2 * diff-delta.c: generate a delta between two buffers
4 * Many parts of this file have been lifted from LibXDiff version 0.10.
5 * http://www.xmailserver.org/xdiff-lib.html
7 * LibXDiff was written by Davide Libenzi <davidel@xmailserver.org>
8 * Copyright (C) 2003 Davide Libenzi
10 * Many mods for GIT usage by Nicolas Pitre <nico@cam.org>, (C) 2005.
12 * This file is free software; you can redistribute it and/or
13 * modify it under the terms of the GNU Lesser General Public
14 * License as published by the Free Software Foundation; either
15 * version 2.1 of the License, or (at your option) any later version.
17 * Use of this within git automatically means that the LGPL
18 * licensing gets turned into GPLv2 within this project.
27 /* block size: min = 16, max = 64k, power of 2 */
30 #define MIN(a, b) ((a) < (b) ? (a) : (b))
32 #define GR_PRIME 0x9e370001
33 #define HASH(v, shift) (((unsigned int)(v) * GR_PRIME) >> (shift))
36 const unsigned char *ptr;
41 static struct index ** delta_index(const unsigned char *buf,
42 unsigned long bufsize,
43 unsigned int *hash_shift)
45 unsigned int hsize, hshift, entries, blksize, i;
46 const unsigned char *data;
47 struct index *entry, **hash;
50 /* determine index hash size */
51 entries = (bufsize + BLK_SIZE - 1) / BLK_SIZE;
53 for (i = 4; (1 << i) < hsize && i < 16; i++);
58 /* allocate lookup index */
59 mem = malloc(hsize * sizeof(*hash) + entries * sizeof(*entry));
63 entry = mem + hsize * sizeof(*hash);
64 memset(hash, 0, hsize * sizeof(*hash));
66 /* then populate it */
67 data = buf + entries * BLK_SIZE - BLK_SIZE;
68 blksize = bufsize - (data - buf);
70 unsigned int val = adler32(0, data, blksize);
71 i = HASH(val, hshift);
74 entry->next = hash[i];
83 /* provide the size of the copy opcode given the block offset and size */
84 #define COPYOP_SIZE(o, s) \
85 (!!(o & 0xff) + !!(o & 0xff00) + !!(o & 0xff0000) + !!(o & 0xff000000) + \
86 !!(s & 0xff) + !!(s & 0xff00) + 1)
88 /* the maximum size for any opcode */
89 #define MAX_OP_SIZE COPYOP_SIZE(0xffffffff, 0xffffffff)
91 void *diff_delta(void *from_buf, unsigned long from_size,
92 void *to_buf, unsigned long to_size,
93 unsigned long *delta_size,
94 unsigned long max_size)
96 unsigned int i, outpos, outsize, inscnt, hash_shift;
97 const unsigned char *ref_data, *ref_top, *data, *top;
99 struct index *entry, **hash;
101 if (!from_size || !to_size)
103 hash = delta_index(from_buf, from_size, &hash_shift);
109 if (max_size && outsize >= max_size)
110 outsize = max_size + MAX_OP_SIZE + 1;
111 out = malloc(outsize);
118 ref_top = from_buf + from_size;
120 top = to_buf + to_size;
122 /* store reference buffer size */
123 out[outpos++] = from_size;
126 out[outpos - 1] |= 0x80;
127 out[outpos++] = from_size;
131 /* store target buffer size */
132 out[outpos++] = to_size;
135 out[outpos - 1] |= 0x80;
136 out[outpos++] = to_size;
143 unsigned int moff = 0, msize = 0;
144 unsigned int blksize = MIN(top - data, BLK_SIZE);
145 unsigned int val = adler32(0, data, blksize);
146 i = HASH(val, hash_shift);
147 for (entry = hash[i]; entry; entry = entry->next) {
148 const unsigned char *ref = entry->ptr;
149 const unsigned char *src = data;
150 unsigned int ref_size = ref_top - ref;
151 if (entry->val != val)
153 if (ref_size > top - src)
154 ref_size = top - src;
155 while (ref_size && *src++ == *ref) {
159 ref_size = ref - entry->ptr;
160 if (ref_size > msize) {
161 /* this is our best match so far */
162 moff = entry->ptr - ref_data;
164 if (msize >= 0x10000) {
171 if (!msize || msize < COPYOP_SIZE(moff, msize)) {
174 out[outpos++] = *data++;
176 if (inscnt == 0x7f) {
177 out[outpos - inscnt - 1] = inscnt;
184 out[outpos - inscnt - 1] = inscnt;
192 if (moff & 0xff) { out[outpos++] = moff; i |= 0x01; }
194 if (moff & 0xff) { out[outpos++] = moff; i |= 0x02; }
196 if (moff & 0xff) { out[outpos++] = moff; i |= 0x04; }
198 if (moff & 0xff) { out[outpos++] = moff; i |= 0x08; }
200 if (msize & 0xff) { out[outpos++] = msize; i |= 0x10; }
202 if (msize & 0xff) { out[outpos++] = msize; i |= 0x20; }
207 if (outpos >= outsize - MAX_OP_SIZE) {
209 outsize = outsize * 3 / 2;
210 if (max_size && outsize >= max_size)
211 outsize = max_size + MAX_OP_SIZE + 1;
212 if (max_size && outpos > max_size)
215 out = realloc(out, outsize);
225 out[outpos - inscnt - 1] = inscnt;
228 *delta_size = outpos;