1 /* $OpenBSD: diffreg.c,v 1.93 2019/06/28 13:35:00 deraadt Exp $ */
4 * SPDX-License-Identifier: BSD-4-Clause
6 * Copyright (C) Caldera International Inc. 2001-2002.
9 * Redistribution and use in source and binary forms, with or without
10 * modification, are permitted provided that the following conditions
12 * 1. Redistributions of source code and documentation must retain the above
13 * copyright notice, this list of conditions and the following disclaimer.
14 * 2. Redistributions in binary form must reproduce the above copyright
15 * notice, this list of conditions and the following disclaimer in the
16 * documentation and/or other materials provided with the distribution.
17 * 3. All advertising materials mentioning features or use of this software
18 * must display the following acknowledgement:
19 * This product includes software developed or owned by Caldera
21 * 4. Neither the name of Caldera International, Inc. nor the names of other
22 * contributors may be used to endorse or promote products derived from
23 * this software without specific prior written permission.
25 * USE OF THE SOFTWARE PROVIDED FOR UNDER THIS LICENSE BY CALDERA
26 * INTERNATIONAL, INC. AND CONTRIBUTORS ``AS IS'' AND ANY EXPRESS OR
27 * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
28 * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
29 * IN NO EVENT SHALL CALDERA INTERNATIONAL, INC. BE LIABLE FOR ANY DIRECT,
30 * INDIRECT INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
31 * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
32 * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
33 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT,
34 * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING
35 * IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
36 * POSSIBILITY OF SUCH DAMAGE.
39 * Copyright (c) 1991, 1993
40 * The Regents of the University of California. All rights reserved.
42 * Redistribution and use in source and binary forms, with or without
43 * modification, are permitted provided that the following conditions
45 * 1. Redistributions of source code must retain the above copyright
46 * notice, this list of conditions and the following disclaimer.
47 * 2. Redistributions in binary form must reproduce the above copyright
48 * notice, this list of conditions and the following disclaimer in the
49 * documentation and/or other materials provided with the distribution.
50 * 3. Neither the name of the University nor the names of its contributors
51 * may be used to endorse or promote products derived from this software
52 * without specific prior written permission.
54 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
55 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
56 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
57 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
58 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
59 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
60 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
61 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
62 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
63 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
67 #include <sys/capsicum.h>
70 #include <capsicum_helpers.h>
90 * diff - compare two files.
94 * Uses an algorithm due to Harold Stone, which finds a pair of longest
95 * identical subsequences in the two files.
97 * The major goal is to generate the match vector J. J[i] is the index of
98 * the line in file1 corresponding to line i file0. J[i] = 0 if there is no
101 * Lines are hashed so as to work in core. All potential matches are
102 * located by sorting the lines of each file on the hash (called
103 * ``value''). In particular, this collects the equivalence classes in
104 * file1 together. Subroutine equiv replaces the value of each line in
105 * file0 by the index of the first element of its matching equivalence in
106 * (the reordered) file1. To save space equiv squeezes file1 into a single
107 * array member in which the equivalence classes are simply concatenated,
108 * except that their first members are flagged by changing sign.
110 * Next the indices that point into member are unsorted into array class
111 * according to the original order of file0.
113 * The cleverness lies in routine stone. This marches through the lines of
114 * file0, developing a vector klist of "k-candidates". At step i
115 * a k-candidate is a matched pair of lines x,y (x in file0 y in file1)
116 * such that there is a common subsequence of length k between the first
117 * i lines of file0 and the first y lines of file1, but there is no such
118 * subsequence for any smaller y. x is the earliest possible mate to y that
119 * occurs in such a subsequence.
121 * Whenever any of the members of the equivalence class of lines in file1
122 * matable to a line in file0 has serial number less than the y of some
123 * k-candidate, that k-candidate with the smallest such y is replaced. The
124 * new k-candidate is chained (via pred) to the current k-1 candidate so
125 * that the actual subsequence can be recovered. When a member has serial
126 * number greater that the y of all k-candidates, the klist is extended. At
127 * the end, the longest subsequence is pulled out and placed in the array J
130 * With J in hand, the matches there recorded are check'ed against reality
131 * to assure that no spurious matches have crept in due to hashing. If they
132 * have, they are broken, and "jackpot" is recorded -- a harmless matter
133 * except that a true match for a spuriously mated line may now be
134 * unnecessarily reported as a change.
136 * Much of the complexity of the program comes simply from trying to
137 * minimize core utilization and maximize the range of doable problems by
138 * dynamically allocating what is needed and reusing what is not. The core
139 * requirements for problems larger than somewhat are (in words)
140 * 2*length(file0) + length(file1) + 3*(number of k-candidates installed),
141 * typically about 6n words for files of length n.
156 * The following struct is used to record change information when
157 * doing a "context" or "unified" diff. (see routine "change" to
158 * understand the highly mnemonic field names)
161 int a; /* start line in old file */
162 int b; /* end line in old file */
163 int c; /* start line in new file */
164 int d; /* end line in new file */
167 enum readhash { RH_BINARY, RH_OK, RH_EOF };
170 static FILE *opentemp(const char *);
171 static void output(char *, FILE *, char *, FILE *, int);
172 static void check(FILE *, FILE *, int);
173 static void range(int, int, const char *);
174 static void uni_range(int, int);
175 static void dump_context_vec(FILE *, FILE *, int);
176 static void dump_unified_vec(FILE *, FILE *, int);
177 static bool prepare(int, FILE *, size_t, int);
178 static void prune(void);
179 static void equiv(struct line *, int, struct line *, int, int *);
180 static void unravel(int);
181 static void unsort(struct line *, int, int *);
182 static void change(char *, FILE *, char *, FILE *, int, int, int, int, int *);
183 static void sort(struct line *, int);
184 static void print_header(const char *, const char *);
185 static void print_space(int, int, int);
186 static bool ignoreline_pattern(char *);
187 static bool ignoreline(char *, bool);
188 static int asciifile(FILE *);
189 static int fetch(long *, int, int, FILE *, int, int, int);
190 static int newcand(int, int, int);
191 static int search(int *, int, int);
192 static int skipline(FILE *);
193 static int stone(int *, int, int *, int *, int);
194 static enum readhash readhash(FILE *, int, unsigned *);
195 static int files_differ(FILE *, FILE *, int);
196 static char *match_function(const long *, int, FILE *);
197 static char *preadline(int, size_t, off_t);
199 static int *J; /* will be overlaid on class */
200 static int *class; /* will be overlaid on file[0] */
201 static int *klist; /* will be overlaid on file[0] after class */
202 static int *member; /* will be overlaid on file[1] */
204 static int inifdef; /* whether or not we are in a #ifdef block */
206 static int pref, suff; /* length of prefix and suffix */
208 static int anychange;
209 static int hw, padding; /* half width and padding */
211 static long *ixnew; /* will be overlaid on file[1] */
212 static long *ixold; /* will be overlaid on klist */
213 static struct cand *clist; /* merely a free storage pot for candidates */
214 static int clistlen; /* the length of clist */
215 static struct line *sfile[2]; /* shortened by pruning common prefix/suffix */
216 static int (*chrtran)(int); /* translation table for case-folding */
217 static struct context_vec *context_vec_start;
218 static struct context_vec *context_vec_end;
219 static struct context_vec *context_vec_ptr;
221 #define FUNCTION_CONTEXT_SIZE 55
222 static char lastbuf[FUNCTION_CONTEXT_SIZE];
224 static int lastmatchline;
241 diffreg(char *file1, char *file2, int flags, int capsicum)
245 struct pr *pr = NULL;
246 cap_rights_t rights_ro;
255 * hw excludes padding and make sure when -t is not used,
256 * the second column always starts from the closest tab stop
258 if (diff_format == D_SIDEBYSIDE) {
260 padding = tabsize - (hw % tabsize);
261 if ((flags & D_EXPANDTABS) != 0 || (padding % tabsize == 0))
265 ((padding == MIN_PAD) ? (padding << 1) : padding) - 1;
269 if (flags & D_IGNORECASE)
273 if (S_ISDIR(stb1.st_mode) != S_ISDIR(stb2.st_mode))
274 return (S_ISDIR(stb1.st_mode) ? D_MISMATCH1 : D_MISMATCH2);
275 if (strcmp(file1, "-") == 0 && strcmp(file2, "-") == 0)
278 if (flags & D_EMPTY1)
279 f1 = fopen(_PATH_DEVNULL, "r");
281 if (!S_ISREG(stb1.st_mode)) {
282 if ((f1 = opentemp(file1)) == NULL ||
283 fstat(fileno(f1), &stb1) == -1) {
289 } else if (strcmp(file1, "-") == 0)
292 f1 = fopen(file1, "r");
301 if (flags & D_EMPTY2)
302 f2 = fopen(_PATH_DEVNULL, "r");
304 if (!S_ISREG(stb2.st_mode)) {
305 if ((f2 = opentemp(file2)) == NULL ||
306 fstat(fileno(f2), &stb2) == -1) {
312 } else if (strcmp(file2, "-") == 0)
315 f2 = fopen(file2, "r");
325 pr = start_pr(file1, file2);
328 cap_rights_init(&rights_ro, CAP_READ, CAP_FSTAT, CAP_SEEK);
329 if (caph_rights_limit(fileno(f1), &rights_ro) < 0)
330 err(2, "unable to limit rights on: %s", file1);
331 if (caph_rights_limit(fileno(f2), &rights_ro) < 0)
332 err(2, "unable to limit rights on: %s", file2);
333 if (fileno(f1) == STDIN_FILENO || fileno(f2) == STDIN_FILENO) {
334 /* stdin has already been limited */
335 if (caph_limit_stderr() == -1)
336 err(2, "unable to limit stderr");
337 if (caph_limit_stdout() == -1)
338 err(2, "unable to limit stdout");
339 } else if (caph_limit_stdio() == -1)
340 err(2, "unable to limit stdio");
342 caph_cache_catpages();
344 if (caph_enter() < 0)
345 err(2, "unable to enter capability mode");
348 switch (files_differ(f1, f2, flags)) {
360 if (diff_format == D_BRIEF && ignore_pats == NULL &&
361 (flags & (D_FOLDBLANKS|D_IGNOREBLANKS|D_IGNORECASE|D_STRIPCR)) == 0)
367 if ((flags & D_FORCEASCII) != 0) {
368 (void)prepare(0, f1, stb1.st_size, flags);
369 (void)prepare(1, f2, stb2.st_size, flags);
370 } else if (!asciifile(f1) || !asciifile(f2) ||
371 !prepare(0, f1, stb1.st_size, flags) ||
372 !prepare(1, f2, stb2.st_size, flags)) {
379 sort(sfile[0], slen[0]);
380 sort(sfile[1], slen[1]);
382 member = (int *)file[1];
383 equiv(sfile[0], slen[0], sfile[1], slen[1], member);
384 member = xreallocarray(member, slen[1] + 2, sizeof(*member));
386 class = (int *)file[0];
387 unsort(sfile[0], slen[0], class);
388 class = xreallocarray(class, slen[0] + 2, sizeof(*class));
390 klist = xcalloc(slen[0] + 2, sizeof(*klist));
393 clist = xcalloc(clistlen, sizeof(*clist));
394 i = stone(class, slen[0], member, klist, flags);
398 J = xreallocarray(J, len[0] + 2, sizeof(*J));
403 ixold = xreallocarray(ixold, len[0] + 2, sizeof(*ixold));
404 ixnew = xreallocarray(ixnew, len[1] + 2, sizeof(*ixnew));
405 check(f1, f2, flags);
406 output(file1, f1, file2, f2, flags);
425 * Check to see if the given files differ.
426 * Returns 0 if they are the same, 1 if different, and -1 on error.
427 * XXX - could use code from cmp(1) [faster]
430 files_differ(FILE *f1, FILE *f2, int flags)
432 char buf1[BUFSIZ], buf2[BUFSIZ];
435 if ((flags & (D_EMPTY1|D_EMPTY2)) || stb1.st_size != stb2.st_size ||
436 (stb1.st_mode & S_IFMT) != (stb2.st_mode & S_IFMT))
439 if (stb1.st_dev == stb2.st_dev && stb1.st_ino == stb2.st_ino)
443 i = fread(buf1, 1, sizeof(buf1), f1);
444 j = fread(buf2, 1, sizeof(buf2), f2);
445 if ((!i && ferror(f1)) || (!j && ferror(f2)))
451 if (memcmp(buf1, buf2, i) != 0)
457 opentemp(const char *f)
459 char buf[BUFSIZ], tempfile[PATH_MAX];
463 if (strcmp(f, "-") == 0)
465 else if ((ifd = open(f, O_RDONLY, 0644)) == -1)
468 (void)strlcpy(tempfile, _PATH_TMP "/diff.XXXXXXXX", sizeof(tempfile));
470 if ((ofd = mkstemp(tempfile)) == -1) {
475 while ((nread = read(ifd, buf, BUFSIZ)) > 0) {
476 if (write(ofd, buf, nread) != nread) {
483 lseek(ofd, (off_t)0, SEEK_SET);
484 return (fdopen(ofd, "r"));
488 prepare(int i, FILE *fd, size_t filesize, int flags)
497 sz = MIN(filesize, SIZE_MAX) / 25;
501 p = xcalloc(sz + 3, sizeof(*p));
502 while ((r = readhash(fd, flags, &h)) != RH_EOF)
504 case RH_EOF: /* otherwise clang complains */
510 p = xreallocarray(p, sz + 3, sizeof(*p));
526 for (pref = 0; pref < len[0] && pref < len[1] &&
527 file[0][pref + 1].value == file[1][pref + 1].value;
530 for (suff = 0; suff < len[0] - pref && suff < len[1] - pref &&
531 file[0][len[0] - suff].value == file[1][len[1] - suff].value;
534 for (j = 0; j < 2; j++) {
535 sfile[j] = file[j] + pref;
536 slen[j] = len[j] - pref - suff;
537 for (i = 0; i <= slen[j]; i++)
538 sfile[j][i].serial = i;
543 equiv(struct line *a, int n, struct line *b, int m, int *c)
548 while (i <= n && j <= m) {
549 if (a[i].value < b[j].value)
551 else if (a[i].value == b[j].value)
562 while (b[j + 1].value == b[j].value) {
571 stone(int *a, int n, int *b, int *c, int flags)
574 int oldc, tc, oldl, sq;
575 unsigned numtries, bound;
577 if (flags & D_MINIMAL)
581 bound = MAX(256, sq);
585 c[0] = newcand(0, 0, 0);
586 for (i = 1; i <= n; i++) {
595 if (y <= clist[oldc].y)
601 if (clist[c[l]].y <= y)
604 c[l] = newcand(i, y, oldc);
609 c[l] = newcand(i, y, oldc);
613 } while ((y = b[++j]) > 0 && numtries < bound);
619 newcand(int x, int y, int pred)
623 if (clen == clistlen) {
624 clistlen = clistlen * 11 / 10;
625 clist = xreallocarray(clist, clistlen, sizeof(*clist));
635 search(int *c, int k, int y)
639 if (clist[c[k]].y < y) /* quick look for typical case */
664 for (i = 0; i <= len[0]; i++)
665 J[i] = i <= pref ? i :
666 i > len[0] - suff ? i + len[1] - len[0] : 0;
667 for (q = clist + p; q->y != 0; q = clist + q->pred)
668 J[q->x + pref] = q->y + pref;
672 * Check does double duty:
673 * 1. ferret out any fortuitous correspondences due to confounding by
674 * hashing (which result in "jackpot")
675 * 2. collect random access indexes to the two files
678 check(FILE *f1, FILE *f2, int flags)
680 int i, j, /* jackpot, */ c, d;
686 ixold[0] = ixnew[0] = 0;
689 for (i = 1; i <= len[0]; i++) {
691 ixold[i] = ctold += skipline(f1);
695 ixnew[j] = ctnew += skipline(f2);
698 if (flags & (D_FOLDBLANKS | D_IGNOREBLANKS | D_IGNORECASE | D_STRIPCR)) {
703 * GNU diff ignores a missing newline
704 * in one file for -b or -w.
706 if (flags & (D_FOLDBLANKS | D_IGNOREBLANKS)) {
707 if (c == EOF && d == '\n') {
710 } else if (c == '\n' && d == EOF) {
717 if (flags & D_STRIPCR && (c == '\r' || d == '\r')) {
719 if ((c = getc(f1)) == '\n') {
726 if ((d = getc(f2)) == '\n') {
734 if ((flags & D_FOLDBLANKS) && isspace(c) &&
740 } while (isspace(c = getc(f1)));
745 } while (isspace(d = getc(f2)));
746 } else if (flags & D_IGNOREBLANKS) {
747 while (isspace(c) && c != '\n') {
751 while (isspace(d) && d != '\n') {
756 if (chrtran(c) != chrtran(d)) {
759 if (c != '\n' && c != EOF)
760 ctold += skipline(f1);
761 if (d != '\n' && c != EOF)
762 ctnew += skipline(f2);
765 if (c == '\n' || c == EOF)
772 if ((c = getc(f1)) != (d = getc(f2))) {
775 if (c != '\n' && c != EOF)
776 ctold += skipline(f1);
777 if (d != '\n' && c != EOF)
778 ctnew += skipline(f2);
781 if (c == '\n' || c == EOF)
789 for (; j <= len[1]; j++) {
790 ixnew[j] = ctnew += skipline(f2);
794 * fprintf(stderr, "jackpot\n");
798 /* shellsort CACM #201 */
800 sort(struct line *a, int n)
802 struct line *ai, *aim, w;
807 for (j = 1; j <= n; j *= 2)
809 for (m /= 2; m != 0; m /= 2) {
811 for (j = 1; j <= k; j++) {
812 for (ai = &a[j]; ai > a; ai -= m) {
815 break; /* wraparound */
816 if (aim->value > ai[0].value ||
817 (aim->value == ai[0].value &&
818 aim->serial > ai[0].serial))
820 w.value = ai[0].value;
821 ai[0].value = aim->value;
822 aim->value = w.value;
823 w.serial = ai[0].serial;
824 ai[0].serial = aim->serial;
825 aim->serial = w.serial;
832 unsort(struct line *f, int l, int *b)
836 a = xcalloc(l + 1, sizeof(*a));
837 for (i = 1; i <= l; i++)
838 a[f[i].serial] = f[i].value;
839 for (i = 1; i <= l; i++)
849 for (i = 1; (c = getc(f)) != '\n' && c != EOF; i++)
855 output(char *file1, FILE *f1, char *file2, FILE *f2, int flags)
857 int i, j, m, i0, i1, j0, j1, nc;
863 J[m + 1] = len[1] + 1;
864 if (diff_format != D_EDIT) {
865 for (i0 = 1; i0 <= m; i0 = i1 + 1) {
866 while (i0 <= m && J[i0] == J[i0 - 1] + 1) {
867 if (diff_format == D_SIDEBYSIDE && suppress_common != 1) {
868 nc = fetch(ixold, i0, i0, f1, '\0', 1, flags);
869 print_space(nc, (hw - nc) + (padding << 1) + 1, flags);
870 fetch(ixnew, J[i0], J[i0], f2, '\0', 0, flags);
877 while (i1 < m && J[i1 + 1] == 0)
883 * When using side-by-side, lines from both of the files are
884 * printed. The algorithm used by diff(1) identifies the ranges
885 * in which two files differ.
886 * See the change() function below.
887 * The for loop below consumes the shorter range, whereas one of
888 * the while loops deals with the longer one.
890 if (diff_format == D_SIDEBYSIDE) {
891 for (i = i0, j = j0; i <= i1 && j <= j1; i++, j++)
892 change(file1, f1, file2, f2, i, i, j, j, &flags);
895 change(file1, f1, file2, f2, i, i, j + 1, j, &flags);
900 change(file1, f1, file2, f2, i + 1, i, j, j, &flags);
904 change(file1, f1, file2, f2, i0, i1, j0, j1, &flags);
907 for (i0 = m; i0 >= 1; i0 = i1 - 1) {
908 while (i0 >= 1 && J[i0] == J[i0 + 1] - 1 && J[i0] != 0)
912 while (i1 > 1 && J[i1 - 1] == 0)
916 change(file1, f1, file2, f2, i1, i0, j1, j0, &flags);
920 change(file1, f1, file2, f2, 1, 0, 1, len[1], &flags);
921 if (diff_format == D_IFDEF || diff_format == D_GFORMAT) {
924 if ((c = getc(f1)) == EOF)
930 if (anychange != 0) {
931 if (diff_format == D_CONTEXT)
932 dump_context_vec(f1, f2, flags);
933 else if (diff_format == D_UNIFIED)
934 dump_unified_vec(f1, f2, flags);
939 range(int a, int b, const char *separator)
941 printf("%d", a > b ? b : a);
943 printf("%s%d", separator, b);
947 uni_range(int a, int b)
950 printf("%d,%d", a, b - a + 1);
958 preadline(int fd, size_t rlen, off_t off)
963 line = xmalloc(rlen + 1);
964 if ((nr = pread(fd, line, rlen, off)) == -1)
966 if (nr > 0 && line[nr-1] == '\n')
973 ignoreline_pattern(char *line)
977 ret = regexec(&ignore_re, line, 0, NULL, 0);
978 return (ret == 0); /* if it matched, it should be ignored. */
982 ignoreline(char *line, bool skip_blanks)
985 if (skip_blanks && *line == '\0')
987 if (ignore_pats != NULL && ignoreline_pattern(line))
993 * Indicate that there is a difference between lines a and b of the from file
994 * to get to lines c to d of the to file. If a is greater then b then there
995 * are no lines in the from file involved and this means that there were
996 * lines appended (beginning at b). If c is greater than d then there are
997 * lines missing from the to file.
1000 change(char *file1, FILE *f1, char *file2, FILE *f2, int a, int b, int c, int d,
1003 static size_t max_context = 64;
1007 bool skip_blanks, ignore;
1009 skip_blanks = (*pflags & D_SKIPBLANKLINES);
1011 if ((diff_format != D_IFDEF || diff_format == D_GFORMAT) &&
1014 if (ignore_pats != NULL || skip_blanks) {
1017 * All lines in the change, insert, or delete must match an ignore
1018 * pattern for the change to be ignored.
1020 if (a <= b) { /* Changes and deletes. */
1021 for (i = a; i <= b; i++) {
1022 line = preadline(fileno(f1),
1023 ixold[i] - ixold[i - 1], ixold[i - 1]);
1024 ignore = ignoreline(line, skip_blanks);
1030 if (a > b || c <= d) { /* Changes and inserts. */
1031 for (i = c; i <= d; i++) {
1032 line = preadline(fileno(f2),
1033 ixnew[i] - ixnew[i - 1], ixnew[i - 1]);
1034 ignore = ignoreline(line, skip_blanks);
1043 if (*pflags & D_HEADER && diff_format != D_BRIEF) {
1044 printf("%s %s %s\n", diffargs, file1, file2);
1045 *pflags &= ~D_HEADER;
1047 if (diff_format == D_CONTEXT || diff_format == D_UNIFIED) {
1049 * Allocate change records as needed.
1051 if (context_vec_start == NULL ||
1052 context_vec_ptr == context_vec_end - 1) {
1053 ptrdiff_t offset = -1;
1055 if (context_vec_start != NULL)
1056 offset = context_vec_ptr - context_vec_start;
1058 context_vec_start = xreallocarray(context_vec_start,
1059 max_context, sizeof(*context_vec_start));
1060 context_vec_end = context_vec_start + max_context;
1061 context_vec_ptr = context_vec_start + offset;
1063 if (anychange == 0) {
1065 * Print the context/unidiff header first time through.
1067 print_header(file1, file2);
1069 } else if (a > context_vec_ptr->b + (2 * diff_context) + 1 &&
1070 c > context_vec_ptr->d + (2 * diff_context) + 1) {
1072 * If this change is more than 'diff_context' lines from the
1073 * previous change, dump the record and reset it.
1075 if (diff_format == D_CONTEXT)
1076 dump_context_vec(f1, f2, *pflags);
1078 dump_unified_vec(f1, f2, *pflags);
1081 context_vec_ptr->a = a;
1082 context_vec_ptr->b = b;
1083 context_vec_ptr->c = c;
1084 context_vec_ptr->d = d;
1089 switch (diff_format) {
1095 printf("%c", a > b ? 'a' : c > d ? 'd' : 'c');
1096 if (diff_format == D_NORMAL)
1101 printf("%c", a > b ? 'a' : c > d ? 'd' : 'c');
1107 printf("a%d %d\n", b, d - c + 1);
1109 printf("d%d %d\n", a, b - a + 1);
1111 /* add changed lines */
1112 printf("a%d %d\n", b, d - c + 1);
1116 if (diff_format == D_GFORMAT) {
1118 /* print through if append (a>b), else to (nb: 0 vs 1 orig) */
1119 nc = ixold[a > b ? b : a - 1] - curpos;
1120 for (i = 0; i < nc; i++)
1121 printf("%c", getc(f1));
1122 for (walk = group_format; *walk != '\0'; walk++) {
1127 fetch(ixold, a, b, f1, '<', 1, *pflags);
1130 fetch(ixnew, c, d, f2, '>', 0, *pflags);
1133 printf("%%%c", *walk);
1138 printf("%c", *walk);
1141 if (diff_format == D_SIDEBYSIDE) {
1143 printf("\033[%sm", add_code);
1144 else if (color && c > d)
1145 printf("\033[%sm", del_code);
1147 print_space(0, hw + padding , *pflags);
1149 nc = fetch(ixold, a, b, f1, '\0', 1, *pflags);
1150 print_space(nc, hw - nc + padding, *pflags);
1153 printf("\033[%sm", add_code);
1154 else if (color && c > d)
1155 printf("\033[%sm", del_code);
1156 printf("%c", (a > b) ? '>' : ((c > d) ? '<' : '|'));
1159 print_space(hw + padding + 1 , padding, *pflags);
1160 fetch(ixnew, c, d, f2, '\0', 0, *pflags);
1163 if (diff_format == D_NORMAL || diff_format == D_IFDEF) {
1164 fetch(ixold, a, b, f1, '<', 1, *pflags);
1165 if (a <= b && c <= d && diff_format == D_NORMAL)
1168 if (diff_format != D_GFORMAT && diff_format != D_SIDEBYSIDE)
1169 fetch(ixnew, c, d, f2, diff_format == D_NORMAL ? '>' : '\0', 0, *pflags);
1170 if (edoffset != 0 && diff_format == D_EDIT) {
1172 * A non-zero edoffset value for D_EDIT indicates that the last line
1173 * printed was a bare dot (".") that has been escaped as ".." to
1174 * prevent ed(1) from misinterpreting it. We have to add a
1175 * substitute command to change this back and restart where we left
1179 printf("%ds/.//\n", a + edoffset - 1);
1180 b = a + edoffset - 1;
1185 if ((diff_format == D_EDIT || diff_format == D_REVERSE) && c <= d)
1188 printf("#endif /* %s */\n", ifdefname);
1194 fetch(long *f, int a, int b, FILE *lb, int ch, int oldfile, int flags)
1196 int i, j, c, lastc, col, nc, newcol;
1201 * When doing #ifdef's, copy down to current line
1202 * if this is the first file, so that stuff makes it to output.
1204 if ((diff_format == D_IFDEF) && oldfile) {
1205 long curpos = ftell(lb);
1206 /* print through if append (a>b), else to (nb: 0 vs 1 orig) */
1207 nc = f[a > b ? b : a - 1] - curpos;
1208 for (i = 0; i < nc; i++)
1209 printf("%c", getc(lb));
1213 if (diff_format == D_IFDEF) {
1215 printf("#else /* %s%s */\n",
1216 oldfile == 1 ? "!" : "", ifdefname);
1219 printf("#ifndef %s\n", ifdefname);
1221 printf("#ifdef %s\n", ifdefname);
1223 inifdef = 1 + oldfile;
1225 for (i = a; i <= b; i++) {
1226 fseek(lb, f[i - 1], SEEK_SET);
1227 nc = f[i] - f[i - 1];
1228 if (diff_format == D_SIDEBYSIDE && hw < nc)
1230 if (diff_format != D_IFDEF && diff_format != D_GFORMAT &&
1232 if (color && (ch == '>' || ch == '+'))
1233 printf("\033[%sm", add_code);
1234 else if (color && (ch == '<' || ch == '-'))
1235 printf("\033[%sm", del_code);
1237 if (Tflag && (diff_format == D_NORMAL ||
1238 diff_format == D_CONTEXT ||
1239 diff_format == D_UNIFIED))
1241 else if (diff_format != D_UNIFIED)
1246 while (j < nc && (hw == 0 || col < hw)) {
1248 if (flags & D_STRIPCR && c == '\r') {
1249 if ((c = getc(lb)) == '\n')
1257 if (diff_format == D_EDIT ||
1258 diff_format == D_REVERSE ||
1259 diff_format == D_NREVERSE)
1260 warnx("No newline at end of file");
1262 printf("\n\\ No newline at end of file\n");
1266 * when using --side-by-side, col needs to be increased
1267 * in any case to keep the columns aligned
1270 if (flags & D_EXPANDTABS) {
1271 newcol = ((col / tabsize) + 1) * tabsize;
1274 } while (++col < newcol && col < hw);
1276 if (diff_format == D_SIDEBYSIDE) {
1277 if ((col + tabsize) > hw) {
1278 printf("%*s", hw - col, "");
1290 if (diff_format == D_EDIT && j == 1 && c == '\n' &&
1293 * Don't print a bare "." line since that will confuse
1294 * ed(1). Print ".." instead and set the, global variable
1295 * edoffset to an offset from which to restart. The
1296 * caller must check the value of edoffset
1299 edoffset = i - a + 1;
1302 /* when side-by-side, do not print a newline */
1303 if (diff_format != D_SIDEBYSIDE || c != '\n') {
1304 if (color && c == '\n')
1305 printf("\033[m%c", c);
1316 if (color && diff_format == D_SIDEBYSIDE)
1322 * Hash function taken from Robert Sedgewick, Algorithms in C, 3d ed., p 578.
1324 static enum readhash
1325 readhash(FILE *f, int flags, unsigned *hash)
1333 switch (t = getc(f)) {
1335 if ((flags & D_FORCEASCII) == 0)
1339 if (flags & D_STRIPCR) {
1350 if ((flags & (D_FOLDBLANKS|D_IGNOREBLANKS)) != 0) {
1357 if (space && (flags & D_IGNOREBLANKS) == 0) {
1361 sum = sum * 127 + chrtran(t);
1380 unsigned char buf[BUFSIZ];
1387 cnt = fread(buf, 1, sizeof(buf), f);
1388 return (memchr(buf, '\0', cnt) == NULL);
1391 #define begins_with(s, pre) (strncmp(s, pre, sizeof(pre) - 1) == 0)
1394 match_function(const long *f, int pos, FILE *fp)
1396 unsigned char buf[FUNCTION_CONTEXT_SIZE];
1398 int last = lastline;
1399 const char *state = NULL;
1402 for (; pos > last; pos--) {
1403 fseek(fp, f[pos - 1], SEEK_SET);
1404 nc = f[pos] - f[pos - 1];
1405 if (nc >= sizeof(buf))
1406 nc = sizeof(buf) - 1;
1407 nc = fread(buf, 1, nc, fp);
1411 buf[strcspn(buf, "\n")] = '\0';
1412 if (most_recent_pat != NULL) {
1413 int ret = regexec(&most_recent_re, buf, 0, NULL, 0);
1417 strlcpy(lastbuf, buf, sizeof(lastbuf));
1418 lastmatchline = pos;
1420 } else if (isalpha(buf[0]) || buf[0] == '_' || buf[0] == '$'
1421 || buf[0] == '-' || buf[0] == '+') {
1422 if (begins_with(buf, "private:")) {
1424 state = " (private)";
1425 } else if (begins_with(buf, "protected:")) {
1427 state = " (protected)";
1428 } else if (begins_with(buf, "public:")) {
1430 state = " (public)";
1432 strlcpy(lastbuf, buf, sizeof(lastbuf));
1434 strlcat(lastbuf, state, sizeof(lastbuf));
1435 lastmatchline = pos;
1440 return (lastmatchline > 0 ? lastbuf : NULL);
1443 /* dump accumulated "context" diff changes */
1445 dump_context_vec(FILE *f1, FILE *f2, int flags)
1447 struct context_vec *cvp = context_vec_start;
1448 int lowa, upb, lowc, upd, do_output;
1452 if (context_vec_start > context_vec_ptr)
1455 b = d = 0; /* gcc */
1456 lowa = MAX(1, cvp->a - diff_context);
1457 upb = MIN(len[0], context_vec_ptr->b + diff_context);
1458 lowc = MAX(1, cvp->c - diff_context);
1459 upd = MIN(len[1], context_vec_ptr->d + diff_context);
1461 printf("***************");
1462 if (flags & (D_PROTOTYPE | D_MATCHLAST)) {
1463 f = match_function(ixold, cvp->a - 1, f1);
1468 range(lowa, upb, ",");
1472 * Output changes to the "old" file. The first loop suppresses
1473 * output if there were no changes to the "old" file (we'll see
1474 * the "old" lines as context in the "new" list).
1477 for (; cvp <= context_vec_ptr; cvp++)
1478 if (cvp->a <= cvp->b) {
1479 cvp = context_vec_start;
1484 while (cvp <= context_vec_ptr) {
1490 if (a <= b && c <= d)
1493 ch = (a <= b) ? 'd' : 'a';
1496 fetch(ixold, lowa, b, f1, ' ', 0, flags);
1498 fetch(ixold, lowa, a - 1, f1, ' ', 0, flags);
1499 fetch(ixold, a, b, f1,
1500 ch == 'c' ? '!' : '-', 0, flags);
1505 fetch(ixold, b + 1, upb, f1, ' ', 0, flags);
1507 /* output changes to the "new" file */
1509 range(lowc, upd, ",");
1513 for (cvp = context_vec_start; cvp <= context_vec_ptr; cvp++)
1514 if (cvp->c <= cvp->d) {
1515 cvp = context_vec_start;
1520 while (cvp <= context_vec_ptr) {
1526 if (a <= b && c <= d)
1529 ch = (a <= b) ? 'd' : 'a';
1532 fetch(ixnew, lowc, d, f2, ' ', 0, flags);
1534 fetch(ixnew, lowc, c - 1, f2, ' ', 0, flags);
1535 fetch(ixnew, c, d, f2,
1536 ch == 'c' ? '!' : '+', 0, flags);
1541 fetch(ixnew, d + 1, upd, f2, ' ', 0, flags);
1543 context_vec_ptr = context_vec_start - 1;
1546 /* dump accumulated "unified" diff changes */
1548 dump_unified_vec(FILE *f1, FILE *f2, int flags)
1550 struct context_vec *cvp = context_vec_start;
1551 int lowa, upb, lowc, upd;
1555 if (context_vec_start > context_vec_ptr)
1558 b = d = 0; /* gcc */
1559 lowa = MAX(1, cvp->a - diff_context);
1560 upb = MIN(len[0], context_vec_ptr->b + diff_context);
1561 lowc = MAX(1, cvp->c - diff_context);
1562 upd = MIN(len[1], context_vec_ptr->d + diff_context);
1565 uni_range(lowa, upb);
1567 uni_range(lowc, upd);
1569 if (flags & (D_PROTOTYPE | D_MATCHLAST)) {
1570 f = match_function(ixold, cvp->a - 1, f1);
1577 * Output changes in "unified" diff format--the old and new lines
1578 * are printed together.
1580 for (; cvp <= context_vec_ptr; cvp++) {
1587 * c: both new and old changes
1588 * d: only changes in the old file
1589 * a: only changes in the new file
1591 if (a <= b && c <= d)
1594 ch = (a <= b) ? 'd' : 'a';
1598 fetch(ixold, lowa, a - 1, f1, ' ', 0, flags);
1599 fetch(ixold, a, b, f1, '-', 0, flags);
1600 fetch(ixnew, c, d, f2, '+', 0, flags);
1603 fetch(ixold, lowa, a - 1, f1, ' ', 0, flags);
1604 fetch(ixold, a, b, f1, '-', 0, flags);
1607 fetch(ixnew, lowc, c - 1, f2, ' ', 0, flags);
1608 fetch(ixnew, c, d, f2, '+', 0, flags);
1614 fetch(ixnew, d + 1, upd, f2, ' ', 0, flags);
1616 context_vec_ptr = context_vec_start - 1;
1620 print_header(const char *file1, const char *file2)
1622 const char *time_format;
1624 struct tm tm1, tm2, *tm_ptr1, *tm_ptr2;
1625 int nsec1 = stb1.st_mtim.tv_nsec;
1626 int nsec2 = stb2.st_mtim.tv_nsec;
1628 time_format = "%Y-%m-%d %H:%M:%S";
1632 tm_ptr1 = localtime_r(&stb1.st_mtime, &tm1);
1633 tm_ptr2 = localtime_r(&stb2.st_mtime, &tm2);
1634 if (label[0] != NULL)
1635 printf("%s %s\n", diff_format == D_CONTEXT ? "***" : "---",
1638 strftime(buf, sizeof(buf), time_format, tm_ptr1);
1639 printf("%s %s\t%s", diff_format == D_CONTEXT ? "***" : "---",
1642 strftime(buf, sizeof(buf), "%z", tm_ptr1);
1643 printf(".%.9d %s", nsec1, buf);
1647 if (label[1] != NULL)
1648 printf("%s %s\n", diff_format == D_CONTEXT ? "---" : "+++",
1651 strftime(buf, sizeof(buf), time_format, tm_ptr2);
1652 printf("%s %s\t%s", diff_format == D_CONTEXT ? "---" : "+++",
1655 strftime(buf, sizeof(buf), "%z", tm_ptr2);
1656 printf(".%.9d %s", nsec2, buf);
1663 * Prints n number of space characters either by using tab
1664 * or single space characters.
1665 * nc is the preceding number of characters
1668 print_space(int nc, int n, int flags) {
1672 if ((flags & D_EXPANDTABS) == 0) {
1673 /* first tabstop may be closer than tabsize */
1674 i = tabsize - (nc % tabsize);
1675 while (col >= tabsize) {
1681 printf("%*s", col, "");