summaryrefslogtreecommitdiff
path: root/test/compression/lzhash.c
blob: 83cbe94df91fc2fa7f786501bbf453d7dad2e9db (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
#include <stdio.h>
/***************************************************************************
*          Lempel, Ziv, Storer, and Szymanski Encoding and Decoding
*
*   File    : hash.c
*   Purpose : Implement hash table optimized matching of uncoded strings
*             for LZSS algorithm.
*   Author  : Michael Dipperstein
*   Date    : February 21, 2004
*
****************************************************************************
*   UPDATES
*   $Id: hash.c,v 1.7 2007/09/20 04:34:25 michael Exp $
*   $Log: hash.c,v $
*   Revision 1.7  2007/09/20 04:34:25  michael
*   Changes required for LGPL v3.
*
*   Revision 1.6  2006/12/26 04:09:09  michael
*   Updated e-mail address and minor text clean-up.
*
*   Revision 1.5  2005/12/29 14:37:56  michael
*   Remove debug statements.
*
*   Revision 1.4  2005/12/29 14:32:56  michael
*   Correct algorithm for replacing characters in dictionary.
*
*   Revision 1.3  2005/12/29 07:06:24  michael
*   Let the hash table size vary with the size of the sliding window.
*
*   Revision 1.2  2005/12/28 06:03:30  michael
*   Use slower but clearer Get/PutBitsInt for reading/writing bits.
*   Replace mod with conditional Wrap macro.
*
*   Revision 1.1  2004/02/22 17:23:02  michael
*   Initial revision of hash table search.  Mostly code from lzhash.c.
*
*
****************************************************************************
*
* Hash: Hash table optimized matching routines used by LZSS
*       Encoding/Decoding Routine
* Copyright (C) 2004-2007 by
* Michael Dipperstein (mdipper@alumni.engr.ucsb.edu)
*
* This file is part of the lzss library.
*
* The lzss library is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public License as
* published by the Free Software Foundation; either version 3 of the
* License, or (at your option) any later version.
*
* The lzss library 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 Lesser
* General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public License
* along with this program.  If not, see <http://www.gnu.org/licenses/>.
*
***************************************************************************/

/***************************************************************************
*                             INCLUDED FILES
***************************************************************************/
#include "lzlocal.h"

/***************************************************************************
*                            TYPE DEFINITIONS
***************************************************************************/

/***************************************************************************
*                                CONSTANTS
***************************************************************************/
#define NULL_INDEX      (WINDOW_SIZE + 1)

#define HASH_SIZE       (WINDOW_SIZE >> 2)  /* size of hash table */

/***************************************************************************
*                            GLOBAL VARIABLES
***************************************************************************/
/* cyclic buffer sliding window of already read characters */
extern unsigned char slidingWindow[];
extern unsigned char uncodedLookahead[];

unsigned int hashTable[HASH_SIZE];          /* list head for each hash key */
unsigned int next[WINDOW_SIZE];             /* indices of next in hash list */

/***************************************************************************
*                               PROTOTYPES
***************************************************************************/

/***************************************************************************
*                                FUNCTIONS
***************************************************************************/

/****************************************************************************
*   Function   : HashKey
*   Description: This function generates a hash key for a (MAX_UNCODED + 1)
*                long string located either in the sliding window or the
*                uncoded lookahead.  The key generation algorithm is
*                supposed to be based on the algorithm used by gzip.  As
*                reported in K. Sadakane, H. Imai. "Improving the Speed of
*                LZ77 Compression by Hashing and Suffix Sorting". IEICE
*                Trans. Fundamentals, Vol. E83-A, No. 12 (December 2000)
*   Parameters : offset - offset into either the uncoded lookahead or the
*                         sliding window.
*                lookahead - TRUE iff offset is an offset into the uncoded
*                            lookahead buffer.
*   Effects    : NONE
*   Returned   : The sliding window index where the match starts and the
*                length of the match.  If there is no match a length of
*                zero will be returned.
****************************************************************************/
unsigned int HashKey(unsigned int offset, unsigned char lookahead)
{
    int i, hashKey;

    hashKey = 0;

    if (lookahead)
    {
        /* string is in the lookahead buffer */
        for (i = 0; i < (MAX_UNCODED + 1); i++)
        {
            hashKey = (hashKey << 5) ^ (uncodedLookahead[offset]);
            hashKey %= HASH_SIZE;
            offset = Wrap((offset + 1), MAX_CODED);
        }
    }
    else
    {
        /* string is in the sliding window */
        for (i = 0; i < (MAX_UNCODED + 1); i++)
        {
            hashKey = (hashKey << 5) ^ (slidingWindow[offset]);
            hashKey %= HASH_SIZE;
            offset = Wrap((offset + 1), WINDOW_SIZE);
        }
    }

    return hashKey;
}

/****************************************************************************
*   Function   : InitializeSearchStructures
*   Description: This function initializes structures used to speed up the
*                process of mathcing uncoded strings to strings in the
*                sliding window.  For hashed searches, this means that a
*                hash table pointing to linked lists is initialized.
*   Parameters : None
*   Effects    : The hash table and next array are initialized.
*   Returned   : None
*
*   NOTE: This function assumes that the sliding window is initially filled
*         with all identical characters.
****************************************************************************/
void InitializeSearchStructures()
{
    unsigned int i;

    /************************************************************************
    * Since the encode routine only fills the sliding window with one
    * character, there is only one hash key for the entier sliding window.
    * That means all positions are in the same linked list.
    ************************************************************************/
    for (i = 0; i < (WINDOW_SIZE - 1); i++)
    {
        next[i] = i + 1;
    }

    /* there's no next for the last character */
    next[WINDOW_SIZE - 1] = NULL_INDEX;

    /* the only list right now is the "   " list */
    for (i = 0; i < HASH_SIZE; i++)
    {
        hashTable[i] = NULL_INDEX;
    }

    hashTable[HashKey(0, FALSE)] = 0;
}

/****************************************************************************
*   Function   : FindMatch
*   Description: This function will search through the slidingWindow
*                dictionary for the longest sequence matching the MAX_CODED
*                long string stored in uncodedLookahead.
*   Parameters : uncodedHead - head of uncoded lookahead buffer
*   Effects    : NONE
*   Returned   : The sliding window index where the match starts and the
*                length of the match.  If there is no match a length of
*                zero will be returned.
****************************************************************************/
/* XL: no struct return */
void FindMatch(encoded_string_t * matchData, 
               unsigned int windowHead, unsigned int uncodedHead)
{
    unsigned int i, j;

    matchData->length = 0;
    matchData->offset = 0;

    i = hashTable[HashKey(uncodedHead, TRUE)];  /* start of proper list */
    j = 0;

    while (i != NULL_INDEX)
    {
        if (slidingWindow[i] == uncodedLookahead[uncodedHead])
        {
            /* we matched one how many more match? */
            j = 1;

            while(slidingWindow[Wrap((i + j), WINDOW_SIZE)] ==
                uncodedLookahead[Wrap((uncodedHead + j), MAX_CODED)])
            {
                if (j >= MAX_CODED)
                {
                    break;
                }
                j++;
            }

            if (j > matchData->length)
            {
                matchData->length = j;
                matchData->offset = i;
            }
        }

        if (j >= MAX_CODED)
        {
            matchData->length = MAX_CODED;
            break;
        }

        i = next[i];    /* try next in list */
    }
}

/****************************************************************************
*   Function   : AddString
*   Description: This function adds the (MAX_UNCODED + 1) long string
*                starting at slidingWindow[charIndex] to the hash table's
*                linked list associated with its hash key.
*   Parameters : charIndex - sliding window index of the string to be
*                            added to the linked list.
*   Effects    : The string starting at slidingWindow[charIndex] is appended
*                to the end of the appropriate linked list.
*   Returned   : NONE
****************************************************************************/
void AddString(unsigned int charIndex)
{
    unsigned int i, hashKey;

    /* inserted character will be at the end of the list */
    next[charIndex] = NULL_INDEX;

    hashKey = HashKey(charIndex, FALSE);

    if (hashTable[hashKey] == NULL_INDEX)
    {
        /* this is the only character in it's list */
        hashTable[hashKey] = charIndex;
        return;
    }

    /* find the end of the list */
    i = hashTable[hashKey];

    while(next[i] != NULL_INDEX)
    {
        i = next[i];
    }

    /* add new character to the list end */
    next[i] = charIndex;
}

/****************************************************************************
*   Function   : RemoveString
*   Description: This function removes the (MAX_UNCODED + 1) long string
*                starting at slidingWindow[charIndex] from the hash table's
*                linked list associated with its hash key.
*   Parameters : charIndex - sliding window index of the string to be
*                            removed from the linked list.
*   Effects    : The string starting at slidingWindow[charIndex] is removed
*                from its linked list.
*   Returned   : NONE
****************************************************************************/
void RemoveString(unsigned int charIndex)
{
    unsigned int i, hashKey;
    unsigned int nextIndex;

    nextIndex = next[charIndex];        /* remember where this points to */
    next[charIndex] = NULL_INDEX;

    hashKey = HashKey(charIndex, FALSE);

    if (hashTable[hashKey] == charIndex)
    {
        /* we're deleting a list head */
        hashTable[hashKey] = nextIndex;
        return;
    }

    /* find character pointing to ours */
    i = hashTable[hashKey];

    while(next[i] != charIndex)
    {
        i = next[i];
    }

    /* point the previous next */
    next[i] = nextIndex;
}

/****************************************************************************
*   Function   : ReplaceChar
*   Description: This function replaces the character stored in
*                slidingWindow[charIndex] with the one specified by
*                replacement.  The hash table entries effected by the
*                replacement are also corrected.
*   Parameters : charIndex - sliding window index of the character to be
*                            removed from the linked list.
*   Effects    : slidingWindow[charIndex] is replaced by replacement.  Old
*                hash entries for strings containing slidingWindow[charIndex]
*                are removed and new ones are added.
*   Returned   : NONE
****************************************************************************/
void ReplaceChar(unsigned int charIndex, unsigned char replacement)
{
    unsigned int firstIndex, i;

    if (charIndex < MAX_UNCODED)
    {
        firstIndex = (WINDOW_SIZE + charIndex) - MAX_UNCODED;
    }
    else
    {
        firstIndex = charIndex - MAX_UNCODED;
    }

    /* remove all hash entries containing character at char index */
    for (i = 0; i < (MAX_UNCODED + 1); i++)
    {
        RemoveString(Wrap((firstIndex + i), WINDOW_SIZE));
    }

    slidingWindow[charIndex] = replacement;

    /* add all hash entries containing character at char index */
    for (i = 0; i < (MAX_UNCODED + 1); i++)
    {
        AddString(Wrap((firstIndex + i), WINDOW_SIZE));
    }
}