2 * Copyright (C) 2010, Google Inc.
3 * and other copyright owners as documented in JGit's IP log.
5 * This program and the accompanying materials are made available
6 * under the terms of the Eclipse Distribution License v1.0 which
7 * accompanies this distribution, is reproduced below, and is
8 * available at http://www.eclipse.org/org/documents/edl-v10.php
10 * All rights reserved.
12 * Redistribution and use in source and binary forms, with or
13 * without modification, are permitted provided that the following
16 * - Redistributions of source code must retain the above copyright
17 * notice, this list of conditions and the following disclaimer.
19 * - Redistributions in binary form must reproduce the above
20 * copyright notice, this list of conditions and the following
21 * disclaimer in the documentation and/or other materials provided
22 * with the distribution.
24 * - Neither the name of the Eclipse Foundation, Inc. nor the
25 * names of its contributors may be used to endorse or promote
26 * products derived from this software without specific prior
29 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND
30 * CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES,
31 * INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
32 * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
33 * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR
34 * CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
35 * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
36 * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
37 * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
38 * CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT,
39 * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
40 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF
41 * ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
48 #define MAX_PTR UINT_MAX
49 #define MAX_CNT UINT_MAX
51 #define LINE_END(n) (line##n + count##n - 1)
52 #define LINE_END_PTR(n) (*line##n + *count##n - 1)
56 unsigned int ptr, cnt;
58 } **records, /* an occurrence */
59 **line_map; /* map of line to record chain */
61 unsigned int *next_ptrs;
62 unsigned int table_bits,
66 unsigned int max_chain_length,
78 unsigned int begin1, end1;
79 unsigned int begin2, end2;
82 #define LINE_MAP(i, a) (i->line_map[(a) - i->ptr_shift])
84 #define NEXT_PTR(index, ptr) \
85 (index->next_ptrs[(ptr) - index->ptr_shift])
87 #define CNT(index, ptr) \
88 ((LINE_MAP(index, ptr))->cnt)
90 #define REC(env, s, l) \
91 (env->xdf##s.recs[l - 1])
93 static int cmp_recs(xpparam_t const *xpp,
94 xrecord_t *r1, xrecord_t *r2)
96 return r1->ha == r2->ha &&
97 xdl_recmatch(r1->ptr, r1->size, r2->ptr, r2->size,
101 #define CMP_ENV(xpp, env, s1, l1, s2, l2) \
102 (cmp_recs(xpp, REC(env, s1, l1), REC(env, s2, l2)))
104 #define CMP(i, s1, l1, s2, l2) \
105 (cmp_recs(i->xpp, REC(i->env, s1, l1), REC(i->env, s2, l2)))
107 #define TABLE_HASH(index, side, line) \
108 XDL_HASHLONG((REC(index->env, side, line))->ha, index->table_bits)
110 static int scanA(struct histindex *index, int line1, int count1)
112 unsigned int ptr, tbl_idx;
113 unsigned int chain_len;
114 struct record **rec_chain, *rec;
116 for (ptr = LINE_END(1); line1 <= ptr; ptr--) {
117 tbl_idx = TABLE_HASH(index, 1, ptr);
118 rec_chain = index->records + tbl_idx;
123 if (CMP(index, 1, rec->ptr, 1, ptr)) {
125 * ptr is identical to another element. Insert
126 * it onto the front of the existing element
129 NEXT_PTR(index, ptr) = rec->ptr;
131 /* cap rec->cnt at MAX_CNT */
132 rec->cnt = XDL_MIN(MAX_CNT, rec->cnt + 1);
133 LINE_MAP(index, ptr) = rec;
141 if (chain_len == index->max_chain_length)
145 * This is the first time we have ever seen this particular
146 * element in the sequence. Construct a new chain for it.
148 if (!(rec = xdl_cha_alloc(&index->rcha)))
152 rec->next = *rec_chain;
154 LINE_MAP(index, ptr) = rec;
163 static int try_lcs(struct histindex *index, struct region *lcs, int b_ptr,
164 int line1, int count1, int line2, int count2)
166 unsigned int b_next = b_ptr + 1;
167 struct record *rec = index->records[TABLE_HASH(index, 2, b_ptr)];
168 unsigned int as, ae, bs, be, np, rc;
171 for (; rec; rec = rec->next) {
172 if (rec->cnt > index->cnt) {
173 if (!index->has_common)
174 index->has_common = CMP(index, 1, rec->ptr, 2, b_ptr);
179 if (!CMP(index, 1, as, 2, b_ptr))
182 index->has_common = 1;
185 np = NEXT_PTR(index, as);
191 while (line1 < as && line2 < bs
192 && CMP(index, 1, as - 1, 2, bs - 1)) {
196 rc = XDL_MIN(rc, CNT(index, as));
198 while (ae < LINE_END(1) && be < LINE_END(2)
199 && CMP(index, 1, ae + 1, 2, be + 1)) {
203 rc = XDL_MIN(rc, CNT(index, ae));
208 if (lcs->end1 - lcs->begin1 < ae - as || rc < index->cnt) {
220 np = NEXT_PTR(index, np);
236 static int find_lcs(struct histindex *index, struct region *lcs,
237 int line1, int count1, int line2, int count2) {
240 if (scanA(index, line1, count1))
243 index->cnt = index->max_chain_length + 1;
245 for (b_ptr = line2; b_ptr <= LINE_END(2); )
246 b_ptr = try_lcs(index, lcs, b_ptr, line1, count1, line2, count2);
248 return index->has_common && index->max_chain_length < index->cnt;
251 static int fall_back_to_classic_diff(struct histindex *index,
252 int line1, int count1, int line2, int count2)
255 xpp.flags = index->xpp->flags & ~XDF_DIFF_ALGORITHM_MASK;
257 return xdl_fall_back_diff(index->env, &xpp,
258 line1, count1, line2, count2);
261 static int histogram_diff(xpparam_t const *xpp, xdfenv_t *env,
262 int line1, int count1, int line2, int count2)
264 struct histindex index;
269 if (count1 <= 0 && count2 <= 0)
272 if (LINE_END(1) >= MAX_PTR)
277 env->xdf2.rchg[line2++ - 1] = 1;
279 } else if (!count2) {
281 env->xdf1.rchg[line1++ - 1] = 1;
285 memset(&index, 0, sizeof(index));
290 index.records = NULL;
291 index.line_map = NULL;
292 /* in case of early xdl_cha_free() */
293 index.rcha.head = NULL;
295 index.table_bits = xdl_hashbits(count1);
296 sz = index.records_size = 1 << index.table_bits;
297 sz *= sizeof(struct record *);
298 if (!(index.records = (struct record **) xdl_malloc(sz)))
300 memset(index.records, 0, sz);
302 sz = index.line_map_size = count1;
303 sz *= sizeof(struct record *);
304 if (!(index.line_map = (struct record **) xdl_malloc(sz)))
306 memset(index.line_map, 0, sz);
308 sz = index.line_map_size;
309 sz *= sizeof(unsigned int);
310 if (!(index.next_ptrs = (unsigned int *) xdl_malloc(sz)))
312 memset(index.next_ptrs, 0, sz);
314 /* lines / 4 + 1 comes from xprepare.c:xdl_prepare_ctx() */
315 if (xdl_cha_init(&index.rcha, sizeof(struct record), count1 / 4 + 1) < 0)
318 index.ptr_shift = line1;
319 index.max_chain_length = 64;
321 memset(&lcs, 0, sizeof(lcs));
322 if (find_lcs(&index, &lcs, line1, count1, line2, count2))
323 result = fall_back_to_classic_diff(&index, line1, count1, line2, count2);
325 if (lcs.begin1 == 0 && lcs.begin2 == 0) {
327 env->xdf1.rchg[line1++ - 1] = 1;
329 env->xdf2.rchg[line2++ - 1] = 1;
332 result = histogram_diff(xpp, env,
333 line1, lcs.begin1 - line1,
334 line2, lcs.begin2 - line2);
337 result = histogram_diff(xpp, env,
338 lcs.end1 + 1, LINE_END(1) - lcs.end1,
339 lcs.end2 + 1, LINE_END(2) - lcs.end2);
346 xdl_free(index.records);
347 xdl_free(index.line_map);
348 xdl_free(index.next_ptrs);
349 xdl_cha_free(&index.rcha);
354 int xdl_do_histogram_diff(mmfile_t *file1, mmfile_t *file2,
355 xpparam_t const *xpp, xdfenv_t *env)
357 if (xdl_prepare_env(file1, file2, xpp, env) < 0)
360 return histogram_diff(xpp, env,
361 env->xdf1.dstart + 1, env->xdf1.dend - env->xdf1.dstart + 1,
362 env->xdf2.dstart + 1, env->xdf2.dend - env->xdf2.dstart + 1);