netscape-revival
nspr/src/prhash.c
/* -*- Mode: C; tab-width: 4; indent-tabs-mode: nil -*-
*/
/* TODO: 64K limit on win16 */
#include <stdlib.h>
#include <string.h>
#include "prmem.h"
#include "prhash.h"
#include "prglobal.h"
/* Compute the number of buckets in ht */
#define NBUCKETS(ht) (1 << (PR_HASH_BITS - (ht)->shift))
/* The smallest table has 16 buckets */
#define MINBUCKETSLOG2 4
#define MINBUCKETS (1 << MINBUCKETSLOG2)
/* Compute the maximum entries given n buckets that we will tolerate, ~90% */
#define OVERLOADED(n) ((n) - ((n) >> 3))
/* Compute the number of entries below which we shrink the table by half */
#define UNDERLOADED(n) (((n) > MINBUCKETS) ? ((n) >> 2) : 0)
/*
** Stubs for default hash allocator ops.
*/
static void *
DefaultAllocTable(void *pool, size_t size)
{
return malloc(size);
}
static void
DefaultFreeTable(void *pool, void *item)
{
free(item);
}
static PRHashEntry *
DefaultAllocEntry(void *pool)
{
return malloc(sizeof(PRHashEntry));
}
static void
DefaultFreeEntry(void *pool, PRHashEntry *he, int flag)
{
if (flag == HT_FREE_ENTRY)
free(he);
}
static PRHashAllocOps defaultHashAllocOps = {
DefaultAllocTable, DefaultFreeTable,
DefaultAllocEntry, DefaultFreeEntry
};
PR_PUBLIC_API(PRHashTable *)
PR_NewHashTable(uint32 n, PRHashFunction keyHash,
PRHashComparator keyCompare, PRHashComparator valueCompare,
PRHashAllocOps *allocOps, void *allocPool)
{
PRHashTable *ht;
size_t nb;
if (n <= MINBUCKETS) {
n = MINBUCKETSLOG2;
} else {
n = PR_CeilingLog2(n);
if ((int32)n < 0)
return 0;
}
if (!allocOps) allocOps = &defaultHashAllocOps;
ht = (*allocOps->allocTable)(allocPool, sizeof *ht);
if (!ht) return 0;
memset(ht, 0, sizeof *ht);
ht->shift = PR_HASH_BITS - n;
n = 1 << n;
#if defined(XP_PC) && !defined(_WIN32)
if (n > 16000) {
(*allocOps->freeTable)(allocPool, ht);
return 0;
}
#endif /* WIN16 */
nb = n * sizeof(PRHashEntry *);
ht->buckets = (*allocOps->allocTable)(allocPool, nb);
if (!ht->buckets) {
(*allocOps->freeTable)(allocPool, ht);
return 0;
}
memset(ht->buckets, 0, nb);
ht->keyHash = keyHash;
ht->keyCompare = keyCompare;
ht->valueCompare = valueCompare;
ht->allocOps = allocOps;
ht->allocPool = allocPool;
return ht;
}
PR_PUBLIC_API(void)
PR_HashTableDestroy(PRHashTable *ht)
{
uint32 i, n;
PRHashEntry *he, *next;
PRHashAllocOps *allocOps = ht->allocOps;
void *allocPool = ht->allocPool;
n = NBUCKETS(ht);
for (i = 0; i < n; i++) {
for (he = ht->buckets[i]; he; he = next) {
next = he->next;
(*allocOps->freeEntry)(allocPool, he, HT_FREE_ENTRY);
}
}
#ifdef DEBUG
memset(ht->buckets, 0xDB, n * sizeof ht->buckets[0]);
#endif
(*allocOps->freeTable)(allocPool, ht->buckets);
#ifdef DEBUG
memset(ht, 0xDB, sizeof *ht);
#endif
(*allocOps->freeTable)(allocPool, ht);
}
/*
** Multiplicative hash, from Knuth 6.4.
*/
#define GOLDEN_RATIO 0x9E3779B9U
PR_PUBLIC_API(PRHashEntry **)
PR_HashTableRawLookup(PRHashTable *ht, PRHashNumber keyHash, const void *key)
{
PRHashEntry *he, **hep, **hep0;
PRHashNumber h;
#ifdef HASHMETER
ht->nlookups++;
#endif
h = keyHash * GOLDEN_RATIO;
h >>= ht->shift;
hep = hep0 = &ht->buckets[h];
while ((he = *hep) != 0) {
if (he->keyHash == keyHash && (*ht->keyCompare)(key, he->key)) {
/* Move to front of chain if not already there */
if (hep != hep0) {
*hep = he->next;
he->next = *hep0;
*hep0 = he;
}
return hep0;
}
hep = &he->next;
#ifdef HASHMETER
ht->nsteps++;
#endif
}
return hep;
}
PR_PUBLIC_API(PRHashEntry *)
PR_HashTableRawAdd(PRHashTable *ht, PRHashEntry **hep,
PRHashNumber keyHash, const void *key, void *value)
{
uint32 i, n;
PRHashEntry *he, *next, **oldbuckets;
size_t nb;
/* Grow the table if it is overloaded */
n = NBUCKETS(ht);
if (ht->nentries >= OVERLOADED(n)) {
#ifdef HASHMETER
ht->ngrows++;
#endif
ht->shift--;
oldbuckets = ht->buckets;
#if defined(XP_PC) && !defined(_WIN32)
if (2 * n > 16000)
return 0;
#endif /* WIN16 */
nb = 2 * n * sizeof(PRHashEntry *);
ht->buckets = (*ht->allocOps->allocTable)(ht->allocPool, nb);
if (!ht->buckets) {
ht->buckets = oldbuckets;
return 0;
}
memset(ht->buckets, 0, nb);
for (i = 0; i < n; i++) {
for (he = oldbuckets[i]; he; he = next) {
next = he->next;
hep = PR_HashTableRawLookup(ht, he->keyHash, he->key);
PR_ASSERT(*hep == 0);
he->next = 0;
*hep = he;
}
}
#ifdef DEBUG
memset(oldbuckets, 0xDB, n * sizeof oldbuckets[0]);
#endif
(*ht->allocOps->freeTable)(ht->allocPool, oldbuckets);
hep = PR_HashTableRawLookup(ht, keyHash, key);
}
/* Make a new key value entry */
he = (*ht->allocOps->allocEntry)(ht->allocPool);
if (!he) return 0;
he->keyHash = keyHash;
he->key = key;
he->value = value;
he->next = *hep;
*hep = he;
ht->nentries++;
return he;
}
PR_PUBLIC_API(PRHashEntry *)
PR_HashTableAdd(PRHashTable *ht, const void *key, void *value)
{
PRHashNumber keyHash;
PRHashEntry *he, **hep;
keyHash = (*ht->keyHash)(key);
hep = PR_HashTableRawLookup(ht, keyHash, key);
if ((he = *hep) != 0) {
/* Hit; see if values match */
if ((*ht->valueCompare)(he->value, value)) {
/* key,value pair is already present in table */
return he;
}
if (he->value)
(*ht->allocOps->freeEntry)(ht->allocPool, he, HT_FREE_VALUE);
he->value = value;
return he;
}
return PR_HashTableRawAdd(ht, hep, keyHash, key, value);
}
PR_PUBLIC_API(void)
PR_HashTableRawRemove(PRHashTable *ht, PRHashEntry **hep, PRHashEntry *he)
{
uint32 i, n;
PRHashEntry *next, **oldbuckets;
size_t nb;
*hep = he->next;
(*ht->allocOps->freeEntry)(ht->allocPool, he, HT_FREE_ENTRY);
/* Shrink table if it's underloaded */
n = NBUCKETS(ht);
if (--ht->nentries < UNDERLOADED(n)) {
#ifdef HASHMETER
ht->nshrinks++;
#endif
ht->shift++;
oldbuckets = ht->buckets;
nb = n * sizeof(PRHashEntry*) / 2;
ht->buckets = (*ht->allocOps->allocTable)(ht->allocPool, nb);
if (!ht->buckets) {
ht->buckets = oldbuckets;
return;
}
memset(ht->buckets, 0, nb);
for (i = 0; i < n; i++) {
for (he = oldbuckets[i]; he; he = next) {
next = he->next;
hep = PR_HashTableRawLookup(ht, he->keyHash, he->key);
PR_ASSERT(*hep == 0);
he->next = 0;
*hep = he;
}
}
#ifdef DEBUG
memset(oldbuckets, 0xDB, n * sizeof oldbuckets[0]);
#endif
(*ht->allocOps->freeTable)(ht->allocPool, oldbuckets);
}
}
PR_PUBLIC_API(PRBool)
PR_HashTableRemove(PRHashTable *ht, const void *key)
{
PRHashNumber keyHash;
PRHashEntry *he, **hep;
keyHash = (*ht->keyHash)(key);
hep = PR_HashTableRawLookup(ht, keyHash, key);
if ((he = *hep) == 0)
return PR_FALSE;
/* Hit; remove element */
PR_HashTableRawRemove(ht, hep, he);
return PR_TRUE;
}
PR_PUBLIC_API(void *)
PR_HashTableLookup(PRHashTable *ht, const void *key)
{
PRHashNumber keyHash;
PRHashEntry *he, **hep;
keyHash = (*ht->keyHash)(key);
hep = PR_HashTableRawLookup(ht, keyHash, key);
if ((he = *hep) != 0) {
return he->value;
}
return 0;
}
/*
** Iterate over the entries in the hash table calling func for each
** entry found. Stop if "f" says to (return value & PR_ENUMERATE_STOP).
** Return a count of the number of elements scanned.
*/
PR_PUBLIC_API(int)
PR_HashTableEnumerateEntries(PRHashTable *ht, PRHashEnumerator f, void *arg)
{
PRHashEntry *he, **hep;
uint32 i, nbuckets;
int rv, n = 0;
PRHashEntry *todo = 0;
nbuckets = NBUCKETS(ht);
for (i = 0; i < nbuckets; i++) {
hep = &ht->buckets[i];
while ((he = *hep) != 0) {
rv = (*f)(he, n, arg);
n++;
if (rv & (HT_ENUMERATE_REMOVE | HT_ENUMERATE_UNHASH)) {
*hep = he->next;
if (rv & HT_ENUMERATE_REMOVE) {
he->next = todo;
todo = he;
}
} else {
hep = &he->next;
}
if (rv & HT_ENUMERATE_STOP) {
goto out;
}
}
}
out:
hep = &todo;
while ((he = *hep) != 0) {
PR_HashTableRawRemove(ht, hep, he);
}
return n;
}
#ifdef HASHMETER
#include <math.h>
#include <stdio.h>
PR_PUBLIC_API(void)
PR_HashTableDumpMeter(PRHashTable *ht, PRHashEnumerator dump, FILE *fp)
{
double mean, variance;
uint32 nchains, nbuckets;
uint32 i, n, maxChain, maxChainLen;
PRHashEntry *he;
variance = 0;
nchains = 0;
maxChainLen = 0;
nbuckets = NBUCKETS(ht);
for (i = 0; i < nbuckets; i++) {
he = ht->buckets[i];
if (!he)
continue;
nchains++;
for (n = 0; he; he = he->next)
n++;
variance += n * n;
if (n > maxChainLen) {
maxChainLen = n;
maxChain = i;
}
}
mean = (double)ht->nentries / nchains;
variance = fabs(variance / nchains - mean * mean);
fprintf(fp, "\nHash table statistics:\n");
fprintf(fp, " number of lookups: %u\n", ht->nlookups);
fprintf(fp, " number of entries: %u\n", ht->nentries);
fprintf(fp, " number of grows: %u\n", ht->ngrows);
fprintf(fp, " number of shrinks: %u\n", ht->nshrinks);
fprintf(fp, " mean steps per hash: %g\n", (double)ht->nsteps
/ ht->nlookups);
fprintf(fp, "mean hash chain length: %g\n", mean);
fprintf(fp, " standard deviation: %g\n", sqrt(variance));
fprintf(fp, " max hash chain length: %u\n", maxChainLen);
fprintf(fp, " max hash chain: [%u]\n", maxChain);
for (he = ht->buckets[maxChain], i = 0; he; he = he->next, i++)
if ((*dump)(he, i, fp) != HT_ENUMERATE_NEXT)
break;
}
#endif /* HASHMETER */
PR_PUBLIC_API(int)
PR_HashTableDump(PRHashTable *ht, PRHashEnumerator dump, FILE *fp)
{
int count;
count = PR_HashTableEnumerateEntries(ht, dump, fp);
#ifdef HASHMETER
PR_HashTableDumpMeter(ht, dump, fp);
#endif
return count;
}
/*
** Rotating xor-accumulate string hash function. The rotate unit of 4 gave
** best results for a large dictionary of strings used by the Java compiler
** and kipp's homebrew Java runtime. [brendan, 21-May-96]
*/
PR_PUBLIC_API(PRHashNumber)
PR_HashString(const void *key)
{
PRHashNumber h;
const unsigned char *s;
h = 0;
for (s = key; *s; s++)
h = (h >> 28) ^ (h << 4) ^ *s;
return h;
}
PR_PUBLIC_API(int)
PR_CompareStrings(const void *v1, const void *v2)
{
return strcmp(v1, v2) == 0;
}
PR_PUBLIC_API(int)
PR_CompareValues(const void *v1, const void *v2)
{
return v1 == v2;
}