Hash functions

src/hash_functions.c

@@ -0,0 +1,84 @@
+#include <stdlib.h>
+#include <stdint.h>
+#include <stdio.h>
+#include <string.h>
+#include <stdbool.h>
+
+#include "hash_functions.h"
+#include "memory.h"
+
+size_t hash_map_default_hash_func(const void *key, size_t capacity, int len) {
+	return *((size_t *) key) % capacity;
+}
+
+size_t xPear16(const void *x_copy, size_t capacity, int len) {
+	int h, i, j;
+	unsigned char ch;
+	size_t hex;
+
+	// to store h values
+	struct {
+		int a;
+	} hh[8];
+
+	// 256 values 0-255 in any (random) order suffices
+	struct DAT {
+		int a;
+	}
+
+	T[256] = {
+		98,  6, 85, 150, 36, 23, 112, 164, 135, 207, 169,  5, 26, 64, 165, 219, //  1
+		61, 20, 68, 89, 130, 63, 52, 102, 24, 229, 132, 245, 80, 216, 195, 115, //  2
+		90, 168, 156, 203, 177, 120,  2, 190, 188,  7, 100, 185, 174, 243, 162, 10, //  3
+		237, 18, 253, 225,  8, 208, 172, 244, 255, 126, 101, 79, 145, 235, 228, 121, //  4
+		123, 251, 67, 250, 161,  0, 107, 97, 241, 111, 181, 82, 249, 33, 69, 55, //  5
+		59, 153, 29,  9, 213, 167, 84, 93, 30, 46, 94, 75, 151, 114, 73, 222, //  6
+		197, 96, 210, 45, 16, 227, 248, 202, 51, 152, 252, 125, 81, 206, 215, 186, //  7
+		39, 158, 178, 187, 131, 136,  1, 49, 50, 17, 141, 91, 47, 129, 60, 99, //  8
+		154, 35, 86, 171, 105, 34, 38, 200, 147, 58, 77, 118, 173, 246, 76, 254, //  9
+		133, 232, 196, 144, 198, 124, 53,  4, 108, 74, 223, 234, 134, 230, 157, 139, // 10
+		189, 205, 199, 128, 176, 19, 211, 236, 127, 192, 231, 70, 233, 88, 146, 44, // 11
+		183, 201, 22, 83, 13, 214, 116, 109, 159, 32, 95, 226, 140, 220, 57, 12, // 12
+		221, 31, 209, 182, 143, 92, 149, 184, 148, 62, 113, 65, 37, 27, 106, 166, // 13
+		3, 14, 204, 72, 21, 41, 56, 66, 28, 193, 40, 217, 25, 54, 179, 117, // 14
+		238, 87, 240, 155, 180, 170, 242, 212, 191, 163, 78, 218, 137, 194, 175, 110, // 15
+		43, 119, 224, 71, 122, 142, 42, 160, 104, 48, 247, 103, 15, 11, 138, 239 // 16
+	};
+
+	char *x = safe_malloc(len);
+	memcpy(x, x_copy, len);
+
+	int r;
+	ch = x[0]; // save first byte
+	for (j = 0 ; j < 8; j++) {
+		// standard Pearson hash (output is h)
+		h = 0;
+
+		for (i = 0; i < len; i++) {
+			r = abs(h ^ x[i]);
+			h = T[r].a;
+		}
+
+		hh[j].a = h;	// store result
+		x[0] = x[0] + 1;	// increment first data byte by 1
+	}
+
+	x[0] = ch; // restore first byte
+
+	safe_free(x);
+
+	// concatenate the 8 stored values of h
+	/*sprintf(hex,"%02X%02X%02X%02X%02X%02X%02X%02X",
+	        hh[0].a, hh[1].a,
+	        hh[2].a, hh[3].a,
+	        hh[4].a, hh[5].a,
+	        hh[6].a, hh[7].a);*/
+
+	// For now let me do it this way, but Liu doesn't think this is a good thing
+	for (int i = 0; i < 8; i++) {
+		hex = hex << 8;
+		hex = hex & hh[i].a;
+	}
+
+	return hex; // output 64-bit 16 hex bytes hash
+}

src/hash_functions.h

@@ -0,0 +1,27 @@
+/**
+ * @file
+ * Implementation of different hash functions.
+ */
+
+#ifndef HASH_FUNCTIONS_H
+#define HASH_FUNCTIONS_H
+
+/**
+ * The default hash function. This is the function to be used if the user does not explicitly provide any hash function. It is a basic <EM> mod </EM> operator.
+ * @param key key to hash
+ * @param capacity maximum size of the map
+ * @param len the size of the key passed as the first argument
+ * @return an offset within the range `[0, capacity)`
+ */
+size_t hash_map_default_hash_func(const void *key, size_t capacity, int len);
+
+/**
+ * The Pearson hash function. Both the explanation and code are derived from <A HREF="http://en.wikipedia.org/wiki/Pearson_hashing">Wikipedia</A>
+ * @param key key to hash
+ * @param capacity maximum size of the map
+ * @param len the size of the key passed as the first argument
+ * @return an offset within the range `[0, capacity)`
+ */
+size_t xPear16(const void *x, size_t capacity, int len);
+
+#endif /*HASH_FUNCTIONS_H*/

src/hash_map.c

@@ -2,11 +2,13 @@
 #include <stdint.h>
 #include <string.h>
 #include <stdbool.h>
+#include <stdio.h>
 
 #include "memory.h"
 #include "linked_list.h"
 
 #include "hash_map.h"
+#include "hash_functions.h"
 
 // when "-g" is used redefine inline to static or else the linker will try to link the function and fail (since it's inline)
 #ifdef TEST
@@ -17,11 +19,7 @@ inline int hash_map_default_comparator(const void *l, const void *r) {
 	return *((unsigned long *) l) - *((unsigned long *) r);
 }
 
-inline size_t hash_map_default_hash_func(const void *key, size_t capacity) {
-	return *((size_t *) key) % capacity;
-}
-
-void hash_map_init(hash_map *map, size_t capacity, hash_map_comparator comparator, hash_map_hash_func hash_func) {
+void hash_map_init(hash_map *map, size_t capacity, hash_map_comparator comparator, hash_map_hash_func hash_func, hash_map_key_size key_size) {
 	map->capacity = capacity;
 	map->size = 0;
 
@@ -40,6 +38,8 @@ void hash_map_init(hash_map *map, size_t capacity, hash_map_comparator comparato
 		map->hash_func = hash_map_default_hash_func;
 	}
 
+	map->key_size = key_size;
+
 	map->keys = (linked_list *) safe_malloc(sizeof(linked_list));
 	// No free_data func here because keys will be free'd by linked_list_free for **table
 	linked_list_init(map->keys, NULL);
@@ -60,7 +60,7 @@ void hash_map_free(hash_map *map) {
 }
 
 void *hash_map_get(hash_map *map, void *key) {
-	linked_list *list = map->table[map->hash_func(key, map->capacity)];
+	linked_list *list = map->table[map->hash_func(key, map->capacity, map->key_size(key))];
 
 	if (!list) {
 		return NULL;
@@ -82,12 +82,13 @@ void *hash_map_get(hash_map *map, void *key) {
 }
 
 void hash_map_put(hash_map *map, void *key, void *value) {
-	linked_list *list = map->table[map->hash_func(key, map->capacity)];
+	linked_list *list = map->table[map->hash_func(key, map->capacity, map->key_size(key))];
 
 	if (!list) {
 		list = (linked_list *) safe_malloc(sizeof(linked_list));
+
 		linked_list_init(list, (linked_list_destructor) safe_free);
-		map->table[map->hash_func(key, map->capacity)] = list;
+		map->table[map->hash_func(key, map->capacity, map->key_size(key))] = list;
 	}
 
 	linked_list_node *head = linked_list_head(list);
@@ -118,7 +119,7 @@ void hash_map_put(hash_map *map, void *key, void *value) {
 }
 
 void hash_map_remove(hash_map *map, void *key) {
-	size_t offset = map->hash_func(key, map->capacity);
+	size_t offset = map->hash_func(key, map->capacity, map->key_size(key));
 	linked_list *list = map->table[offset];
 
 	if (!list) {
@@ -175,7 +176,7 @@ void hash_map_clear(hash_map *map) {
 }
 
 bool hash_map_contains_key(hash_map *map, void *key) {
-	linked_list *list = map->table[map->hash_func(key, map->capacity)];
+	linked_list *list = map->table[map->hash_func(key, map->capacity, map->key_size(key))];
 
 	if (!list) {
 		return false;

src/hash_map.h

@@ -22,9 +22,16 @@ typedef int (*hash_map_comparator)(const void *l, const void *r);
  * Hash function
  * @param key key to hash
  * @param capacity maximum size of the map
+ * @param len the size of the key passed as the first argument
  * @return an offset within the range `[0, capacity)`
  */
-typedef size_t (*hash_map_hash_func)(const void *key, size_t capacity);
+typedef size_t (*hash_map_hash_func)(const void *key, size_t capacity, int len);
+
+/**
+ * Size function to determine the size of each keys in the hash function
+ * @return the size of each key`
+ */
+typedef size_t (*hash_map_key_size)();
 
 /**
  * Hash map object
@@ -40,6 +47,8 @@ typedef struct {
 	hash_map_comparator comparator;
 	/** Key hash function */
 	hash_map_hash_func hash_func;
+	/** Key size function */
+	hash_map_key_size key_size;
 	/** Keys */
 	linked_list *keys;
 } hash_map;
@@ -61,7 +70,7 @@ typedef struct {
  * @param comparator key comparator function
  * @param hash_func key hash function
  */
-void hash_map_init(hash_map *map, size_t capacity, hash_map_comparator comparator, hash_map_hash_func hash_func);
+void hash_map_init(hash_map *map, size_t capacity, hash_map_comparator comparator, hash_map_hash_func hash_func, hash_map_key_size);
 
 /**
  * Free the hash map.

src/memory.c

@@ -1,5 +1,4 @@
 #include "memory.h"
-
 #include <stdio.h>
 #include <stdlib.h>