summaryrefslogtreecommitdiff
path: root/utils/hashtable.c
blob: 6f06332b72ee1c0d70673f1e63c3a34a91c81245 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
/*
 * This file is part of NetSurf, http://netsurf.sourceforge.net/
 * Licensed under the GNU General Public License,
 *                http://www.opensource.org/licenses/gpl-license
 * Copyright 2006 Rob Kendrick <rjek@rjek.com>
 */

/** Write-Once hash table for string to string mappings */

#include <stdlib.h>
#include <string.h>
#ifdef TEST_RIG
#include <assert.h>
#include <stdio.h>
#endif
#include "netsurf/utils/hashtable.h"
#include "netsurf/utils/log.h"

struct hash_table *hash_create(unsigned int chains)
{
	struct hash_table *r = (struct hash_table *)malloc(
						sizeof(struct hash_table));

	if (r == NULL) {
		LOG(("Not enough memory for hash table."));
		return NULL;
	}

	r->nchains = chains;
	r->chain = (struct hash_entry **)
	  calloc(chains, sizeof(struct hash_entry));

	if (r->chain == NULL) {
		LOG(("Not enough memory for %d hash table chains.", chains));
		free(r);
		return NULL;
	}

	return r;
}

void hash_destroy(struct hash_table *ht)
{
	int i;

	for (i = 0; i < ht->nchains; i++) {
		if (ht->chain[i] != NULL) {
			struct hash_entry *e = ht->chain[i];
			while (e) {
			  	struct hash_entry *n = e->next;
				free(e->key);
				free(e->value);
				free(e);
				e = n;
			}
		}
	}

	free(ht->chain);
	free(ht);
}

void hash_add(struct hash_table *ht, const char *key, const char *value)
{
	unsigned int h = hash_string_fnv(key);
	unsigned int c = h % ht->nchains;
	struct hash_entry *e = (struct hash_entry *)malloc(
						sizeof(struct hash_entry));

	e->key = strdup(key);
	e->value = strdup(value);
	e->next = ht->chain[c];
	ht->chain[c] = e;
}

const char *hash_get(struct hash_table *ht, const char *key)
{
	unsigned int h = hash_string_fnv(key);
	unsigned int c = h % ht->nchains;
	struct hash_entry *e = ht->chain[c];
	
	while (e) {
		if (!strcmp(key, e->key))
			return e->value;
		e = e->next;
	}

	return NULL;

}

unsigned int hash_string_fnv(const char *datum)
{
	unsigned int z = 0x01000193, i = 0;

	while (datum[i]) {
		z *= 0x01000193;
		z ^= datum[i];
		datum++;
	}

	return z;
}

#ifdef TEST_RIG

int main(int argc, char *argv[])
{
	struct hash_table *a, *b;
	FILE *dict;
	char keybuf[BUFSIZ], valbuf[BUFSIZ];

	a = hash_create(79);
	assert(a != NULL);

	b = hash_create(103);
	assert(b != NULL);

	hash_add(a, "cow", "moo");
	hash_add(b, "moo", "cow");

	hash_add(a, "pig", "oink");
	hash_add(b, "oink", "pig");

	hash_add(a, "chicken", "cluck");
	hash_add(b, "cluck", "chicken");

	hash_add(a, "dog", "woof");
	hash_add(b, "woof", "dog");

	hash_add(a, "cat", "meow");
	hash_add(b, "meow", "cat");

#define MATCH(x,y) assert(!strcmp(hash_get(a, x), y)); assert(!strcmp(hash_get(b, y), x))
	MATCH("cow", "moo");
	MATCH("pig", "oink");
	MATCH("chicken", "cluck");
	MATCH("dog", "woof");
	MATCH("cat", "meow");

	hash_destroy(a);
	hash_destroy(b);

	/* this test requires /usr/share/dict/words - a large list of English
	 * words.  We load the entire file - odd lines are used as keys, and
	 * even lines are used as the values for the previous line.  we then
	 * work through it again making sure everything matches.
	 *
	 * We do this twice - once in a hash table with many chains, and once
	 * with a hash table with fewer chains.
	 */

	a = hash_create(1031);
	b = hash_create(7919);
	
	dict = fopen("/usr/share/dict/words", "r");
	if (dict == NULL) {
		fprintf(stderr, "Unable to open /usr/share/dict/words - extensive testing skipped.\n");
		exit(0);
	}

	while (!feof(dict)) {
		fscanf(dict, "%s", keybuf);
		fscanf(dict, "%s", valbuf);
		hash_add(a, keybuf, valbuf);
		hash_add(b, keybuf, valbuf);
	}

	fseek(dict, 0, SEEK_SET);

	while (!feof(dict)) {
		fscanf(dict, "%s", keybuf);
		fscanf(dict, "%s", valbuf);
		assert(strcmp(hash_get(a, keybuf), valbuf) == 0);
		assert(strcmp(hash_get(b, keybuf), valbuf) == 0);
	}

	hash_destroy(a);
	hash_destroy(b);

	fclose(dict);

	return 0;
}

#endif