aboutsummaryrefslogtreecommitdiff
path: root/common/hashmap.h
blob: 87ed1b11730b346364545d8ae9ff6bd62768411c (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
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
/* ScummVM - Scumm Interpreter
 * Copyright (C) 2006 The ScummVM project
 *
 * This program is free software; you can redistribute it and/or
 * modify it under the terms of the GNU General Public License
 * as published by the Free Software Foundation; either version 2
 * of the License, or (at your option) any later version.

 * This program is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 * GNU General Public License for more details.

 * You should have received a copy of the GNU General Public License
 * along with this program; if not, write to the Free Software
 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
 *
 * $URL$
 * $Id$
 *
 */

// The hash map (associative array) implementation in this file is
// based on code by Andrew Y. Ng, 1996:

/* 
 * Copyright (c) 1998-2003 Massachusetts Institute of Technology. 
 * This code was developed as part of the Haystack research project 
 * (http://haystack.lcs.mit.edu/). Permission is hereby granted, 
 * free of charge, to any person obtaining a copy of this software 
 * and associated documentation files (the "Software"), to deal in 
 * the Software without restriction, including without limitation 
 * the rights to use, copy, modify, merge, publish, distribute, 
 * sublicense, and/or sell copies of the Software, and to permit 
 * persons to whom the Software is furnished to do so, subject to 
 * the following conditions: 
 * 
 * The above copyright notice and this permission notice shall be 
 * included in all copies or substantial portions of the Software. 
 * 
 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, 
 * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES 
 * OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND 
 * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT 
 * HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, 
 * WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING 
 * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR 
 * OTHER DEALINGS IN THE SOFTWARE. 
 */

#ifndef COMMON_HASHMAP_H
#define COMMON_HASHMAP_H

#include "common/stdafx.h"
#include "common/func.h"
#include "common/str.h"
#include "common/util.h"

namespace Common { 

// data structure used by HashMap internally to keep
// track of what's mapped to what.
template <class Key, class Val>
struct BaseNode {
	Key _key;
	Val _value;
	BaseNode() {}
	BaseNode(const Key &key) : _key(key) {}
};

// The table sizes ideally are primes. We use a helper function to find
// suitable table sizes.
uint nextTableSize(uint x);


// Enable the following #define if you want to check how many collisions the
// code produces (many collisions indicate either a bad hash function, or a
// hash table that is too small).
//#define DEBUG_HASH_COLLISIONS

/**
 * HashMap<Key,Val> maps objects of type Key to objects of type Val.
 * For each used Key type, we need an "uint hashit(Key,uint)" function
 * that computes a hash for the given Key object and returns it as an
 * an integer from 0 to hashsize-1, and also an "equality functor".
 * that returns true if if its two arguments are to be considered
 * equal. Also, we assume that "=" works on Val objects for assignment.
 *
 * If aa is an HashMap<Key,Val>, then space is allocated each time aa[key] is
 * referenced, for a new key. If the object is const, then an assertion is
 * triggered instead. Hence if you are not sure whether a key is contained in
 * the map, use contains() first to check for its presence.
 */ 
template <class Key, class Val, class HashFunc = Hash<Key>, class EqualFunc = EqualTo<Key>, class BaseNodeType = BaseNode<Key, Val> >
class HashMap {
private:
#if defined (_WIN32_WCE) || defined (_MSC_VER) || defined (__SYMBIAN32__) || defined (PALMOS_MODE) || defined (__MINT__)
//FIXME evc4, msvc6,msvc7 & GCC 2.9x doesn't like it as private member
public:
#endif
	BaseNodeType **_arr;	// hashtable of size arrsize.
	uint _arrsize, _nele;
	
	HashFunc _hash;
	EqualFunc _equal;
	
#ifdef DEBUG_HASH_COLLISIONS
	mutable int _collisions, _lookups;
#endif

	int lookup(const Key &key) const;
	void expand_array(uint newsize);

public:
	class const_iterator {
		typedef const HashMap<Key, Val, HashFunc, EqualFunc, BaseNodeType> * hashmap_t;
		friend class HashMap<Key, Val, HashFunc, EqualFunc, BaseNodeType>;
	protected:
		hashmap_t _hashmap;
		uint _idx;
		const_iterator(uint idx, hashmap_t hashmap) : _idx(idx), _hashmap(hashmap) {}

		const BaseNodeType *deref() const {
			assert(_hashmap != 0);
			BaseNodeType *node = _hashmap->_arr[_idx];
			assert(node != 0);
			return node;
		}

	public:
		const_iterator() : _idx(0), _hashmap(0) {}

		const BaseNodeType &operator *() const { return *deref(); }
		const BaseNodeType *operator->() const { return deref(); }
		bool operator ==(const const_iterator &iter) const { return _idx == iter._idx && _hashmap == iter._hashmap; }
		bool operator !=(const const_iterator &iter) const { return !(*this == iter); }
		const_iterator operator ++() {
			assert(_hashmap);
			do {
				_idx++;
			} while (_idx < _hashmap->_arrsize && _hashmap->_arr[_idx] == 0);
			if (_idx >= _hashmap->_arrsize)
				_idx = (uint)-1;
			
			return *this;
		}
	};

	HashMap();
	~HashMap();

	bool contains(const Key &key) const;

	Val &operator [](const Key &key);
	const Val &operator [](const Key &key) const;
	const Val &queryVal(const Key &key) const;

	void clear(bool shrinkArray = 0);

	size_t erase(const Key &key);

	uint size() const { return _nele; }

	const_iterator	begin() const {
		// Find and return the first non-empty entry
		for (uint ctr = 0; ctr < _arrsize; ++ctr) {
			if (_arr[ctr])
				return const_iterator(ctr, this);
		}
		return end();
	}
	const_iterator	end() const {
		return const_iterator((uint)-1, this);
	}
	
	const_iterator	find(const Key &key) const {
		uint ctr = lookup(key);
		if (_arr[ctr])
			return const_iterator(ctr, this);
		return end();
	}
	
	// TODO: insert() method?

	bool empty() const {
		return (_nele == 0);
	}
};

//-------------------------------------------------------
// HashMap functions

template <class Key, class Val, class HashFunc, class EqualFunc, class BaseNodeType>
HashMap<Key, Val, HashFunc, EqualFunc, BaseNodeType>::HashMap() {
	_arrsize = nextTableSize(0);
	_arr = new BaseNodeType *[_arrsize];
	assert(_arr != NULL);
	memset(_arr, 0, _arrsize * sizeof(BaseNodeType *));

	_nele = 0;
	
#ifdef DEBUG_HASH_COLLISIONS
	_collisions = 0;
	_lookups = 0;
#endif
}

template <class Key, class Val, class HashFunc, class EqualFunc, class BaseNodeType>
HashMap<Key, Val, HashFunc, EqualFunc, BaseNodeType>::~HashMap() {
	uint ctr;

	for (ctr = 0; ctr < _arrsize; ctr++)
		if (_arr[ctr] != NULL)
			delete _arr[ctr];

	delete[] _arr;
}

template <class Key, class Val, class HashFunc, class EqualFunc, class BaseNodeType>
void HashMap<Key, Val, HashFunc, EqualFunc, BaseNodeType>::clear(bool shrinkArray) {
	for (uint ctr = 0; ctr < _arrsize; ctr++) {
		if (_arr[ctr] != NULL) {
			delete _arr[ctr];
			_arr[ctr] = NULL;
		}
	}

	if (shrinkArray && _arrsize > nextTableSize(0)) {
		delete[] _arr;

		_arrsize = nextTableSize(0);
		_arr = new BaseNodeType *[_arrsize];
		assert(_arr != NULL);
		memset(_arr, 0, _arrsize * sizeof(BaseNodeType *));
	}

	_nele = 0;
}

template <class Key, class Val, class HashFunc, class EqualFunc, class BaseNodeType>
void HashMap<Key, Val, HashFunc, EqualFunc, BaseNodeType>::expand_array(uint newsize) {
	assert(newsize > _arrsize);
	BaseNodeType **old_arr;
	uint old_arrsize, old_nele, ctr, dex;

	old_nele = _nele;
	old_arr = _arr;
	old_arrsize = _arrsize;

	// allocate a new array 
	_arrsize = newsize;
	_arr = new BaseNodeType *[_arrsize];
	assert(_arr != NULL);
	memset(_arr, 0, _arrsize * sizeof(BaseNodeType *));

	_nele = 0;

	// rehash all the old elements
	for (ctr = 0; ctr < old_arrsize; ctr++) {
		if (old_arr[ctr] == NULL)
			continue;

		// Insert the element from the old table into the new table.
		// Since we know that no key exists twice the old table, we
		// can do this slightly better than by calling lookup, since we
		// don't have to call _equal().
		dex = _hash(old_arr[ctr]->_key) % _arrsize;
		while (_arr[dex] != NULL) {
			dex = (dex + 1) % _arrsize;
		}

		_arr[dex] = old_arr[ctr];
		_nele++;
	}

	// Perform a sanity check: Old number of elements should match the new one!
	assert(_nele == old_nele);

	delete[] old_arr;

	return;
}

template <class Key, class Val, class HashFunc, class EqualFunc, class BaseNodeType>
int HashMap<Key, Val, HashFunc, EqualFunc, BaseNodeType>::lookup(const Key &key) const {
	uint ctr = _hash(key) % _arrsize;

	while (_arr[ctr] != NULL && !_equal(_arr[ctr]->_key, key)) {
		ctr = (ctr + 1) % _arrsize;

#ifdef DEBUG_HASH_COLLISIONS
		_collisions++;
#endif
	}
	
#ifdef DEBUG_HASH_COLLISIONS
	_lookups++;
	fprintf(stderr, "collisions %d, lookups %d, ratio %f in HashMap %p; size %d num elements %d\n",
		_collisions, _lookups, ((double) _collisions / (double)_lookups),
		(const void *)this, _arrsize, _nele);
#endif

	return ctr;
}

template <class Key, class Val, class HashFunc, class EqualFunc, class BaseNodeType>
bool HashMap<Key, Val, HashFunc, EqualFunc, BaseNodeType>::contains(const Key &key) const {
	uint ctr = lookup(key);
	return (_arr[ctr] != NULL);
}

template <class Key, class Val, class HashFunc, class EqualFunc, class BaseNodeType>
Val &HashMap<Key, Val, HashFunc, EqualFunc, BaseNodeType>::operator [](const Key &key) {
	uint ctr = lookup(key);

	if (_arr[ctr] == NULL) {
		_arr[ctr] = new BaseNodeType(key);
		_nele++;

		// Keep the load factor below 75%.
		if (_nele > _arrsize * 75 / 100) {
			expand_array(nextTableSize(_arrsize));
			ctr = lookup(key);
		}
	}

	return _arr[ctr]->_value;
}

template <class Key, class Val, class HashFunc, class EqualFunc, class BaseNodeType>
const Val &HashMap<Key, Val, HashFunc, EqualFunc, BaseNodeType>::operator [](const Key &key) const {
	return queryVal(key);
}

template <class Key, class Val, class HashFunc, class EqualFunc, class BaseNodeType>
const Val &HashMap<Key, Val, HashFunc, EqualFunc, BaseNodeType>::queryVal(const Key &key) const {
	uint ctr = lookup(key);
	assert(_arr[ctr] != NULL);
	return _arr[ctr]->_value;
}

template <class Key, class Val, class HashFunc, class EqualFunc, class BaseNodeType>
size_t HashMap<Key, Val, HashFunc, EqualFunc, BaseNodeType>::erase(const Key &key) {
	// This is based on code in the Wikipedia article on Hash tables.
	uint i = lookup(key);
	if (_arr[i] == NULL)
		return 0; // key wasn't present, so no work has to be done
	uint j = i;
	while (true) {
		j = (j + 1) % _arrsize;
		if (_arr[j] == NULL)
			break;
		uint k = _hash(_arr[j]->_key) % _arrsize;
		if ((j > i && (k <= i || k > j)) ||
		    (j < i && (k <= i && k > j)) ) {
			_arr[i] = _arr[j];
			i = j;
		}
	}
	_arr[i] = NULL;
	return 1;
}

}	// End of namespace Common

#endif