/* $OpenBSD: diff3.c,v 1.48 2008/03/09 01:52:55 joris Exp $ */ /* * Copyright (C) Caldera International Inc. 2001-2002. * All rights reserved. * * Redistribution and use in source and binary forms, with or without * modification, are permitted provided that the following conditions * are met: * 1. Redistributions of source code and documentation must retain the above * copyright notice, this list of conditions and the following disclaimer. * 2. Redistributions in binary form must reproduce the above copyright * notice, this list of conditions and the following disclaimer in the * documentation and/or other materials provided with the distribution. * 3. All advertising materials mentioning features or use of this software * must display the following acknowledgement: * This product includes software developed or owned by Caldera * International, Inc. * 4. Neither the name of Caldera International, Inc. nor the names of other * contributors may be used to endorse or promote products derived from * this software without specific prior written permission. * * USE OF THE SOFTWARE PROVIDED FOR UNDER THIS LICENSE BY CALDERA * INTERNATIONAL, INC. AND CONTRIBUTORS ``AS IS'' AND ANY EXPRESS OR * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. * IN NO EVENT SHALL CALDERA INTERNATIONAL, INC. BE LIABLE FOR ANY DIRECT, * INDIRECT INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING * IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE * POSSIBILITY OF SUCH DAMAGE. */ /*- * Copyright (c) 1991, 1993 * The Regents of the University of California. All rights reserved. * * Redistribution and use in source and binary forms, with or without * modification, are permitted provided that the following conditions * are met: * 1. Redistributions of source code must retain the above copyright * notice, this list of conditions and the following disclaimer. * 2. Redistributions in binary form must reproduce the above copyright * notice, this list of conditions and the following disclaimer in the * documentation and/or other materials provided with the distribution. * 3. Neither the name of the University nor the names of its contributors * may be used to endorse or promote products derived from this software * without specific prior written permission. * * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF * SUCH DAMAGE. * * @(#)diff3.c 8.1 (Berkeley) 6/6/93 */ #ifndef lint static const char copyright[] = "@(#) Copyright (c) 1991, 1993\n\ The Regents of the University of California. All rights reserved.\n"; #endif /* not lint */ #ifndef lint static const char rcsid[] = "$OpenBSD: diff3.c,v 1.48 2008/03/09 01:52:55 joris Exp $"; #endif /* not lint */ #include #include #include #include #include #include #include #include "atomicio.h" #include "cvs.h" #include "diff.h" /* diff3 - 3-way differential file comparison */ /* diff3 [-ex3EX] d13 d23 f1 f2 f3 [m1 m3] * * d13 = diff report on f1 vs f3 * d23 = diff report on f2 vs f3 * f1, f2, f3 the 3 files * if changes in f1 overlap with changes in f3, m1 and m3 are used * to mark the overlaps; otherwise, the file names f1 and f3 are used * (only for options E and X). */ /* * "from" is first in range of changed lines; "to" is last+1 * from=to=line after point of insertion for added lines. */ struct range { int from; int to; }; struct diff { struct range old; struct range new; }; static size_t szchanges; static struct diff *d13 = NULL; static struct diff *d23 = NULL; /* * "de" is used to gather editing scripts. These are later spewed out in * reverse order. Its first element must be all zero, the "new" component * of "de" contains line positions or byte positions depending on when you * look (!?). Array overlap indicates which sections in "de" correspond to * lines that are different in all three files. */ static struct diff *de = NULL; static char *overlap = NULL; static int overlapcnt = 0; static FILE *fp[3]; static int cline[3]; /* # of the last-read line in each file (0-2) */ /* * the latest known correspondence between line numbers of the 3 files * is stored in last[1-3]; */ static int last[4]; static int eflag; static int oflag; /* indicates whether to mark overlaps (-E or -X)*/ static int debug = 0; static char f1mark[MAXPATHLEN], f3mark[MAXPATHLEN]; /* markers for -E and -X */ static int duplicate(struct range *, struct range *); static int edit(struct diff *, int, int); static char *getchange(FILE *); static char *getline(FILE *, size_t *); static int number(char **); static size_t readin(int, struct diff **); static int skip(int, int, char *); static int edscript(int); static int merge(size_t, size_t); static void change(int, struct range *, int); static void keep(int, struct range *); static void prange(struct range *); static void repos(int); static void separate(const char *); static void increase(void); static int diff3_internal(int, char **, const char *, const char *); int diff3_conflicts = 0; RCSNUM *d3rev1 = NULL; RCSNUM *d3rev2 = NULL; static int fds[5]; void cvs_merge_file(struct cvs_file *cf, int verbose) { int i, argc; char *data, *patch; char *argv[5], r1[CVS_REV_BUFSZ], r2[CVS_REV_BUFSZ]; char *dp13, *dp23, *path1, *path2, *path3; BUF *b1, *b2, *b3, *d1, *d2, *diffb; size_t dlen, plen; struct cvs_line *lp; struct cvs_lines *dlines, *plines; overlapcnt = 0; b1 = b2 = b3 = d1 = d2 = diffb = NULL; rcsnum_tostr(d3rev1, r1, sizeof(r1)); rcsnum_tostr(d3rev2, r2, sizeof(r2)); b1 = cvs_buf_load_fd(cf->fd); d1 = cvs_buf_alloc(128); d2 = cvs_buf_alloc(128); diffb = cvs_buf_alloc(128); (void)xasprintf(&path1, "%s/diff1.XXXXXXXXXX", cvs_tmpdir); (void)xasprintf(&path2, "%s/diff2.XXXXXXXXXX", cvs_tmpdir); (void)xasprintf(&path3, "%s/diff3.XXXXXXXXXX", cvs_tmpdir); fds[2] = cvs_buf_write_stmp(b1, path1, NULL); if (verbose == 1) cvs_printf("Retrieving revision %s\n", r1); fds[3] = rcs_rev_write_stmp(cf->file_rcs, d3rev1, path2, 0); if (verbose == 1) cvs_printf("Retrieving revision %s\n", r2); fds[4] = rcs_rev_write_stmp(cf->file_rcs, d3rev2, path3, 0); cvs_diffreg(path1, path3, fds[2], fds[4], d1); cvs_diffreg(path2, path3, fds[3], fds[4], d2); (void)xasprintf(&dp13, "%s/d13.XXXXXXXXXX", cvs_tmpdir); fds[0] = cvs_buf_write_stmp(d1, dp13, NULL); cvs_buf_free(d1); (void)xasprintf(&dp23, "%s/d23.XXXXXXXXXX", cvs_tmpdir); fds[1] = cvs_buf_write_stmp(d2, dp23, NULL); cvs_buf_free(d2); argc = 0; diffbuf = diffb; argv[argc++] = dp13; argv[argc++] = dp23; argv[argc++] = path1; argv[argc++] = path2; argv[argc++] = path3; if (lseek(fds[2], 0, SEEK_SET) < 0) fatal("cvs_merge_file: lseek fds[2]: %s", strerror(errno)); if (lseek(fds[3], 0, SEEK_SET) < 0) fatal("cvs_merge_file: lseek fds[3]: %s", strerror(errno)); if (lseek(fds[4], 0, SEEK_SET) < 0) fatal("cvs_merge_file: lseek fds[4]: %s", strerror(errno)); diff3_conflicts = diff3_internal(argc, argv, cf->file_path, r2); if (diff3_conflicts < 0) fatal("cvs_merge_file: merging failed for an unknown reason"); plen = cvs_buf_len(diffb); patch = cvs_buf_release(diffb); dlen = cvs_buf_len(b1); data = cvs_buf_release(b1); if (verbose == 1) cvs_printf("Merging differences between %s and %s into `%s'\n", r1, r2, cf->file_path); dlines = cvs_splitlines(data, dlen); plines = cvs_splitlines(patch, plen); ed_patch_lines(dlines, plines); cvs_freelines(plines); if (verbose == 1 && diff3_conflicts != 0) { cvs_log(LP_ERR, "%d conflict%s found during merge, " "please correct.", diff3_conflicts, (diff3_conflicts > 1) ? "s" : ""); } (void)close(cf->fd); cf->fd = open(cf->file_path, O_CREAT | O_RDWR | O_TRUNC, 0644); if (cf->fd == -1) { fatal("cvs_merge_file: failed to reopen fd for writing: %s", strerror(errno)); } TAILQ_FOREACH(lp, &(dlines->l_lines), l_list) { if (lp->l_line == NULL) continue; if (atomicio(vwrite, cf->fd, lp->l_line, lp->l_len) != lp->l_len) fatal("cvs_merge_file: %s", strerror(errno)); } cvs_freelines(dlines); if (data != NULL) xfree(data); xfree(patch); for (i = 0; i < 3; i++) fclose(fp[i]); cvs_worklist_run(&temp_files, cvs_worklist_unlink); xfree(path1); xfree(path2); xfree(path3); xfree(dp13); xfree(dp23); } static int diff3_internal(int argc, char **argv, const char *fmark, const char *rmark) { size_t m, n; int i; eflag = 3; oflag = 1; if (argc < 5) return (-1); (void)xsnprintf(f1mark, sizeof(f1mark), "<<<<<<< %s", fmark); (void)xsnprintf(f3mark, sizeof(f3mark), ">>>>>>> %s", rmark); szchanges = 0; memset(last, 0, sizeof(last)); memset(cline, 0, sizeof(cline)); if (d13 != NULL) xfree(d13); if (d23 != NULL) xfree(d23); if (overlap != NULL) xfree(overlap); if (de != NULL) xfree(de); overlap = NULL; de = d13 = d23 = NULL; increase(); /* fds[0] and fds[1] are closed in readin() */ m = readin(fds[0], &d13); n = readin(fds[1], &d23); for (i = 0; i <= 2; i++) { if ((fp[i] = fdopen(fds[i + 2], "r")) == NULL) { cvs_log(LP_ERR, "%s", argv[i + 2]); return (-1); } } return (merge(m, n)); } int ed_patch_lines(struct cvs_lines *dlines, struct cvs_lines *plines) { char op, *ep; struct cvs_line *sort, *lp, *dlp, *ndlp, *insert_after; int start, end, i, lineno; u_char tmp; dlp = TAILQ_FIRST(&(dlines->l_lines)); lp = TAILQ_FIRST(&(plines->l_lines)); end = 0; for (lp = TAILQ_NEXT(lp, l_list); lp != NULL; lp = TAILQ_NEXT(lp, l_list)) { /* Skip blank lines */ if (lp->l_len < 2) continue; /* NUL-terminate line buffer for strtol() safety. */ tmp = lp->l_line[lp->l_len - 1]; lp->l_line[lp->l_len - 1] = '\0'; op = lp->l_line[strlen(lp->l_line) - 1]; start = (int)strtol(lp->l_line, &ep, 10); /* Restore the last byte of the buffer */ lp->l_line[lp->l_len - 1] = tmp; if (op == 'a') { if (start > dlines->l_nblines || start < 0 || *ep != 'a') fatal("ed_patch_lines %d", start); } else if (op == 'c') { if (start > dlines->l_nblines || start < 0 || (*ep != ',' && *ep != 'c')) fatal("ed_patch_lines"); if (*ep == ',') { ep++; end = (int)strtol(ep, &ep, 10); if (end < 0 || *ep != 'c') fatal("ed_patch_lines"); } else { end = start; } } else { fatal("invalid op %c found while merging", op); } for (;;) { if (dlp == NULL) break; if (dlp->l_lineno == start) break; if (dlp->l_lineno > start) { dlp = TAILQ_PREV(dlp, cvs_tqh, l_list); } else if (dlp->l_lineno < start) { ndlp = TAILQ_NEXT(dlp, l_list); if (ndlp->l_lineno > start) break; dlp = ndlp; } } if (dlp == NULL) fatal("ed_patch_lines"); if (op == 'c') { insert_after = TAILQ_PREV(dlp, cvs_tqh, l_list); for (i = 0; i <= (end - start); i++) { ndlp = TAILQ_NEXT(dlp, l_list); TAILQ_REMOVE(&(dlines->l_lines), dlp, l_list); dlp = ndlp; } dlp = insert_after; } if (op == 'a' || op == 'c') { for (;;) { ndlp = lp; lp = TAILQ_NEXT(lp, l_list); if (lp == NULL) fatal("ed_patch_lines"); if (lp->l_len == 2 && lp->l_line[0] == '.' && lp->l_line[1] == '\n') break; TAILQ_REMOVE(&(plines->l_lines), lp, l_list); TAILQ_INSERT_AFTER(&(dlines->l_lines), dlp, lp, l_list); dlp = lp; lp->l_lineno = start; lp = ndlp; } } /* * always resort lines as the markers might be put at the * same line as we first started editing. */ lineno = 0; TAILQ_FOREACH(sort, &(dlines->l_lines), l_list) sort->l_lineno = lineno++; dlines->l_nblines = lineno - 1; } return (0); } /* * Pick up the line numbers of all changes from one change file. * (This puts the numbers in a vector, which is not strictly necessary, * since the vector is processed in one sequential pass. * The vector could be optimized out of existence) */ static size_t readin(int fd, struct diff **dd) { int a, b, c, d; char kind, *p; size_t i; fp[0] = fdopen(fd, "r"); if (fp[0] == NULL) fatal("readin: fdopen: %s", strerror(errno)); for (i = 0; (p = getchange(fp[0])); i++) { if (i >= szchanges - 1) increase(); a = b = number(&p); if (*p == ',') { p++; b = number(&p); } kind = *p++; c = d = number(&p); if (*p==',') { p++; d = number(&p); } if (kind == 'a') a++; if (kind == 'd') c++; b++; d++; (*dd)[i].old.from = a; (*dd)[i].old.to = b; (*dd)[i].new.from = c; (*dd)[i].new.to = d; } if (i) { (*dd)[i].old.from = (*dd)[i-1].old.to; (*dd)[i].new.from = (*dd)[i-1].new.to; } (void)fclose(fp[0]); return (i); } static int number(char **lc) { int nn; nn = 0; while (isdigit((unsigned char)(**lc))) nn = nn*10 + *(*lc)++ - '0'; return (nn); } static char * getchange(FILE *b) { char *line; while ((line = getline(b, NULL))) { if (isdigit((unsigned char)line[0])) return (line); } return (NULL); } static char * getline(FILE *b, size_t *n) { char *cp; size_t len; static char *buf; static size_t bufsize; if ((cp = fgetln(b, &len)) == NULL) return (NULL); if (cp[len - 1] != '\n') len++; if (len + 1 > bufsize) { do { bufsize += 1024; } while (len + 1 > bufsize); buf = xrealloc(buf, 1, bufsize); } memcpy(buf, cp, len - 1); buf[len - 1] = '\n'; buf[len] = '\0'; if (n != NULL) *n = len; return (buf); } static int merge(size_t m1, size_t m2) { struct diff *d1, *d2, *d3; int dpl, j, t1, t2; d1 = d13; d2 = d23; j = 0; while ((t1 = d1 < d13 + m1) | (t2 = d2 < d23 + m2)) { if (debug) { printf("%d,%d=%d,%d %d,%d=%d,%d\n", d1->old.from, d1->old.to, d1->new.from, d1->new.to, d2->old.from, d2->old.to, d2->new.from, d2->new.to); } /* first file is different from others */ if (!t2 || (t1 && d1->new.to < d2->new.from)) { /* stuff peculiar to 1st file */ if (eflag==0) { separate("1"); change(1, &d1->old, 0); keep(2, &d1->new); change(3, &d1->new, 0); } d1++; continue; } /* second file is different from others */ if (!t1 || (t2 && d2->new.to < d1->new.from)) { if (eflag==0) { separate("2"); keep(1, &d2->new); change(2, &d2->old, 0); change(3, &d2->new, 0); } d2++; continue; } /* * Merge overlapping changes in first file * this happens after extension (see below). */ if (d1 + 1 < d13 + m1 && d1->new.to >= d1[1].new.from) { d1[1].old.from = d1->old.from; d1[1].new.from = d1->new.from; d1++; continue; } /* merge overlapping changes in second */ if (d2 + 1 < d23 + m2 && d2->new.to >= d2[1].new.from) { d2[1].old.from = d2->old.from; d2[1].new.from = d2->new.from; d2++; continue; } /* stuff peculiar to third file or different in all */ if (d1->new.from == d2->new.from && d1->new.to == d2->new.to) { dpl = duplicate(&d1->old,&d2->old); if (dpl == -1) return (-1); /* * dpl = 0 means all files differ * dpl = 1 means files 1 and 2 identical */ if (eflag==0) { separate(dpl ? "3" : ""); change(1, &d1->old, dpl); change(2, &d2->old, 0); d3 = d1->old.to > d1->old.from ? d1 : d2; change(3, &d3->new, 0); } else j = edit(d1, dpl, j); d1++; d2++; continue; } /* * Overlapping changes from file 1 and 2; extend changes * appropriately to make them coincide. */ if (d1->new.from < d2->new.from) { d2->old.from -= d2->new.from-d1->new.from; d2->new.from = d1->new.from; } else if (d2->new.from < d1->new.from) { d1->old.from -= d1->new.from-d2->new.from; d1->new.from = d2->new.from; } if (d1->new.to > d2->new.to) { d2->old.to += d1->new.to - d2->new.to; d2->new.to = d1->new.to; } else if (d2->new.to > d1->new.to) { d1->old.to += d2->new.to - d1->new.to; d1->new.to = d2->new.to; } } return (edscript(j)); } static void separate(const char *s) { diff_output("====%s\n", s); } /* * The range of lines rold.from thru rold.to in file i is to be changed. * It is to be printed only if it does not duplicate something to be * printed later. */ static void change(int i, struct range *rold, int fdup) { diff_output("%d:", i); last[i] = rold->to; prange(rold); if (fdup || debug) return; i--; (void)skip(i, rold->from, NULL); (void)skip(i, rold->to, " "); } /* * print the range of line numbers, rold.from thru rold.to, as n1,n2 or n1 */ static void prange(struct range *rold) { if (rold->to <= rold->from) diff_output("%da\n", rold->from - 1); else { diff_output("%d", rold->from); if (rold->to > rold->from+1) diff_output(",%d", rold->to - 1); diff_output("c\n"); } } /* * No difference was reported by diff between file 1 (or 2) and file 3, * and an artificial dummy difference (trange) must be ginned up to * correspond to the change reported in the other file. */ static void keep(int i, struct range *rnew) { int delta; struct range trange; delta = last[3] - last[i]; trange.from = rnew->from - delta; trange.to = rnew->to - delta; change(i, &trange, 1); } /* * skip to just before line number from in file "i". If "pr" is non-NULL, * print all skipped stuff with string pr as a prefix. */ static int skip(int i, int from, char *pr) { size_t j, n; char *line; for (n = 0; cline[i] < from - 1; n += j) { if ((line = getline(fp[i], &j)) == NULL) return (-1); if (pr != NULL) diff_output("%s%s", pr, line); cline[i]++; } return ((int) n); } /* * Return 1 or 0 according as the old range (in file 1) contains exactly * the same data as the new range (in file 2). */ static int duplicate(struct range *r1, struct range *r2) { int c,d; int nchar; int nline; if (r1->to-r1->from != r2->to-r2->from) return (0); (void)skip(0, r1->from, NULL); (void)skip(1, r2->from, NULL); nchar = 0; for (nline=0; nline < r1->to - r1->from; nline++) { do { c = getc(fp[0]); d = getc(fp[1]); if (c == -1 || d== -1) return (-1); nchar++; if (c != d) { repos(nchar); return (0); } } while (c != '\n'); } repos(nchar); return (1); } static void repos(int nchar) { int i; for (i = 0; i < 2; i++) (void)fseek(fp[i], (long)-nchar, SEEK_CUR); } /* * collect an editing script for later regurgitation */ static int edit(struct diff *diff, int fdup, int j) { if (((fdup + 1) & eflag) == 0) return (j); j++; overlap[j] = !fdup; if (!fdup) overlapcnt++; de[j].old.from = diff->old.from; de[j].old.to = diff->old.to; de[j].new.from = de[j-1].new.to + skip(2, diff->new.from, NULL); de[j].new.to = de[j].new.from + skip(2, diff->new.to, NULL); return (j); } /* regurgitate */ static int edscript(int n) { int j, k; char block[BUFSIZ+1]; for (n = n; n > 0; n--) { if (!oflag || !overlap[n]) prange(&de[n].old); else diff_output("%da\n=======\n", de[n].old.to -1); (void)fseek(fp[2], (long)de[n].new.from, SEEK_SET); for (k = de[n].new.to-de[n].new.from; k > 0; k-= j) { j = k > BUFSIZ ? BUFSIZ : k; if (fread(block, (size_t)1, (size_t)j, fp[2]) != (size_t)j) return (-1); block[j] = '\0'; diff_output("%s", block); } if (!oflag || !overlap[n]) diff_output(".\n"); else { diff_output("%s\n.\n", f3mark); diff_output("%da\n%s\n.\n", de[n].old.from - 1, f1mark); } } return (overlapcnt); } static void increase(void) { size_t newsz, incr; /* are the memset(3) calls needed? */ newsz = szchanges == 0 ? 64 : 2 * szchanges; incr = newsz - szchanges; d13 = xrealloc(d13, newsz, sizeof(*d13)); memset(d13 + szchanges, 0, incr * sizeof(*d13)); d23 = xrealloc(d23, newsz, sizeof(*d23)); memset(d23 + szchanges, 0, incr * sizeof(*d23)); de = xrealloc(de, newsz, sizeof(*de)); memset(de + szchanges, 0, incr * sizeof(*de)); overlap = xrealloc(overlap, newsz, sizeof(*overlap)); memset(overlap + szchanges, 0, incr * sizeof(*overlap)); szchanges = newsz; }