+ assertTrue(${keyTypeCap}.isNaN(k));
+#else
+ assertEquals(0, k ${keyEpsilon});
+#end
+ }
+
+ @Test
+ public void testKeys() {
+ Open${keyTypeCap}${valueTypeCap}HashMap map = new Open${keyTypeCap}${valueTypeCap}HashMap();
+ map.put(($keyType) 11, (${valueType}) 22);
+ map.put(($keyType) 12, (${valueType}) 22);
+ ${keyTypeCap}ArrayList keys = new ${keyTypeCap}ArrayList();
+ map.keys(keys);
+ keys.sort();
+ assertEquals(11, keys.get(0) ${keyEpsilon});
+ assertEquals(12, keys.get(1) ${keyEpsilon});
+ ${keyTypeCap}ArrayList k2 = map.keys();
+ k2.sort();
+ assertEquals(keys, k2);
+ }
+
+ @Test
+ public void testPairsMatching() {
+ ${keyTypeCap}ArrayList keyList = new ${keyTypeCap}ArrayList();
+ ${valueTypeCap}ArrayList valueList = new ${valueTypeCap}ArrayList();
+ Open${keyTypeCap}${valueTypeCap}HashMap map = new Open${keyTypeCap}${valueTypeCap}HashMap();
+ map.put(($keyType) 11, (${valueType}) 22);
+ map.put(($keyType) 12, (${valueType}) 23);
+ map.put(($keyType) 13, (${valueType}) 24);
+ map.put(($keyType) 14, (${valueType}) 25);
+ map.removeKey(($keyType) 13);
+ map.pairsMatching(new ${keyTypeCap}${valueTypeCap}Procedure() {
+
+ @Override
+ public boolean apply(${keyType} first, ${valueType} second) {
+ return (first % 2) == 0;
+ }},
+ keyList, valueList);
+ keyList.sort();
+ valueList.sort();
+ assertEquals(2, keyList.size());
+ assertEquals(2, valueList.size());
+ assertEquals(12, keyList.get(0) ${keyEpsilon});
+ assertEquals(14, keyList.get(1) ${keyEpsilon});
+ assertEquals(23, valueList.get(0) ${valueEpsilon});
+ assertEquals(25, valueList.get(1) ${valueEpsilon});
+ }
+
+ @Test
+ public void testValues() {
+ Open${keyTypeCap}${valueTypeCap}HashMap map = new Open${keyTypeCap}${valueTypeCap}HashMap();
+ map.put(($keyType) 11, (${valueType}) 22);
+ map.put(($keyType) 12, (${valueType}) 23);
+ map.put(($keyType) 13, (${valueType}) 24);
+ map.put(($keyType) 14, (${valueType}) 25);
+ map.removeKey(($keyType) 13);
+ ${valueTypeCap}ArrayList values = new ${valueTypeCap}ArrayList(100);
+ map.values(values);
+ assertEquals(3, values.size());
+ values.sort();
+ assertEquals(22, values.get(0) ${valueEpsilon});
+ assertEquals(23, values.get(1) ${valueEpsilon});
+ assertEquals(25, values.get(2) ${valueEpsilon});
+ }
+
+ // tests of the code in the abstract class
+
+ @Test
+ public void testCopy() {
+ Open${keyTypeCap}${valueTypeCap}HashMap map = new Open${keyTypeCap}${valueTypeCap}HashMap();
+ map.put(($keyType) 11, (${valueType}) 22);
+ Open${keyTypeCap}${valueTypeCap}HashMap map2 = (Open${keyTypeCap}${valueTypeCap}HashMap) map.copy();
+ map.clear();
+ assertEquals(1, map2.size());
+ }
+
+ @Test
+ public void testEquals() {
+ // since there are no other subclasses of
+ // Abstractxxx available, we have to just test the
+ // obvious.
+ Open${keyTypeCap}${valueTypeCap}HashMap map = new Open${keyTypeCap}${valueTypeCap}HashMap();
+ map.put(($keyType) 11, (${valueType}) 22);
+ map.put(($keyType) 12, (${valueType}) 23);
+ map.put(($keyType) 13, (${valueType}) 24);
+ map.put(($keyType) 14, (${valueType}) 25);
+ map.removeKey(($keyType) 13);
+ Open${keyTypeCap}${valueTypeCap}HashMap map2 = (Open${keyTypeCap}${valueTypeCap}HashMap) map.copy();
+ assertTrue(map.equals(map2));
+ assertTrue(map2.equals(map));
+ assertFalse("Hello Sailor".equals(map));
+ assertFalse(map.equals("hello sailor"));
+ map2.removeKey(($keyType) 11);
+ assertFalse(map.equals(map2));
+ assertFalse(map2.equals(map));
+ }
+
+ // keys() tested in testKeys
+
+ @Test
+ public void testKeysSortedByValue() {
+ Open${keyTypeCap}${valueTypeCap}HashMap map = new Open${keyTypeCap}${valueTypeCap}HashMap();
+ map.put(($keyType) 11, (${valueType}) 22);
+ map.put(($keyType) 12, (${valueType}) 23);
+ map.put(($keyType) 13, (${valueType}) 24);
+ map.put(($keyType) 14, (${valueType}) 25);
+ map.removeKey(($keyType) 13);
+ ${keyTypeCap}ArrayList keys = new ${keyTypeCap}ArrayList();
+ map.keysSortedByValue(keys);
+ ${keyType}[] keysArray = keys.toArray(new ${keyType}[keys.size()]);
+ assertArrayEquals(new ${keyType}[] {11, 12, 14},
+ keysArray ${keyEpsilon});
+ }
+
+ @Test
+ public void testPairsSortedByKey() {
+ Open${keyTypeCap}${valueTypeCap}HashMap map = new Open${keyTypeCap}${valueTypeCap}HashMap();
+ map.put(($keyType) 11, (${valueType}) 100);
+ map.put(($keyType) 12, (${valueType}) 70);
+ map.put(($keyType) 13, (${valueType}) 30);
+ map.put(($keyType) 14, (${valueType}) 3);
+
+ ${keyTypeCap}ArrayList keys = new ${keyTypeCap}ArrayList();
+ ${valueTypeCap}ArrayList values = new ${valueTypeCap}ArrayList();
+ map.pairsSortedByKey(keys, values);
+
+ assertEquals(4, keys.size());
+ assertEquals(4, values.size());
+ assertEquals(($keyType) 11, keys.get(0) ${keyEpsilon});
+ assertEquals((${valueType}) 100, values.get(0) ${valueEpsilon});
+ assertEquals(($keyType) 12, keys.get(1) ${keyEpsilon});
+ assertEquals((${valueType}) 70, values.get(1) ${valueEpsilon});
+ assertEquals(($keyType) 13, keys.get(2) ${keyEpsilon});
+ assertEquals((${valueType}) 30, values.get(2) ${valueEpsilon});
+ assertEquals(($keyType) 14, keys.get(3) ${keyEpsilon});
+ assertEquals((${valueType}) 3, values.get(3) ${valueEpsilon});
+ keys.clear();
+ values.clear();
+ map.pairsSortedByValue(keys, values);
+ assertEquals(($keyType) 11, keys.get(3) ${keyEpsilon});
+ assertEquals((${valueType}) 100, values.get(3) ${valueEpsilon});
+ assertEquals(($keyType) 12, keys.get(2) ${keyEpsilon});
+ assertEquals((${valueType}) 70, values.get(2) ${valueEpsilon});
+ assertEquals(($keyType) 13, keys.get(1) ${keyEpsilon});
+ assertEquals((${valueType}) 30, values.get(1) ${valueEpsilon});
+ assertEquals(($keyType) 14, keys.get(0) ${keyEpsilon});
+ assertEquals(($valueType) 3, values.get(0) ${valueEpsilon});
+ }
+
+ }