@matheusbsbs / matheus-bagliano / commits / 8c912eea94

teach --histogram to diff

Port JGit's HistogramDiff algorithm over to C. Rough numbers (TODO) show that it is faster than its --patience cousin, as well as the default Meyers algorithm. The implementation has been reworked to use structs and pointers, instead of bitmasks, thus doing away with JGit's 2^28 line limit. We also use xdiff's default hash table implementation (xdl_hash_bits() with XDL_HASHLONG()) for convenience. Signed-off-by: Tay Ray Chuan <rctay89@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>

Tay Ray Chuan committed Jul 12, 2011 at 14:10 UTC 8c912eea94a2138e8bc608f7c390eb0b313effb0
8 files changed +407 -1
Makefile
+1 -1
@@ -1838,7 +1838,7 @@ ifndef NO_CURL
1838 GIT_OBJS += http.o http-walker.o remote-curl.o
1839 endif
1840 XDIFF_OBJS = xdiff/xdiffi.o xdiff/xprepare.o xdiff/xutils.o xdiff/xemit.o \
1841 - xdiff/xmerge.o xdiff/xpatience.o
1841 + xdiff/xmerge.o xdiff/xpatience.o xdiff/xhistogram.o
1842 VCSSVN_OBJS = vcs-svn/string_pool.o vcs-svn/line_buffer.o \
1843 vcs-svn/repo_tree.o vcs-svn/fast_export.o vcs-svn/svndump.o
1844 VCSSVN_TEST_OBJS = test-obj-pool.o test-string-pool.o \
diff.c
+2
@@ -3369,6 +3369,8 @@ int diff_opt_parse(struct diff_options *options, const char **av, int ac)
3369 DIFF_XDL_SET(options, IGNORE_WHITESPACE_AT_EOL);
3370 else if (!strcmp(arg, "--patience"))
3371 DIFF_XDL_SET(options, PATIENCE_DIFF);
3372 + else if (!strcmp(arg, "--histogram"))
3373 + DIFF_XDL_SET(options, HISTOGRAM_DIFF);
3374
3375 /* flags options */
3376 else if (!strcmp(arg, "--binary")) {
merge-recursive.c
+2
@@ -1759,6 +1759,8 @@ int parse_merge_opt(struct merge_options *o, const char *s)
1759 o->subtree_shift = s + strlen("subtree=");
1760 else if (!strcmp(s, "patience"))
1761 o->xdl_opts |= XDF_PATIENCE_DIFF;
1762 + else if (!strcmp(s, "histogram"))
1763 + o->xdl_opts |= XDF_HISTOGRAM_DIFF;
1764 else if (!strcmp(s, "ignore-space-change"))
1765 o->xdl_opts |= XDF_IGNORE_WHITESPACE_CHANGE;
1766 else if (!strcmp(s, "ignore-all-space"))
t/t4049-diff-histogram.sh new
+12
@@ -0,0 +1,12 @@
1 +#!/bin/sh
2 +
3 +test_description='histogram diff algorithm'
4 +
5 +. ./test-lib.sh
6 +. "$TEST_DIRECTORY"/lib-diff-alternative.sh
7 +
8 +test_diff_frobnitz "histogram"
9 +
10 +test_diff_unique "histogram"
11 +
12 +test_done
xdiff/xdiff.h
+1
@@ -33,6 +33,7 @@ extern "C" {
33 #define XDF_IGNORE_WHITESPACE_CHANGE (1 << 3)
34 #define XDF_IGNORE_WHITESPACE_AT_EOL (1 << 4)
35 #define XDF_PATIENCE_DIFF (1 << 5)
36 +#define XDF_HISTOGRAM_DIFF (1 << 6)
37 #define XDF_WHITESPACE_FLAGS (XDF_IGNORE_WHITESPACE | XDF_IGNORE_WHITESPACE_CHANGE | XDF_IGNORE_WHITESPACE_AT_EOL)
38
39 #define XDL_PATCH_NORMAL '-'
xdiff/xdiffi.c
+3
@@ -331,6 +331,9 @@ int xdl_do_diff(mmfile_t *mf1, mmfile_t *mf2, xpparam_t const *xpp,
331 if (xpp->flags & XDF_PATIENCE_DIFF)
332 return xdl_do_patience_diff(mf1, mf2, xpp, xe);
333
334 + if (xpp->flags & XDF_HISTOGRAM_DIFF)
335 + return xdl_do_histogram_diff(mf1, mf2, xpp, xe);
336 +
337 if (xdl_prepare_env(mf1, mf2, xpp, xe) < 0) {
338
339 return -1;
xdiff/xdiffi.h
+2
@@ -57,5 +57,7 @@ int xdl_emit_diff(xdfenv_t *xe, xdchange_t *xscr, xdemitcb_t *ecb,
57 xdemitconf_t const *xecfg);
58 int xdl_do_patience_diff(mmfile_t *mf1, mmfile_t *mf2, xpparam_t const *xpp,
59 xdfenv_t *env);
60 +int xdl_do_histogram_diff(mmfile_t *mf1, mmfile_t *mf2, xpparam_t const *xpp,
61 + xdfenv_t *env);
62
63 #endif /* #if !defined(XDIFFI_H) */
xdiff/xhistogram.c new
+384
@@ -0,0 +1,384 @@
1 +/*
2 + * Copyright (C) 2010, Google Inc.
3 + * and other copyright owners as documented in JGit's IP log.
4 + *
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
9 + *
10 + * All rights reserved.
11 + *
12 + * Redistribution and use in source and binary forms, with or
13 + * without modification, are permitted provided that the following
14 + * conditions are met:
15 + *
16 + * - Redistributions of source code must retain the above copyright
17 + * notice, this list of conditions and the following disclaimer.
18 + *
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.
23 + *
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
27 + * written permission.
28 + *
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.
42 + */
43 +
44 +#include "xinclude.h"
45 +#include "xtypes.h"
46 +#include "xdiff.h"
47 +
48 +#define MAX_PTR UINT_MAX
49 +#define MAX_CNT UINT_MAX
50 +
51 +#define LINE_END(n) (line##n + count##n - 1)
52 +#define LINE_END_PTR(n) (*line##n + *count##n - 1)
53 +
54 +struct histindex {
55 + struct record {
56 + unsigned int ptr, cnt;
57 + struct record *next;
58 + } **records, /* an ocurrence */
59 + **line_map; /* map of line to record chain */
60 + chastore_t rcha;
61 + unsigned int *next_ptrs;
62 + unsigned int table_bits,
63 + records_size,
64 + line_map_size;
65 +
66 + unsigned int max_chain_length,
67 + key_shift,
68 + ptr_shift;
69 +
70 + unsigned int cnt,
71 + has_common;
72 +
73 + xdfenv_t *env;
74 + xpparam_t const *xpp;
75 +};
76 +
77 +struct region {
78 + unsigned int begin1, end1;
79 + unsigned int begin2, end2;
80 +};
81 +
82 +#define LINE_MAP(i, a) (i->line_map[(a) - i->ptr_shift])
83 +
84 +#define NEXT_PTR(index, ptr) \
85 + (index->next_ptrs[(ptr) - index->ptr_shift])
86 +
87 +#define CNT(index, ptr) \
88 + ((LINE_MAP(index, ptr))->cnt)
89 +
90 +#define REC(env, s, l) \
91 + (env->xdf##s.recs[l - 1])
92 +
93 +static int cmp_recs(xpparam_t const *xpp,
94 + xrecord_t *r1, xrecord_t *r2)
95 +{
96 + return r1->ha == r2->ha &&
97 + xdl_recmatch(r1->ptr, r1->size, r2->ptr, r2->size,
98 + xpp->flags);
99 +}
100 +
101 +#define CMP_ENV(xpp, env, s1, l1, s2, l2) \
102 + (cmp_recs(xpp, REC(env, s1, l1), REC(env, s2, l2)))
103 +
104 +#define CMP(i, s1, l1, s2, l2) \
105 + (CMP_ENV(i->xpp, i->env, s1, l1, s2, l2))
106 +
107 +#define TABLE_HASH(index, side, line) \
108 + XDL_HASHLONG((REC(index->env, side, line))->ha, index->table_bits)
109 +
110 +static int scanA(struct histindex *index, int line1, int count1)
111 +{
112 + unsigned int ptr, tbl_idx;
113 + unsigned int chain_len;
114 + struct record **rec_chain, *rec;
115 +
116 + for (ptr = LINE_END(1); line1 <= ptr; ptr--) {
117 + tbl_idx = TABLE_HASH(index, 1, ptr);
118 + rec_chain = index->records + tbl_idx;
119 + rec = *rec_chain;
120 +
121 + chain_len = 0;
122 + while (rec) {
123 + if (CMP(index, 1, rec->ptr, 1, ptr)) {
124 + /*
125 + * ptr is identical to another element. Insert
126 + * it onto the front of the existing element
127 + * chain.
128 + */
129 + NEXT_PTR(index, ptr) = rec->ptr;
130 + rec->ptr = ptr;
131 + /* cap rec->cnt at MAX_CNT */
132 + rec->cnt = XDL_MIN(MAX_CNT, rec->cnt + 1);
133 + LINE_MAP(index, ptr) = rec;
134 + goto continue_scan;
135 + }
136 +
137 + rec = rec->next;
138 + chain_len++;
139 + }
140 +
141 + if (chain_len == index->max_chain_length)
142 + return -1;
143 +
144 + /*
145 + * This is the first time we have ever seen this particular
146 + * element in the sequence. Construct a new chain for it.
147 + */
148 + if (!(rec = xdl_cha_alloc(&index->rcha)))
149 + return -1;
150 + rec->ptr = ptr;
151 + rec->cnt = 1;
152 + rec->next = *rec_chain;
153 + *rec_chain = rec;
154 + LINE_MAP(index, ptr) = rec;
155 +
156 +continue_scan:
157 + ; /* no op */
158 + }
159 +
160 + return 0;
161 +}
162 +
163 +static int try_lcs(struct histindex *index, struct region *lcs, int b_ptr,
164 + int line1, int count1, int line2, int count2)
165 +{
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;
169 + int should_break;
170 +
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);
175 + continue;
176 + }
177 +
178 + as = rec->ptr;
179 + if (!CMP(index, 1, as, 2, b_ptr))
180 + continue;
181 +
182 + index->has_common = 1;
183 + for (;;) {
184 + should_break = 0;
185 + np = NEXT_PTR(index, as);
186 + bs = b_ptr;
187 + ae = as;
188 + be = bs;
189 + rc = rec->cnt;
190 +
191 + while (line1 < as && line2 < bs
192 + && CMP(index, 1, as - 1, 2, bs - 1)) {
193 + as--;
194 + bs--;
195 + if (1 < rc)
196 + rc = XDL_MIN(rc, CNT(index, as));
197 + }
198 + while (ae < LINE_END(1) && be < LINE_END(2)
199 + && CMP(index, 1, ae + 1, 2, be + 1)) {
200 + ae++;
201 + be++;
202 + if (1 < rc)
203 + rc = XDL_MIN(rc, CNT(index, ae));
204 + }
205 +
206 + if (b_next <= be)
207 + b_next = be + 1;
208 + if (lcs->end1 - lcs->begin1 < ae - as || rc < index->cnt) {
209 + lcs->begin1 = as;
210 + lcs->begin2 = bs;
211 + lcs->end1 = ae;
212 + lcs->end2 = be;
213 + index->cnt = rc;
214 + }
215 +
216 + if (np == 0)
217 + break;
218 +
219 + while (np <= ae) {
220 + np = NEXT_PTR(index, np);
221 + if (np == 0) {
222 + should_break = 1;
223 + break;
224 + }
225 + }
226 +
227 + if (should_break)
228 + break;
229 +
230 + as = np;
231 + }
232 + }
233 + return b_next;
234 +}
235 +
236 +static int find_lcs(struct histindex *index, struct region *lcs,
237 + int line1, int count1, int line2, int count2) {
238 + int b_ptr;
239 +
240 + if (scanA(index, line1, count1))
241 + return -1;
242 +
243 + index->cnt = index->max_chain_length + 1;
244 +
245 + for (b_ptr = line2; b_ptr <= LINE_END(2); )
246 + b_ptr = try_lcs(index, lcs, b_ptr, line1, count1, line2, count2);
247 +
248 + return index->has_common && index->max_chain_length < index->cnt;
249 +}
250 +
251 +static void reduce_common_start_end(xpparam_t const *xpp, xdfenv_t *env,
252 + int *line1, int *count1, int *line2, int *count2)
253 +{
254 + if (*count1 <= 1 || *count2 <= 1)
255 + return;
256 + while (*count1 > 1 && *count2 > 1 && CMP_ENV(xpp, env, 1, *line1, 2, *line2)) {
257 + (*line1)++;
258 + (*count1)--;
259 + (*line2)++;
260 + (*count2)--;
261 + }
262 + while (*count1 > 1 && *count2 > 1 && CMP_ENV(xpp, env, 1, LINE_END_PTR(1), 2, LINE_END_PTR(2))) {
263 + (*count1)--;
264 + (*count2)--;
265 + }
266 +}
267 +
268 +static int fall_back_to_classic_diff(struct histindex *index,
269 + int line1, int count1, int line2, int count2)
270 +{
271 + xpparam_t xpp;
272 + xpp.flags = index->xpp->flags & ~XDF_HISTOGRAM_DIFF;
273 +
274 + return xdl_fall_back_diff(index->env, &xpp,
275 + line1, count1, line2, count2);
276 +}
277 +
278 +static int histogram_diff(xpparam_t const *xpp, xdfenv_t *env,
279 + int line1, int count1, int line2, int count2)
280 +{
281 + struct histindex index;
282 + struct region lcs;
283 + int sz;
284 + int result = -1;
285 +
286 + if (count1 <= 0 && count2 <= 0)
287 + return 0;
288 +
289 + if (LINE_END(1) >= MAX_PTR)
290 + return -1;
291 +
292 + if (!count1) {
293 + while(count2--)
294 + env->xdf2.rchg[line2++ - 1] = 1;
295 + return 0;
296 + } else if (!count2) {
297 + while(count1--)
298 + env->xdf1.rchg[line1++ - 1] = 1;
299 + return 0;
300 + }
301 +
302 + memset(&index, 0, sizeof(index));
303 +
304 + index.env = env;
305 + index.xpp = xpp;
306 +
307 + index.records = NULL;
308 + index.line_map = NULL;
309 + /* in case of early xdl_cha_free() */
310 + index.rcha.head = NULL;
311 +
312 + index.table_bits = xdl_hashbits(count1);
313 + sz = index.records_size = 1 << index.table_bits;
314 + sz *= sizeof(struct record *);
315 + if (!(index.records = (struct record **) xdl_malloc(sz)))
316 + goto cleanup;
317 + memset(index.records, 0, sz);
318 +
319 + sz = index.line_map_size = count1;
320 + sz *= sizeof(struct record *);
321 + if (!(index.line_map = (struct record **) xdl_malloc(sz)))
322 + goto cleanup;
323 + memset(index.line_map, 0, sz);
324 +
325 + sz = index.line_map_size;
326 + sz *= sizeof(unsigned int);
327 + if (!(index.next_ptrs = (unsigned int *) xdl_malloc(sz)))
328 + goto cleanup;
329 + memset(index.next_ptrs, 0, sz);
330 +
331 + /* lines / 4 + 1 comes from xprepare.c:xdl_prepare_ctx() */
332 + if (xdl_cha_init(&index.rcha, sizeof(struct record), count1 / 4 + 1) < 0)
333 + goto cleanup;
334 +
335 + index.ptr_shift = line1;
336 + index.max_chain_length = 64;
337 +
338 + memset(&lcs, 0, sizeof(lcs));
339 + if (find_lcs(&index, &lcs, line1, count1, line2, count2))
340 + result = fall_back_to_classic_diff(&index, line1, count1, line2, count2);
341 + else {
342 + result = 0;
343 + if (lcs.begin1 == 0 && lcs.begin2 == 0) {
344 + int ptr;
345 + for (ptr = 0; ptr < count1; ptr++)
346 + env->xdf1.rchg[line1 + ptr - 1] = 1;
347 + for (ptr = 0; ptr < count2; ptr++)
348 + env->xdf2.rchg[line2 + ptr - 1] = 1;
349 + } else {
350 + result = histogram_diff(xpp, env,
351 + line1, lcs.begin1 - line1,
352 + line2, lcs.begin2 - line2);
353 + result = histogram_diff(xpp, env,
354 + lcs.end1 + 1, LINE_END(1) - lcs.end1,
355 + lcs.end2 + 1, LINE_END(2) - lcs.end2);
356 + result *= -1;
357 + }
358 + }
359 +
360 +cleanup:
361 + xdl_free(index.records);
362 + xdl_free(index.line_map);
363 + xdl_free(index.next_ptrs);
364 + xdl_cha_free(&index.rcha);
365 +
366 + return result;
367 +}
368 +
369 +int xdl_do_histogram_diff(mmfile_t *file1, mmfile_t *file2,
370 + xpparam_t const *xpp, xdfenv_t *env)
371 +{
372 + int line1, line2, count1, count2;
373 +
374 + if (xdl_prepare_env(file1, file2, xpp, env) < 0)
375 + return -1;
376 +
377 + line1 = line2 = 1;
378 + count1 = env->xdf1.nrec;
379 + count2 = env->xdf2.nrec;
380 +
381 + reduce_common_start_end(xpp, env, &line1, &count1, &line2, &count2);
382 +
383 + return histogram_diff(xpp, env, line1, count1, line2, count2);
384 +}