[GitHub] [commons-collections] Claudenw commented on a diff in pull request #320: Collections 824: Optimize SimpleHasher.forEachIndex and SimpleHasher name change

[GitBox <gi...@apache.org>]

Claudenw commented on code in PR #320:
URL: https://github.com/apache/commons-collections/pull/320#discussion_r939504995


##########
src/main/java/org/apache/commons/collections4/bloomfilter/EnhancedDoubleHasher.java:
##########
@@ -0,0 +1,229 @@
+/*
+ * 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.
+ */
+package org.apache.commons.collections4.bloomfilter;
+
+import java.util.Objects;
+import java.util.function.IntPredicate;
+
+/**
+ * A Hasher that implements combinatorial hashing as as described by
+ * <a href="https://www.eecs.harvard.edu/~michaelm/postscripts/tr-02-05.pdf">Krisch and Mitzenmacher</a> using the enhanced double hashing technique
+ * described in the wikipedia article  <a href="https://en.wikipedia.org/wiki/Double_hashing#Enhanced_double_hashing">Double Hashing</a>.
+ * <p>
+ * Common use for this hasher is to generate bit indices from a byte array output of a hashing
+ * or MessageDigest algorithm.</p>
+ *
+ * <h2>Thoughts on the hasher input</h2>
+ *
+ *<p>Note that it is worse to create smaller numbers for the <code>initial</code> and <code>increment</code>. If the <code>initial</code> is smaller than

Review Comment:
   fixed



##########
src/main/java/org/apache/commons/collections4/bloomfilter/EnhancedDoubleHasher.java:
##########
@@ -0,0 +1,229 @@
+/*
+ * 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.
+ */
+package org.apache.commons.collections4.bloomfilter;
+
+import java.util.Objects;
+import java.util.function.IntPredicate;
+
+/**
+ * A Hasher that implements combinatorial hashing as as described by
+ * <a href="https://www.eecs.harvard.edu/~michaelm/postscripts/tr-02-05.pdf">Krisch and Mitzenmacher</a> using the enhanced double hashing technique
+ * described in the wikipedia article  <a href="https://en.wikipedia.org/wiki/Double_hashing#Enhanced_double_hashing">Double Hashing</a>.
+ * <p>
+ * Common use for this hasher is to generate bit indices from a byte array output of a hashing
+ * or MessageDigest algorithm.</p>
+ *
+ * <h2>Thoughts on the hasher input</h2>
+ *
+ *<p>Note that it is worse to create smaller numbers for the <code>initial</code> and <code>increment</code>. If the <code>initial</code> is smaller than
+ * the number of bits in a filter then hashing will start at the same point when the size increases; likewise the <code>increment</code> will be
+ * the same if it remains smaller than the number of bits in the filter and so the first few indices will be the same if the number of bits
+ * changes (but is still larger than the <code>increment</code>). In a worse case scenario with small <code>initial</code> and <code>increment</code> for
+ * all items, hashing may not create indices that fill the full region within a much larger filter. Imagine hashers created with <code>initial</code>
+ * and <code>increment</code> values less than 255 with a filter size of 30000 and number of hash functions as 5. Ignoring the
+ * tetrahedral addition (a maximum of 20 for k=5) the max index is 255 * 4 + 255 = 1275, this covers 4.25% of the filter. This also
+ * ignores the negative wrapping but the behaviour is the same, some bits cannot be reached.
+ * </p><p>
+ * So this needs to be avoided as the filter probability assumptions will be void. If the <code>initial</code> and <code>increment</code> are larger
+ * than the number of bits then the modulus will create a 'random' position and increment within the size.
+ * </p>
+ *
+ * @since 4.5
+ */
+public class EnhancedDoubleHasher implements Hasher {
+
+    /**
+     * The initial hash value.
+     */
+    private final long initial;
+
+    /**
+     * The value to increment the hash value by.
+     */
+    private final long increment;
+
+    /**
+     * Convert bytes to big-endian long filling with zero bytes as necessary..
+     * @param byteArray the byte array to extract the values from.
+     * @param offset the offset to start extraction from.
+     * @param len the length of the extraction, may be longer than 8.
+     * @return
+     */
+    private static long toLong(byte[] byteArray, int offset, int len) {
+        long val = 0;
+        len = Math.min(len, Long.BYTES);

Review Comment:
   fixed



-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: issues-unsubscribe@commons.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org