Changeset 17090 for stable/HeuristicLab.Problems.DataAnalysis.Symbolic/3.4/Hashing/HashUtil.cs

Timestamp:

07/05/19 15:44:04 (5 years ago)

Author:

abeham

Message:

#2950: merged revisions 16218, 16252, 16255, 16258, 16259, 16260, 16261, 16263, 16267, 16270, 16271, 16272, 16273, 16284, 16290, 16291, 16302, 16305, 16382, 16478 to stable

Location:

stable

Files:

• . (modified) (1 prop)
• HeuristicLab.Problems.DataAnalysis.Symbolic (modified) (1 prop)
• HeuristicLab.Problems.DataAnalysis.Symbolic/3.4/Hashing (copied) (copied from trunk/HeuristicLab.Problems.DataAnalysis.Symbolic/3.4/Hashing)
• HeuristicLab.Problems.DataAnalysis.Symbolic/3.4/Hashing/HashUtil.cs (modified) (8 diffs)

stable/HeuristicLab.Problems.DataAnalysis.Symbolic/3.4/Hashing/HashUtil.cs

@@ -20 +20 @@
 #endregion
 
+
 using System;
 using System.Security.Cryptography;
@@ -28 +29 @@
 
     // A simple hash function from Robert Sedgwicks Algorithms in C book.I've added some simple optimizations to the algorithm in order to speed up its hashing process. 
-    public static int RSHash(int[] input) {
+    public static ulong RSHash(byte[] input) {
       const int b = 378551;
-      int a = 63689;
-      int hash = 0;
+      ulong a = 63689;
+      ulong hash = 0;
 
       foreach (var v in input) {
@@ -41 +42 @@
 
     // A bitwise hash function written by Justin Sobel
-    public static int JSHash(int[] input) {
-      int hash = 1315423911;
+    public static ulong JSHash(byte[] input) {
+      ulong hash = 1315423911;
       for (int i = 0; i < input.Length; ++i)
         hash ^= (hash << 5) + input[i] + (hash >> 2);
@@ -49 +50 @@
 
     // This hash function comes from Brian Kernighan and Dennis Ritchie's book "The C Programming Language". It is a simple hash function using a strange set of possible seeds which all constitute a pattern of 31....31...31 etc, it seems to be very similar to the DJB hash function. 
-    public static int BKDRHash(int[] input) {
-      const int seed = 131;
-      int hash = 0;
+    public static ulong BKDRHash(byte[] input) {
+      ulong seed = 131;
+      ulong hash = 0;
       foreach (var v in input) {
         hash = (hash * seed) + v;
@@ -59 +60 @@
 
     // This is the algorithm of choice which is used in the open source SDBM project. The hash function seems to have a good over-all distribution for many different data sets. It seems to work well in situations where there is a high variance in the MSBs of the elements in a data set. 
-    public static int SDBMHash(int[] input) {
-      int hash = 0;
+    public static ulong SDBMHash(byte[] input) {
+      ulong hash = 0;
       foreach (var v in input) {
         hash = v + (hash << 6) + (hash << 16) - hash;
@@ -68 +69 @@
 
     // An algorithm produced by Professor Daniel J. Bernstein and shown first to the world on the usenet newsgroup comp.lang.c. It is one of the most efficient hash functions ever published. 
-    public static int DJBHash(int[] input) {
-      int hash = 5381;
+    public static ulong DJBHash(byte[] input) {
+      ulong hash = 5381;
       foreach (var v in input) {
         hash = (hash << 5) + hash + v;
@@ -77 +78 @@
 
     // An algorithm proposed by Donald E.Knuth in The Art Of Computer Programming Volume 3, under the topic of sorting and search chapter 6.4. 
-    public static int DEKHash(int[] input) {
-      int hash = input.Length;
+    public static ulong DEKHash(byte[] input) {
+      ulong hash = (ulong)input.Length;
       foreach (var v in input) {
         hash = (hash << 5) ^ (hash >> 27) ^ v;
@@ -85 +86 @@
     }
 
-    public static int CryptoHash(HashAlgorithm ha, int[] input) {
-      return BitConverter.ToInt32(ha.ComputeHash(input.ToByteArray()), 0);
-    }
-
-    public static byte[] ToByteArray(this int[] input) {
-      const int size = sizeof(int);
-      var bytes = new byte[input.Length * sizeof(int)];
-      for (int i = 0; i < input.Length; ++i) {
-        Array.Copy(BitConverter.GetBytes(bytes[i]), 0, bytes, i * size, size);
-      }
-      return bytes;
+    public static ulong CryptoHash(HashAlgorithm ha, byte[] input) {
+      return BitConverter.ToUInt64(ha.ComputeHash(input), 0);
     }
   }