Rewrite synopsis to clarify the two primary uses of git-checkout.
[git/git.git] / diff-delta.c
CommitLineData
a310d434
NP
1/*
2 * diff-delta.c: generate a delta between two buffers
3 *
4 * Many parts of this file have been lifted from LibXDiff version 0.10.
5 * http://www.xmailserver.org/xdiff-lib.html
6 *
7 * LibXDiff was written by Davide Libenzi <davidel@xmailserver.org>
8 * Copyright (C) 2003 Davide Libenzi
9 *
10 * Many mods for GIT usage by Nicolas Pitre <nico@cam.org>, (C) 2005.
11 *
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.
16 *
17 * Use of this within git automatically means that the LGPL
18 * licensing gets turned into GPLv2 within this project.
19 */
20
21#include <stdlib.h>
8e1454b5
NP
22#include <string.h>
23#include <zlib.h>
a310d434
NP
24#include "delta.h"
25
26
27/* block size: min = 16, max = 64k, power of 2 */
28#define BLK_SIZE 16
29
30#define MIN(a, b) ((a) < (b) ? (a) : (b))
31
32#define GR_PRIME 0x9e370001
8e1454b5 33#define HASH(v, shift) (((unsigned int)(v) * GR_PRIME) >> (shift))
a310d434 34
8e1454b5 35struct index {
a310d434 36 const unsigned char *ptr;
8e1454b5
NP
37 unsigned int val;
38 struct index *next;
39};
a310d434 40
8e1454b5
NP
41static struct index ** delta_index(const unsigned char *buf,
42 unsigned long bufsize,
c13c6bf7 43 unsigned long trg_bufsize,
8e1454b5 44 unsigned int *hash_shift)
a310d434 45{
c13c6bf7 46 unsigned int i, hsize, hshift, hlimit, entries, *hash_count;
8e1454b5
NP
47 const unsigned char *data;
48 struct index *entry, **hash;
49 void *mem;
50
51 /* determine index hash size */
c13c6bf7 52 entries = bufsize / BLK_SIZE;
8e1454b5 53 hsize = entries / 4;
c13c6bf7 54 for (i = 4; (1 << i) < hsize && i < 31; i++);
8e1454b5
NP
55 hsize = 1 << i;
56 hshift = 32 - i;
57 *hash_shift = hshift;
58
59 /* allocate lookup index */
60 mem = malloc(hsize * sizeof(*hash) + entries * sizeof(*entry));
61 if (!mem)
62 return NULL;
63 hash = mem;
64 entry = mem + hsize * sizeof(*hash);
65 memset(hash, 0, hsize * sizeof(*hash));
66
c13c6bf7
NP
67 /* allocate an array to count hash entries */
68 hash_count = calloc(hsize, sizeof(*hash_count));
69 if (!hash_count) {
70 free(hash);
71 return NULL;
72 }
73
74 /* then populate the index */
8e1454b5 75 data = buf + entries * BLK_SIZE - BLK_SIZE;
8e1454b5 76 while (data >= buf) {
c13c6bf7 77 unsigned int val = adler32(0, data, BLK_SIZE);
8e1454b5
NP
78 i = HASH(val, hshift);
79 entry->ptr = data;
80 entry->val = val;
81 entry->next = hash[i];
82 hash[i] = entry++;
c13c6bf7 83 hash_count[i]++;
a310d434 84 data -= BLK_SIZE;
8e1454b5 85 }
a310d434 86
c13c6bf7
NP
87 /*
88 * Determine a limit on the number of entries in the same hash
89 * bucket. This guard us against patological data sets causing
90 * really bad hash distribution with most entries in the same hash
91 * bucket that would bring us to O(m*n) computing costs (m and n
92 * corresponding to reference and target buffer sizes).
93 *
94 * The more the target buffer is large, the more it is important to
95 * have small entry lists for each hash buckets. With such a limit
96 * the cost is bounded to something more like O(m+n).
97 */
98 hlimit = (1 << 26) / trg_bufsize;
99 if (hlimit < 4*BLK_SIZE)
100 hlimit = 4*BLK_SIZE;
101
102 /*
103 * Now make sure none of the hash buckets has more entries than
104 * we're willing to test. Otherwise we cull the entry list
105 * uniformly to still preserve a good repartition across
106 * the reference buffer.
107 */
108 for (i = 0; i < hsize; i++) {
109 if (hash_count[i] < hlimit)
110 continue;
111 entry = hash[i];
112 do {
113 struct index *keep = entry;
114 int skip = hash_count[i] / hlimit / 2;
115 do {
116 entry = entry->next;
117 } while(--skip && entry);
118 keep->next = entry;
119 } while(entry);
120 }
121 free(hash_count);
122
8e1454b5 123 return hash;
a310d434
NP
124}
125
fe474b58 126/* provide the size of the copy opcode given the block offset and size */
a310d434
NP
127#define COPYOP_SIZE(o, s) \
128 (!!(o & 0xff) + !!(o & 0xff00) + !!(o & 0xff0000) + !!(o & 0xff000000) + \
129 !!(s & 0xff) + !!(s & 0xff00) + 1)
130
fe474b58
NP
131/* the maximum size for any opcode */
132#define MAX_OP_SIZE COPYOP_SIZE(0xffffffff, 0xffffffff)
133
a310d434
NP
134void *diff_delta(void *from_buf, unsigned long from_size,
135 void *to_buf, unsigned long to_size,
75c42d8c
LT
136 unsigned long *delta_size,
137 unsigned long max_size)
a310d434 138{
8e1454b5
NP
139 unsigned int i, outpos, outsize, inscnt, hash_shift;
140 const unsigned char *ref_data, *ref_top, *data, *top;
141 unsigned char *out;
142 struct index *entry, **hash;
a310d434 143
8e1454b5 144 if (!from_size || !to_size)
a310d434 145 return NULL;
c13c6bf7 146 hash = delta_index(from_buf, from_size, to_size, &hash_shift);
8e1454b5
NP
147 if (!hash)
148 return NULL;
149
a310d434
NP
150 outpos = 0;
151 outsize = 8192;
fe474b58
NP
152 if (max_size && outsize >= max_size)
153 outsize = max_size + MAX_OP_SIZE + 1;
a310d434
NP
154 out = malloc(outsize);
155 if (!out) {
8e1454b5 156 free(hash);
a310d434
NP
157 return NULL;
158 }
159
e5e3a9d8
NP
160 ref_data = from_buf;
161 ref_top = from_buf + from_size;
a310d434
NP
162 data = to_buf;
163 top = to_buf + to_size;
164
165 /* store reference buffer size */
69a2d426
NP
166 out[outpos++] = from_size;
167 from_size >>= 7;
168 while (from_size) {
169 out[outpos - 1] |= 0x80;
170 out[outpos++] = from_size;
171 from_size >>= 7;
172 }
a310d434
NP
173
174 /* store target buffer size */
69a2d426
NP
175 out[outpos++] = to_size;
176 to_size >>= 7;
177 while (to_size) {
178 out[outpos - 1] |= 0x80;
179 out[outpos++] = to_size;
180 to_size >>= 7;
181 }
a310d434
NP
182
183 inscnt = 0;
a310d434 184
8e1454b5
NP
185 while (data < top) {
186 unsigned int moff = 0, msize = 0;
c13c6bf7
NP
187 if (data + BLK_SIZE <= top) {
188 unsigned int val = adler32(0, data, BLK_SIZE);
189 i = HASH(val, hash_shift);
190 for (entry = hash[i]; entry; entry = entry->next) {
191 const unsigned char *ref = entry->ptr;
192 const unsigned char *src = data;
193 unsigned int ref_size = ref_top - ref;
194 if (entry->val != val)
195 continue;
196 if (ref_size > top - src)
197 ref_size = top - src;
198 if (ref_size > 0x10000)
199 ref_size = 0x10000;
200 if (ref_size <= msize)
8e1454b5 201 break;
c13c6bf7
NP
202 while (ref_size-- && *src++ == *ref)
203 ref++;
204 if (msize < ref - entry->ptr) {
205 /* this is our best match so far */
206 msize = ref - entry->ptr;
207 moff = entry->ptr - ref_data;
a310d434
NP
208 }
209 }
210 }
211
212 if (!msize || msize < COPYOP_SIZE(moff, msize)) {
213 if (!inscnt)
214 outpos++;
215 out[outpos++] = *data++;
216 inscnt++;
217 if (inscnt == 0x7f) {
218 out[outpos - inscnt - 1] = inscnt;
219 inscnt = 0;
220 }
221 } else {
8e1454b5
NP
222 unsigned char *op;
223
a310d434
NP
224 if (inscnt) {
225 out[outpos - inscnt - 1] = inscnt;
226 inscnt = 0;
227 }
228
229 data += msize;
8e1454b5 230 op = out + outpos++;
a310d434
NP
231 i = 0x80;
232
233 if (moff & 0xff) { out[outpos++] = moff; i |= 0x01; }
234 moff >>= 8;
235 if (moff & 0xff) { out[outpos++] = moff; i |= 0x02; }
236 moff >>= 8;
237 if (moff & 0xff) { out[outpos++] = moff; i |= 0x04; }
238 moff >>= 8;
239 if (moff & 0xff) { out[outpos++] = moff; i |= 0x08; }
240
241 if (msize & 0xff) { out[outpos++] = msize; i |= 0x10; }
242 msize >>= 8;
243 if (msize & 0xff) { out[outpos++] = msize; i |= 0x20; }
244
8e1454b5 245 *op = i;
a310d434
NP
246 }
247
fe474b58 248 if (outpos >= outsize - MAX_OP_SIZE) {
a310d434
NP
249 void *tmp = out;
250 outsize = outsize * 3 / 2;
fe474b58
NP
251 if (max_size && outsize >= max_size)
252 outsize = max_size + MAX_OP_SIZE + 1;
253 if (max_size && outpos > max_size)
254 out = NULL;
255 else
256 out = realloc(out, outsize);
a310d434
NP
257 if (!out) {
258 free(tmp);
8e1454b5 259 free(hash);
a310d434
NP
260 return NULL;
261 }
262 }
263 }
264
265 if (inscnt)
266 out[outpos - inscnt - 1] = inscnt;
267
8e1454b5 268 free(hash);
a310d434
NP
269 *delta_size = outpos;
270 return out;
271}