summaryrefslogtreecommitdiffstats
path: root/lib/lz4hc.c
blob: a1d1a55fdced3c3e0127f0159d5a0ba243a6fa37 (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
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
/*
    LZ4 HC - High Compression Mode of LZ4
    Copyright (C) 2011-2015, Yann Collet.

    BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php)

    Redistribution and use in source and binary forms, with or without
    modification, are permitted provided that the following conditions are
    met:

    * Redistributions of source code must retain the above copyright
    notice, this list of conditions and the following disclaimer.
    * Redistributions in binary form must reproduce the above
    copyright notice, this list of conditions and the following disclaimer
    in the documentation and/or other materials provided with the
    distribution.

    THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
    "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
    LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
    A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
    OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
    SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
    LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
    DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
    THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
    (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
    OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

    You can contact the author at :
       - LZ4 source repository : https://github.com/lz4/lz4
       - LZ4 public forum : https://groups.google.com/forum/#!forum/lz4c
*/



/* *************************************
*  Tuning Parameter
***************************************/

/*!
 * HEAPMODE :
 * Select how default compression function will allocate workplace memory,
 * in stack (0:fastest), or in heap (1:requires malloc()).
 * Since workplace is rather large, heap mode is recommended.
 */
#define LZ4HC_HEAPMODE 0


/* *************************************
*  Includes
***************************************/
#include "lz4hc.h"


/* *************************************
*  Local Compiler Options
***************************************/
#if defined(__GNUC__)
#  pragma GCC diagnostic ignored "-Wunused-function"
#endif

#if defined (__clang__)
#  pragma clang diagnostic ignored "-Wunused-function"
#endif


/* *************************************
*  Common LZ4 definition
***************************************/
#define LZ4_COMMONDEFS_ONLY
#include "lz4.c"


/* *************************************
*  Local Constants
***************************************/
#define DICTIONARY_LOGSIZE 16
#define MAXD (1<<DICTIONARY_LOGSIZE)
#define MAXD_MASK (MAXD - 1)

#define HASH_LOG (DICTIONARY_LOGSIZE-1)
#define HASHTABLESIZE (1 << HASH_LOG)
#define HASH_MASK (HASHTABLESIZE - 1)

#define OPTIMAL_ML (int)((ML_MASK-1)+MINMATCH)


/**************************************
*  Local Types
**************************************/
typedef struct
{
    U32   hashTable[HASHTABLESIZE];
    U16   chainTable[MAXD];
    const BYTE* end;        /* next block here to continue on current prefix */
    const BYTE* base;       /* All index relative to this position */
    const BYTE* dictBase;   /* alternate base for extDict */
    BYTE* inputBuffer;      /* deprecated */
    U32   dictLimit;        /* below that point, need extDict */
    U32   lowLimit;         /* below that point, no more dict */
    U32   nextToUpdate;     /* index from which to continue dictionary update */
    U32   compressionLevel;
} LZ4HC_Data_Structure;


/**************************************
*  Local Macros
**************************************/
#define HASH_FUNCTION(i)       (((i) * 2654435761U) >> ((MINMATCH*8)-HASH_LOG))
//#define DELTANEXTU16(p)        chainTable[(p) & MAXD_MASK]   /* flexible, MAXD dependent */
#define DELTANEXTU16(p)        chainTable[(U16)(p)]   /* faster */

static U32 LZ4HC_hashPtr(const void* ptr) { return HASH_FUNCTION(LZ4_read32(ptr)); }



/**************************************
*  HC Compression
**************************************/
static void LZ4HC_init (LZ4HC_Data_Structure* hc4, const BYTE* start)
{
    MEM_INIT((void*)hc4->hashTable, 0, sizeof(hc4->hashTable));
    MEM_INIT(hc4->chainTable, 0xFF, sizeof(hc4->chainTable));
    hc4->nextToUpdate = 64 KB;
    hc4->base = start - 64 KB;
    hc4->end = start;
    hc4->dictBase = start - 64 KB;
    hc4->dictLimit = 64 KB;
    hc4->lowLimit = 64 KB;
}


/* Update chains up to ip (excluded) */
FORCE_INLINE void LZ4HC_Insert (LZ4HC_Data_Structure* hc4, const BYTE* ip)
{
    U16* const chainTable = hc4->chainTable;
    U32* const hashTable  = hc4->hashTable;
    const BYTE* const base = hc4->base;
    U32 const target = (U32)(ip - base);
    U32 idx = hc4->nextToUpdate;

    while (idx < target) {
        U32 const h = LZ4HC_hashPtr(base+idx);
        size_t delta = idx - hashTable[h];
        if (delta>MAX_DISTANCE) delta = MAX_DISTANCE;
        DELTANEXTU16(idx) = (U16)delta;
        hashTable[h] = idx;
        idx++;
    }

    hc4->nextToUpdate = target;
}


FORCE_INLINE int LZ4HC_InsertAndFindBestMatch (LZ4HC_Data_Structure* hc4,   /* Index table will be updated */
                                               const BYTE* ip, const BYTE* const iLimit,
                                               const BYTE** matchpos,
                                               const int maxNbAttempts)
{
    U16* const chainTable = hc4->chainTable;
    U32* const HashTable = hc4->hashTable;
    const BYTE* const base = hc4->base;
    const BYTE* const dictBase = hc4->dictBase;
    const U32 dictLimit = hc4->dictLimit;
    const U32 lowLimit = (hc4->lowLimit + 64 KB > (U32)(ip-base)) ? hc4->lowLimit : (U32)(ip - base) - (64 KB - 1);
    U32 matchIndex;
    const BYTE* match;
    int nbAttempts=maxNbAttempts;
    size_t ml=0;

    /* HC4 match finder */
    LZ4HC_Insert(hc4, ip);
    matchIndex = HashTable[LZ4HC_hashPtr(ip)];

    while ((matchIndex>=lowLimit) && (nbAttempts)) {
        nbAttempts--;
        if (matchIndex >= dictLimit) {
            match = base + matchIndex;
            if (*(match+ml) == *(ip+ml)
                && (LZ4_read32(match) == LZ4_read32(ip)))
            {
                size_t const mlt = LZ4_count(ip+MINMATCH, match+MINMATCH, iLimit) + MINMATCH;
                if (mlt > ml) { ml = mlt; *matchpos = match; }
            }
        } else {
            match = dictBase + matchIndex;
            if (LZ4_read32(match) == LZ4_read32(ip)) {
                size_t mlt;
                const BYTE* vLimit = ip + (dictLimit - matchIndex);
                if (vLimit > iLimit) vLimit = iLimit;
                mlt = LZ4_count(ip+MINMATCH, match+MINMATCH, vLimit) + MINMATCH;
                if ((ip+mlt == vLimit) && (vLimit < iLimit))
                    mlt += LZ4_count(ip+mlt, base+dictLimit, iLimit);
                if (mlt > ml) { ml = mlt; *matchpos = base + matchIndex; }   /* virtual matchpos */
            }
        }
        matchIndex -= DELTANEXTU16(matchIndex);
    }

    return (int)ml;
}


FORCE_INLINE int LZ4HC_InsertAndGetWiderMatch (
    LZ4HC_Data_Structure* hc4,
    const BYTE* const ip,
    const BYTE* const iLowLimit,
    const BYTE* const iHighLimit,
    int longest,
    const BYTE** matchpos,
    const BYTE** startpos,
    const int maxNbAttempts)
{
    U16* const chainTable = hc4->chainTable;
    U32* const HashTable = hc4->hashTable;
    const BYTE* const base = hc4->base;
    const U32 dictLimit = hc4->dictLimit;
    const BYTE* const lowPrefixPtr = base + dictLimit;
    const U32 lowLimit = (hc4->lowLimit + 64 KB > (U32)(ip-base)) ? hc4->lowLimit : (U32)(ip - base) - (64 KB - 1);
    const BYTE* const dictBase = hc4->dictBase;
    U32   matchIndex;
    int nbAttempts = maxNbAttempts;
    int delta = (int)(ip-iLowLimit);


    /* First Match */
    LZ4HC_Insert(hc4, ip);
    matchIndex = HashTable[LZ4HC_hashPtr(ip)];

    while ((matchIndex>=lowLimit) && (nbAttempts)) {
        nbAttempts--;
        if (matchIndex >= dictLimit) {
            const BYTE* matchPtr = base + matchIndex;
            if (*(iLowLimit + longest) == *(matchPtr - delta + longest)) {
                if (LZ4_read32(matchPtr) == LZ4_read32(ip)) {
                    int mlt = MINMATCH + LZ4_count(ip+MINMATCH, matchPtr+MINMATCH, iHighLimit);
                    int back = 0;

                    while ((ip+back > iLowLimit)
                           && (matchPtr+back > lowPrefixPtr)
                           && (ip[back-1] == matchPtr[back-1]))
                            back--;

                    mlt -= back;

                    if (mlt > longest) {
                        longest = (int)mlt;
                        *matchpos = matchPtr+back;
                        *startpos = ip+back;
                    }
                }
            }
        } else {
            const BYTE* matchPtr = dictBase + matchIndex;
            if (LZ4_read32(matchPtr) == LZ4_read32(ip)) {
                size_t mlt;
                int back=0;
                const BYTE* vLimit = ip + (dictLimit - matchIndex);
                if (vLimit > iHighLimit) vLimit = iHighLimit;
                mlt = LZ4_count(ip+MINMATCH, matchPtr+MINMATCH, vLimit) + MINMATCH;
                if ((ip+mlt == vLimit) && (vLimit < iHighLimit))
                    mlt += LZ4_count(ip+mlt, base+dictLimit, iHighLimit);
                while ((ip+back > iLowLimit) && (matchIndex+back > lowLimit) && (ip[back-1] == matchPtr[back-1])) back--;
                mlt -= back;
                if ((int)mlt > longest) { longest = (int)mlt; *matchpos = base + matchIndex + back; *startpos = ip+back; }
            }
        }
        matchIndex -= DELTANEXTU16(matchIndex);
    }

    return longest;
}


typedef enum { noLimit = 0, limitedOutput = 1 } limitedOutput_directive;

#define LZ4HC_DEBUG 0
#if LZ4HC_DEBUG
static unsigned debug = 0;
#endif

FORCE_INLINE int LZ4HC_encodeSequence (
    const BYTE** ip,
    BYTE** op,
    const BYTE** anchor,
    int matchLength,
    const BYTE* const match,
    limitedOutput_directive limitedOutputBuffer,
    BYTE* oend)
{
    int length;
    BYTE* token;

#if LZ4HC_DEBUG
    if (debug) printf("literal : %u  --  match : %u  --  offset : %u\n", (U32)(*ip - *anchor), (U32)matchLength, (U32)(*ip-match));
#endif

    /* Encode Literal length */
    length = (int)(*ip - *anchor);
    token = (*op)++;
    if ((limitedOutputBuffer) && ((*op + (length>>8) + length + (2 + 1 + LASTLITERALS)) > oend)) return 1;   /* Check output limit */
    if (length>=(int)RUN_MASK) { int len; *token=(RUN_MASK<<ML_BITS); len = length-RUN_MASK; for(; len > 254 ; len-=255) *(*op)++ = 255;  *(*op)++ = (BYTE)len; }
    else *token = (BYTE)(length<<ML_BITS);

    /* Copy Literals */
    LZ4_wildCopy(*op, *anchor, (*op) + length);
    *op += length;

    /* Encode Offset */
    LZ4_writeLE16(*op, (U16)(*ip-match)); *op += 2;

    /* Encode MatchLength */
    length = (int)(matchLength-MINMATCH);
    if ((limitedOutputBuffer) && (*op + (length>>8) + (1 + LASTLITERALS) > oend)) return 1;   /* Check output limit */
    if (length>=(int)ML_MASK) {
        *token += ML_MASK;
        length -= ML_MASK;
        for(; length > 509 ; length-=510) { *(*op)++ = 255; *(*op)++ = 255; }
        if (length > 254) { length-=255; *(*op)++ = 255; }
        *(*op)++ = (BYTE)length;
    } else {
        *token += (BYTE)(length);
    }

    /* Prepare next loop */
    *ip += matchLength;
    *anchor = *ip;

    return 0;
}


static int LZ4HC_compress_generic (
    void* ctxvoid,
    const char* source,
    char* dest,
    int inputSize,
    int maxOutputSize,
    int compressionLevel,
    limitedOutput_directive limit
    )
{
    LZ4HC_Data_Structure* ctx = (LZ4HC_Data_Structure*) ctxvoid;
    const BYTE* ip = (const BYTE*) source;
    const BYTE* anchor = ip;
    const BYTE* const iend = ip + inputSize;
    const BYTE* const mflimit = iend - MFLIMIT;
    const BYTE* const matchlimit = (iend - LASTLITERALS);

    BYTE* op = (BYTE*) dest;
    BYTE* const oend = op + maxOutputSize;

    unsigned maxNbAttempts;
    int   ml, ml2, ml3, ml0;
    const BYTE* ref = NULL;
    const BYTE* start2 = NULL;
    const BYTE* ref2 = NULL;
    const BYTE* start3 = NULL;
    const BYTE* ref3 = NULL;
    const BYTE* start0;
    const BYTE* ref0;

    /* init */
    if (compressionLevel > LZ4HC_MAX_CLEVEL) compressionLevel = LZ4HC_MAX_CLEVEL;
    if (compressionLevel < 1) compressionLevel = LZ4HC_DEFAULT_CLEVEL;
    maxNbAttempts = 1 << (compressionLevel-1);
    ctx->end += inputSize;

    ip++;

    /* Main Loop */
    while (ip < mflimit) {
        ml = LZ4HC_InsertAndFindBestMatch (ctx, ip, matchlimit, (&ref), maxNbAttempts);
        if (!ml) { ip++; continue; }

        /* saved, in case we would skip too much */
        start0 = ip;
        ref0 = ref;
        ml0 = ml;

_Search2:
        if (ip+ml < mflimit)
            ml2 = LZ4HC_InsertAndGetWiderMatch(ctx, ip + ml - 2, ip + 1, matchlimit, ml, &ref2, &start2, maxNbAttempts);
        else ml2 = ml;

        if (ml2 == ml) { /* No better match */
            if (LZ4HC_encodeSequence(&ip, &op, &anchor, ml, ref, limit, oend)) return 0;
            continue;
        }

        if (start0 < ip) {
            if (start2 < ip + ml0) {  /* empirical */
                ip = start0;
                ref = ref0;
                ml = ml0;
            }
        }

        /* Here, start0==ip */
        if ((start2 - ip) < 3) {  /* First Match too small : removed */
            ml = ml2;
            ip = start2;
            ref =ref2;
            goto _Search2;
        }

_Search3:
        /*
        * Currently we have :
        * ml2 > ml1, and
        * ip1+3 <= ip2 (usually < ip1+ml1)
        */
        if ((start2 - ip) < OPTIMAL_ML) {
            int correction;
            int new_ml = ml;
            if (new_ml > OPTIMAL_ML) new_ml = OPTIMAL_ML;
            if (ip+new_ml > start2 + ml2 - MINMATCH) new_ml = (int)(start2 - ip) + ml2 - MINMATCH;
            correction = new_ml - (int)(start2 - ip);
            if (correction > 0) {
                start2 += correction;
                ref2 += correction;
                ml2 -= correction;
            }
        }
        /* Now, we have start2 = ip+new_ml, with new_ml = min(ml, OPTIMAL_ML=18) */

        if (start2 + ml2 < mflimit)
            ml3 = LZ4HC_InsertAndGetWiderMatch(ctx, start2 + ml2 - 3, start2, matchlimit, ml2, &ref3, &start3, maxNbAttempts);
        else ml3 = ml2;

        if (ml3 == ml2) {  /* No better match : 2 sequences to encode */
            /* ip & ref are known; Now for ml */
            if (start2 < ip+ml)  ml = (int)(start2 - ip);
            /* Now, encode 2 sequences */
            if (LZ4HC_encodeSequence(&ip, &op, &anchor, ml, ref, limit, oend)) return 0;
            ip = start2;
            if (LZ4HC_encodeSequence(&ip, &op, &anchor, ml2, ref2, limit, oend)) return 0;
            continue;
        }

        if (start3 < ip+ml+3) {  /* Not enough space for match 2 : remove it */
            if (start3 >= (ip+ml)) {  /* can write Seq1 immediately ==> Seq2 is removed, so Seq3 becomes Seq1 */
                if (start2 < ip+ml) {
                    int correction = (int)(ip+ml - start2);
                    start2 += correction;
                    ref2 += correction;
                    ml2 -= correction;
                    if (ml2 < MINMATCH) {
                        start2 = start3;
                        ref2 = ref3;
                        ml2 = ml3;
                    }
                }

                if (LZ4HC_encodeSequence(&ip, &op, &anchor, ml, ref, limit, oend)) return 0;
                ip  = start3;
                ref = ref3;
                ml  = ml3;

                start0 = start2;
                ref0 = ref2;
                ml0 = ml2;
                goto _Search2;
            }

            start2 = start3;
            ref2 = ref3;
            ml2 = ml3;
            goto _Search3;
        }

        /*
        * OK, now we have 3 ascending matches; let's write at least the first one
        * ip & ref are known; Now for ml
        */
        if (start2 < ip+ml) {
            if ((start2 - ip) < (int)ML_MASK) {
                int correction;
                if (ml > OPTIMAL_ML) ml = OPTIMAL_ML;
                if (ip + ml > start2 + ml2 - MINMATCH) ml = (int)(start2 - ip) + ml2 - MINMATCH;
                correction = ml - (int)(start2 - ip);
                if (correction > 0) {
                    start2 += correction;
                    ref2 += correction;
                    ml2 -= correction;
                }
            } else {
                ml = (int)(start2 - ip);
            }
        }
        if (LZ4HC_encodeSequence(&ip, &op, &anchor, ml, ref, limit, oend)) return 0;

        ip = start2;
        ref = ref2;
        ml = ml2;

        start2 = start3;
        ref2 = ref3;
        ml2 = ml3;

        goto _Search3;
    }

    /* Encode Last Literals */
    {   int lastRun = (int)(iend - anchor);
        if ((limit) && (((char*)op - dest) + lastRun + 1 + ((lastRun+255-RUN_MASK)/255) > (U32)maxOutputSize)) return 0;  /* Check output limit */
        if (lastRun>=(int)RUN_MASK) { *op++=(RUN_MASK<<ML_BITS); lastRun-=RUN_MASK; for(; lastRun > 254 ; lastRun-=255) *op++ = 255; *op++ = (BYTE) lastRun; }
        else *op++ = (BYTE)(lastRun<<ML_BITS);
        memcpy(op, anchor, iend - anchor);
        op += iend-anchor;
    }

    /* End */
    return (int) (((char*)op)-dest);
}


int LZ4_sizeofStateHC(void) { return sizeof(LZ4HC_Data_Structure); }

int LZ4_compress_HC_extStateHC (void* state, const char* src, char* dst, int srcSize, int maxDstSize, int compressionLevel)
{
    if (((size_t)(state)&(sizeof(void*)-1)) != 0) return 0;   /* Error : state is not aligned for pointers (32 or 64 bits) */
    LZ4HC_init ((LZ4HC_Data_Structure*)state, (const BYTE*)src);
    if (maxDstSize < LZ4_compressBound(srcSize))
        return LZ4HC_compress_generic (state, src, dst, srcSize, maxDstSize, compressionLevel, limitedOutput);
    else
        return LZ4HC_compress_generic (state, src, dst, srcSize, maxDstSize, compressionLevel, noLimit);
}

int LZ4_compress_HC(const char* src, char* dst, int srcSize, int maxDstSize, int compressionLevel)
{
#if LZ4HC_HEAPMODE==1
    LZ4HC_Data_Structure* statePtr = malloc(sizeof(LZ4HC_Data_Structure));
#else
    LZ4HC_Data_Structure state;
    LZ4HC_Data_Structure* const statePtr = &state;
#endif
    int cSize = LZ4_compress_HC_extStateHC(statePtr, src, dst, srcSize, maxDstSize, compressionLevel);
#if LZ4HC_HEAPMODE==1
    free(statePtr);
#endif
    return cSize;
}



/**************************************
*  Streaming Functions
**************************************/
/* allocation */
LZ4_streamHC_t* LZ4_createStreamHC(void) { return (LZ4_streamHC_t*)malloc(sizeof(LZ4_streamHC_t)); }
int             LZ4_freeStreamHC (LZ4_streamHC_t* LZ4_streamHCPtr) { free(LZ4_streamHCPtr); return 0; }


/* initialization */
void LZ4_resetStreamHC (LZ4_streamHC_t* LZ4_streamHCPtr, int compressionLevel)
{
    LZ4_STATIC_ASSERT(sizeof(LZ4HC_Data_Structure) <= sizeof(LZ4_streamHC_t));   /* if compilation fails here, LZ4_STREAMHCSIZE must be increased */
    ((LZ4HC_Data_Structure*)LZ4_streamHCPtr)->base = NULL;
    ((LZ4HC_Data_Structure*)LZ4_streamHCPtr)->compressionLevel = (unsigned)compressionLevel;
}

int LZ4_loadDictHC (LZ4_streamHC_t* LZ4_streamHCPtr, const char* dictionary, int dictSize)
{
    LZ4HC_Data_Structure* ctxPtr = (LZ4HC_Data_Structure*) LZ4_streamHCPtr;
    if (dictSize > 64 KB) {
        dictionary += dictSize - 64 KB;
        dictSize = 64 KB;
    }
    LZ4HC_init (ctxPtr, (const BYTE*)dictionary);
    if (dictSize >= 4) LZ4HC_Insert (ctxPtr, (const BYTE*)dictionary +(dictSize-3));
    ctxPtr->end = (const BYTE*)dictionary + dictSize;
    return dictSize;
}


/* compression */

static void LZ4HC_setExternalDict(LZ4HC_Data_Structure* ctxPtr, const BYTE* newBlock)
{
    if (ctxPtr->end >= ctxPtr->base + 4) LZ4HC_Insert (ctxPtr, ctxPtr->end-3);   /* Referencing remaining dictionary content */
    /* Only one memory segment for extDict, so any previous extDict is lost at this stage */
    ctxPtr->lowLimit  = ctxPtr->dictLimit;
    ctxPtr->dictLimit = (U32)(ctxPtr->end - ctxPtr->base);
    ctxPtr->dictBase  = ctxPtr->base;
    ctxPtr->base = newBlock - ctxPtr->dictLimit;
    ctxPtr->end  = newBlock;
    ctxPtr->nextToUpdate = ctxPtr->dictLimit;   /* match referencing will resume from there */
}

static int LZ4_compressHC_continue_generic (LZ4HC_Data_Structure* ctxPtr,
                                            const char* source, char* dest,
                                            int inputSize, int maxOutputSize, limitedOutput_directive limit)
{
    /* auto-init if forgotten */
    if (ctxPtr->base == NULL) LZ4HC_init (ctxPtr, (const BYTE*) source);

    /* Check overflow */
    if ((size_t)(ctxPtr->end - ctxPtr->base) > 2 GB) {
        size_t dictSize = (size_t)(ctxPtr->end - ctxPtr->base) - ctxPtr->dictLimit;
        if (dictSize > 64 KB) dictSize = 64 KB;
        LZ4_loadDictHC((LZ4_streamHC_t*)ctxPtr, (const char*)(ctxPtr->end) - dictSize, (int)dictSize);
    }

    /* Check if blocks follow each other */
    if ((const BYTE*)source != ctxPtr->end) LZ4HC_setExternalDict(ctxPtr, (const BYTE*)source);

    /* Check overlapping input/dictionary space */
    {   const BYTE* sourceEnd = (const BYTE*) source + inputSize;
        const BYTE* const dictBegin = ctxPtr->dictBase + ctxPtr->lowLimit;
        const BYTE* const dictEnd   = ctxPtr->dictBase + ctxPtr->dictLimit;
        if ((sourceEnd > dictBegin) && ((const BYTE*)source < dictEnd)) {
            if (sourceEnd > dictEnd) sourceEnd = dictEnd;
            ctxPtr->lowLimit = (U32)(sourceEnd - ctxPtr->dictBase);
            if (ctxPtr->dictLimit - ctxPtr->lowLimit < 4) ctxPtr->lowLimit = ctxPtr->dictLimit;
        }
    }

    return LZ4HC_compress_generic (ctxPtr, source, dest, inputSize, maxOutputSize, ctxPtr->compressionLevel, limit);
}

int LZ4_compress_HC_continue (LZ4_streamHC_t* LZ4_streamHCPtr, const char* source, char* dest, int inputSize, int maxOutputSize)
{
    if (maxOutputSize < LZ4_compressBound(inputSize))
        return LZ4_compressHC_continue_generic ((LZ4HC_Data_Structure*)LZ4_streamHCPtr, source, dest, inputSize, maxOutputSize, limitedOutput);
    else
        return LZ4_compressHC_continue_generic ((LZ4HC_Data_Structure*)LZ4_streamHCPtr, source, dest, inputSize, maxOutputSize, noLimit);
}


/* dictionary saving */

int LZ4_saveDictHC (LZ4_streamHC_t* LZ4_streamHCPtr, char* safeBuffer, int dictSize)
{
    LZ4HC_Data_Structure* const streamPtr = (LZ4HC_Data_Structure*)LZ4_streamHCPtr;
    int const prefixSize = (int)(streamPtr->end - (streamPtr->base + streamPtr->dictLimit));
    if (dictSize > 64 KB) dictSize = 64 KB;
    if (dictSize < 4) dictSize = 0;
    if (dictSize > prefixSize) dictSize = prefixSize;
    memmove(safeBuffer, streamPtr->end - dictSize, dictSize);
    {   U32 const endIndex = (U32)(streamPtr->end - streamPtr->base);
        streamPtr->end = (const BYTE*)safeBuffer + dictSize;
        streamPtr->base = streamPtr->end - endIndex;
        streamPtr->dictLimit = endIndex - dictSize;
        streamPtr->lowLimit = endIndex - dictSize;
        if (streamPtr->nextToUpdate < streamPtr->dictLimit) streamPtr->nextToUpdate = streamPtr->dictLimit;
    }
    return dictSize;
}


/***********************************
*  Deprecated Functions
***********************************/
/* Deprecated compression functions */
/* These functions are planned to start generate warnings by r131 approximately */
int LZ4_compressHC(const char* src, char* dst, int srcSize) { return LZ4_compress_HC (src, dst, srcSize, LZ4_compressBound(srcSize), 0); }
int LZ4_compressHC_limitedOutput(const char* src, char* dst, int srcSize, int maxDstSize) { return LZ4_compress_HC(src, dst, srcSize, maxDstSize, 0); }
int LZ4_compressHC2(const char* src, char* dst, int srcSize, int cLevel) { return LZ4_compress_HC (src, dst, srcSize, LZ4_compressBound(srcSize), cLevel); }
int LZ4_compressHC2_limitedOutput(const char* src, char* dst, int srcSize, int maxDstSize, int cLevel) { return LZ4_compress_HC(src, dst, srcSize, maxDstSize, cLevel); }
int LZ4_compressHC_withStateHC (void* state, const char* src, char* dst, int srcSize) { return LZ4_compress_HC_extStateHC (state, src, dst, srcSize, LZ4_compressBound(srcSize), 0); }
int LZ4_compressHC_limitedOutput_withStateHC (void* state, const char* src, char* dst, int srcSize, int maxDstSize) { return LZ4_compress_HC_extStateHC (state, src, dst, srcSize, maxDstSize, 0); }
int LZ4_compressHC2_withStateHC (void* state, const char* src, char* dst, int srcSize, int cLevel) { return LZ4_compress_HC_extStateHC(state, src, dst, srcSize, LZ4_compressBound(srcSize), cLevel); }
int LZ4_compressHC2_limitedOutput_withStateHC (void* state, const char* src, char* dst, int srcSize, int maxDstSize, int cLevel) { return LZ4_compress_HC_extStateHC(state, src, dst, srcSize, maxDstSize, cLevel); }
int LZ4_compressHC_continue (LZ4_streamHC_t* ctx, const char* src, char* dst, int srcSize) { return LZ4_compress_HC_continue (ctx, src, dst, srcSize, LZ4_compressBound(srcSize)); }
int LZ4_compressHC_limitedOutput_continue (LZ4_streamHC_t* ctx, const char* src, char* dst, int srcSize, int maxDstSize) { return LZ4_compress_HC_continue (ctx, src, dst, srcSize, maxDstSize); }


/* Deprecated streaming functions */
/* These functions currently generate deprecation warnings */
int LZ4_sizeofStreamStateHC(void) { return LZ4_STREAMHCSIZE; }

int LZ4_resetStreamStateHC(void* state, char* inputBuffer)
{
    if ((((size_t)state) & (sizeof(void*)-1)) != 0) return 1;   /* Error : pointer is not aligned for pointer (32 or 64 bits) */
    LZ4HC_init((LZ4HC_Data_Structure*)state, (const BYTE*)inputBuffer);
    ((LZ4HC_Data_Structure*)state)->inputBuffer = (BYTE*)inputBuffer;
    return 0;
}

void* LZ4_createHC (char* inputBuffer)
{
    void* hc4 = ALLOCATOR(1, sizeof(LZ4HC_Data_Structure));
    if (hc4 == NULL) return NULL;   /* not enough memory */
    LZ4HC_init ((LZ4HC_Data_Structure*)hc4, (const BYTE*)inputBuffer);
    ((LZ4HC_Data_Structure*)hc4)->inputBuffer = (BYTE*)inputBuffer;
    return hc4;
}

int LZ4_freeHC (void* LZ4HC_Data)
{
    FREEMEM(LZ4HC_Data);
    return (0);
}

int LZ4_compressHC2_continue (void* LZ4HC_Data, const char* source, char* dest, int inputSize, int compressionLevel)
{
    return LZ4HC_compress_generic (LZ4HC_Data, source, dest, inputSize, 0, compressionLevel, noLimit);
}

int LZ4_compressHC2_limitedOutput_continue (void* LZ4HC_Data, const char* source, char* dest, int inputSize, int maxOutputSize, int compressionLevel)
{
    return LZ4HC_compress_generic (LZ4HC_Data, source, dest, inputSize, maxOutputSize, compressionLevel, limitedOutput);
}

char* LZ4_slideInputBufferHC(void* LZ4HC_Data)
{
    LZ4HC_Data_Structure* hc4 = (LZ4HC_Data_Structure*)LZ4HC_Data;
    int dictSize = LZ4_saveDictHC((LZ4_streamHC_t*)LZ4HC_Data, (char*)(hc4->inputBuffer), 64 KB);
    return (char*)(hc4->inputBuffer + dictSize);
}