aboutsummaryrefslogtreecommitdiff
path: root/usr.bin/diff
diff options
context:
space:
mode:
authorCem Keylan <cem@ckyln.com>2020-10-16 17:47:01 +0300
committerCem Keylan <cem@ckyln.com>2020-10-16 17:47:01 +0300
commit5d69c6a2661bba0a22f3ecfd517e2e9767a38346 (patch)
tree1f479b2714e127835db7f33a3bfed4c38c52f883 /usr.bin/diff
parente2abcdca396661cbe0ae2ddb13d5c2b85682c13a (diff)
downloadotools-5d69c6a2661bba0a22f3ecfd517e2e9767a38346.tar.gz
add tools
Diffstat (limited to 'usr.bin/diff')
-rw-r--r--usr.bin/diff/CVS/Entries9
-rw-r--r--usr.bin/diff/CVS/Repository1
-rw-r--r--usr.bin/diff/CVS/Root1
-rw-r--r--usr.bin/diff/Makefile7
-rw-r--r--usr.bin/diff/diff.1474
-rw-r--r--usr.bin/diff/diff.c402
-rw-r--r--usr.bin/diff/diff.h98
-rw-r--r--usr.bin/diff/diffdir.c237
-rw-r--r--usr.bin/diff/diffreg.c1485
-rw-r--r--usr.bin/diff/xmalloc.c85
-rw-r--r--usr.bin/diff/xmalloc.h30
11 files changed, 2829 insertions, 0 deletions
diff --git a/usr.bin/diff/CVS/Entries b/usr.bin/diff/CVS/Entries
new file mode 100644
index 0000000..912da0e
--- /dev/null
+++ b/usr.bin/diff/CVS/Entries
@@ -0,0 +1,9 @@
+/Makefile/1.3/Tue May 29 18:24:56 2007//
+/diff.1/1.49/Sat Feb 8 01:09:58 2020//
+/diff.c/1.67/Fri Jun 28 13:35:00 2019//
+/diff.h/1.33/Mon Oct 5 20:15:00 2015//
+/diffdir.c/1.47/Fri Jan 25 00:19:26 2019//
+/diffreg.c/1.93/Fri Jun 28 13:35:00 2019//
+/xmalloc.c/1.10/Fri Jun 28 05:44:09 2019//
+/xmalloc.h/1.4/Thu Nov 12 16:30:30 2015//
+D
diff --git a/usr.bin/diff/CVS/Repository b/usr.bin/diff/CVS/Repository
new file mode 100644
index 0000000..088ef75
--- /dev/null
+++ b/usr.bin/diff/CVS/Repository
@@ -0,0 +1 @@
+src/usr.bin/diff
diff --git a/usr.bin/diff/CVS/Root b/usr.bin/diff/CVS/Root
new file mode 100644
index 0000000..3811072
--- /dev/null
+++ b/usr.bin/diff/CVS/Root
@@ -0,0 +1 @@
+/cvs
diff --git a/usr.bin/diff/Makefile b/usr.bin/diff/Makefile
new file mode 100644
index 0000000..4f1c9d5
--- /dev/null
+++ b/usr.bin/diff/Makefile
@@ -0,0 +1,7 @@
+# $OpenBSD: Makefile,v 1.3 2007/05/29 18:24:56 ray Exp $
+
+PROG= diff
+SRCS= diff.c diffdir.c diffreg.c xmalloc.c
+COPTS+= -Wall
+
+.include <bsd.prog.mk>
diff --git a/usr.bin/diff/diff.1 b/usr.bin/diff/diff.1
new file mode 100644
index 0000000..353c770
--- /dev/null
+++ b/usr.bin/diff/diff.1
@@ -0,0 +1,474 @@
+.\" $OpenBSD: diff.1,v 1.49 2020/02/08 01:09:58 jsg Exp $
+.\"
+.\" Copyright (c) 1980, 1990, 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.
+.\"
+.\" @(#)diff.1 8.1 (Berkeley) 6/30/93
+.\"
+.Dd $Mdocdate: February 8 2020 $
+.Dt DIFF 1
+.Os
+.Sh NAME
+.Nm diff
+.Nd differential file and directory comparator
+.Sh SYNOPSIS
+.Nm diff
+.Op Fl abdipTtw
+.Oo
+.Fl c | e | f |
+.Fl n | q | u
+.Oc
+.Op Fl I Ar pattern
+.Op Fl L Ar label
+.Ar file1 file2
+.Nm diff
+.Op Fl abdilpTtw
+.Op Fl I Ar pattern
+.Op Fl L Ar label
+.Fl C Ar number
+.Ar file1 file2
+.Nm diff
+.Op Fl abdiltw
+.Op Fl I Ar pattern
+.Fl D Ar string
+.Ar file1 file2
+.Nm diff
+.Op Fl abdilpTtw
+.Op Fl I Ar pattern
+.Op Fl L Ar label
+.Fl U Ar number
+.Ar file1 file2
+.Nm diff
+.Op Fl abdilNPprsTtw
+.Oo
+.Fl c | e | f |
+.Fl n | q | u
+.Oc
+.Op Fl I Ar pattern
+.Bk -words
+.Op Fl L Ar label
+.Op Fl S Ar name
+.Op Fl X Ar file
+.Op Fl x Ar pattern
+.Ek
+.Ar dir1 dir2
+.Sh DESCRIPTION
+The
+.Nm
+utility compares the contents of
+.Ar file1
+and
+.Ar file2
+and writes to the standard output the list of changes necessary to
+convert one file into the other.
+No output is produced if the files are identical.
+.Pp
+Output options (mutually exclusive):
+.Bl -tag -width Ds
+.It Fl C Ar number
+Like
+.Fl c
+but produces a diff with
+.Ar number
+lines of context.
+.It Fl c
+Produces a diff with 3 lines of context.
+With
+.Fl c
+the output format is modified slightly:
+the output begins with identification of the files involved and
+their creation dates and then each change is separated
+by a line with fifteen
+.Li * Ns 's .
+The lines removed from
+.Ar file1
+are marked with
+.Sq \-\ \& ;
+those added to
+.Ar file2
+are marked
+.Sq +\ \& .
+Lines which are changed from one file to the other are marked in
+both files with
+.Sq !\ \& .
+Changes which lie within 3 lines of each other are grouped together on
+output.
+.It Fl D Ar string
+Creates a merged version of
+.Ar file1
+and
+.Ar file2
+on the standard output, with C preprocessor controls included so that
+a compilation of the result without defining
+.Ar string
+is equivalent to compiling
+.Ar file1 ,
+while defining
+.Ar string
+will yield
+.Ar file2 .
+.It Fl e
+Produces output in a form suitable as input for the editor utility,
+.Xr ed 1 ,
+which can then be used to convert file1 into file2.
+.Pp
+Extra commands are added to the output when comparing directories with
+.Fl e ,
+so that the result is a
+.Xr sh 1
+script for converting text files which are common to the two directories
+from their state in
+.Ar dir1
+to their state in
+.Ar dir2 .
+.It Fl f
+Identical output to that of the
+.Fl e
+flag, but in reverse order.
+It cannot be digested by
+.Xr ed 1 .
+.It Fl n
+Produces a script similar to that of
+.Fl e ,
+but in the opposite order and with a count of changed lines on each
+insert or delete command.
+This is the form used by
+.Xr rcsdiff 1 .
+.It Fl q
+Just print a line when the files differ.
+Does not output a list of changes.
+.It Fl U Ar number
+Like
+.Fl u
+but produces a diff with
+.Ar number
+lines of context.
+.It Fl u
+Produces a
+.Em unified
+diff with 3 lines of context.
+A unified diff is similar to the context diff produced by the
+.Fl c
+option.
+However, unlike with
+.Fl c ,
+all lines to be changed (added and/or removed) are present in
+a single section.
+.El
+.Pp
+Comparison options:
+.Bl -tag -width Ds
+.It Fl a
+Treat all files as ASCII text.
+Normally
+.Nm
+will simply print
+.Dq Binary files ... differ
+if files contain binary characters.
+Use of this option forces
+.Nm
+to produce a diff.
+.It Fl b
+Causes trailing blanks (spaces and tabs) to be ignored, and other
+strings of blanks to compare equal.
+.It Fl d
+Try very hard to produce a diff as small as possible.
+This may consume a lot of processing power and memory when processing
+large files with many changes.
+.It Fl I Ar pattern
+Ignores changes, insertions, and deletions whose lines match the
+extended regular expression
+.Ar pattern .
+Multiple
+.Fl I
+patterns may be specified.
+All lines in the change must match some pattern for the change to be
+ignored.
+See
+.Xr re_format 7
+for more information on regular expression patterns.
+.It Fl i
+Ignores the case of letters.
+E.g.,
+.Dq A
+will compare equal to
+.Dq a .
+.It Fl L Ar label
+Print
+.Ar label
+instead of the first (and second, if this option is specified twice)
+file name and time in the context or unified diff header.
+.It Fl p
+With unified and context diffs, show with each change
+the first 40 characters of the last line before the context beginning
+with a letter, an underscore or a dollar sign.
+For C source code following standard layout conventions, this will
+show the prototype of the function the change applies to.
+.It Fl T
+Print a tab rather than a space before the rest of the line for the
+normal, context or unified output formats.
+This makes the alignment of tabs in the line consistent.
+.It Fl t
+Will expand tabs in output lines.
+Normal or
+.Fl c
+output adds character(s) to the front of each line which may screw up
+the indentation of the original source lines and make the output listing
+difficult to interpret.
+This option will preserve the original source's indentation.
+.It Fl w
+Is similar to
+.Fl b
+but causes whitespace (blanks and tabs) to be totally ignored.
+E.g.,
+.Dq if (\ \&a == b \&)
+will compare equal to
+.Dq if(a==b) .
+.El
+.Pp
+Directory comparison options:
+.Bl -tag -width Ds
+.It Fl N
+If a file is found in only one directory, act as if it was found in the
+other directory too but was of zero size.
+.It Fl P
+If a file is found only in
+.Ar dir2 ,
+act as if it was found in
+.Ar dir1
+too but was of zero size.
+.It Fl r
+Causes application of
+.Nm
+recursively to common subdirectories encountered.
+.It Fl S Ar name
+Re-starts a directory
+.Nm
+in the middle, beginning with file
+.Ar name .
+.It Fl s
+Causes
+.Nm
+to report files which are the same, which are otherwise not mentioned.
+.It Fl X Ar file
+Exclude files and subdirectories from comparison whose basenames match
+lines in
+.Ar file .
+Multiple
+.Fl X
+options may be specified.
+.It Fl x Ar pattern
+Exclude files and subdirectories from comparison whose basenames match
+.Ar pattern .
+Patterns are matched using shell-style globbing via
+.Xr fnmatch 3 .
+Multiple
+.Fl x
+options may be specified.
+.El
+.Pp
+If both arguments are directories,
+.Nm
+sorts the contents of the directories by name, and then runs the
+regular file
+.Nm
+algorithm, producing a change list,
+on text files which are different.
+Binary files which differ,
+common subdirectories, and files which appear in only one directory
+are described as such.
+In directory mode only regular files and directories are compared.
+If a non-regular file such as a device special file or FIFO
+is encountered, a diagnostic message is printed.
+.Pp
+If only one of
+.Ar file1
+and
+.Ar file2
+is a directory,
+.Nm
+is applied to the non-directory file and the file contained in
+the directory file with a filename that is the same as the
+last component of the non-directory file.
+.Pp
+If either
+.Ar file1
+or
+.Ar file2
+is
+.Sq - ,
+the standard input is
+used in its place.
+.Ss Output Style
+The default (without
+.Fl e ,
+.Fl c ,
+or
+.Fl n
+.\" -C
+options)
+output contains lines of these forms, where
+.Va XX , YY , ZZ , QQ
+are line numbers respective of file order.
+.Pp
+.Bl -tag -width "XX,YYcZZ,QQ" -compact
+.It Li XX Ns Ic a Ns Li YY
+At (the end of) line
+.Va XX
+of
+.Ar file1 ,
+append the contents
+of line
+.Va YY
+of
+.Ar file2
+to make them equal.
+.It Li XX Ns Ic a Ns Li YY,ZZ
+Same as above, but append the range of lines,
+.Va YY
+through
+.Va ZZ
+of
+.Ar file2
+to line
+.Va XX
+of file1.
+.It Li XX Ns Ic d Ns Li YY
+At line
+.Va XX
+delete
+the line.
+The value
+.Va YY
+tells to which line the change would bring
+.Ar file1
+in line with
+.Ar file2 .
+.It Li XX,YY Ns Ic d Ns Li ZZ
+Delete the range of lines
+.Va XX
+through
+.Va YY
+in
+.Ar file1 .
+.It Li XX Ns Ic c Ns Li YY
+Change the line
+.Va XX
+in
+.Ar file1
+to the line
+.Va YY
+in
+.Ar file2 .
+.It Li XX,YY Ns Ic c Ns Li ZZ
+Replace the range of specified lines with the line
+.Va ZZ .
+.It Li XX,YY Ns Ic c Ns Li ZZ,QQ
+Replace the range
+.Va XX , Ns Va YY
+from
+.Ar file1
+with the range
+.Va ZZ , Ns Va QQ
+from
+.Ar file2 .
+.El
+.Pp
+These lines resemble
+.Xr ed 1
+subcommands to convert
+.Ar file1
+into
+.Ar file2 .
+The line numbers before the action letters pertain to
+.Ar file1 ;
+those after pertain to
+.Ar file2 .
+Thus, by exchanging
+.Ic a
+for
+.Ic d
+and reading the line in reverse order, one can also
+determine how to convert
+.Ar file2
+into
+.Ar file1 .
+As in
+.Xr ed 1 ,
+identical
+pairs (where num1 = num2) are abbreviated as a single
+number.
+.Sh FILES
+.Bl -tag -width /tmp/diff.XXXXXXXX -compact
+.It Pa /tmp/diff. Ns Ar XXXXXXXX
+Temporary file used when comparing a device or the standard input.
+Note that the temporary file is unlinked as soon as it is created
+so it will not show up in a directory listing.
+.El
+.Sh EXIT STATUS
+The
+.Nm
+utility exits with one of the following values:
+.Pp
+.Bl -tag -width Ds -offset indent -compact
+.It 0
+No differences were found.
+.It 1
+Differences were found.
+.It >1
+An error occurred.
+.El
+.Sh SEE ALSO
+.Xr cmp 1 ,
+.Xr comm 1 ,
+.Xr diff3 1 ,
+.Xr ed 1 ,
+.Xr patch 1 ,
+.Xr sdiff 1
+.Rs
+.%A James W. Hunt
+.%A M. Douglas McIlroy
+.%T "An Algorithm for Differential File Comparison"
+.%J Computing Science Technical Report
+.%Q Bell Laboratories 41
+.%D June 1976
+.Re
+.Sh STANDARDS
+The
+.Nm
+utility is compliant with the
+.St -p1003.1-2008
+specification.
+.Pp
+The flags
+.Op Fl aDdIiLlNnPpqSsTtwXx
+are extensions to that specification.
+.Sh HISTORY
+A
+.Nm
+command appeared in
+.At v5 .
diff --git a/usr.bin/diff/diff.c b/usr.bin/diff/diff.c
new file mode 100644
index 0000000..64cdd45
--- /dev/null
+++ b/usr.bin/diff/diff.c
@@ -0,0 +1,402 @@
+/* $OpenBSD: diff.c,v 1.67 2019/06/28 13:35:00 deraadt Exp $ */
+
+/*
+ * Copyright (c) 2003 Todd C. Miller <millert@openbsd.org>
+ *
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose with or without fee is hereby granted, provided that the above
+ * copyright notice and this permission notice appear in all copies.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
+ * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
+ * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
+ * ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
+ * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
+ * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
+ * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
+ *
+ * Sponsored in part by the Defense Advanced Research Projects
+ * Agency (DARPA) and Air Force Research Laboratory, Air Force
+ * Materiel Command, USAF, under agreement number F39502-99-1-0512.
+ */
+
+#include <sys/cdefs.h>
+#include <sys/stat.h>
+
+#include <ctype.h>
+#include <err.h>
+#include <errno.h>
+#include <getopt.h>
+#include <stdlib.h>
+#include <stdio.h>
+#include <stdarg.h>
+#include <string.h>
+#include <unistd.h>
+#include <limits.h>
+
+#include "diff.h"
+#include "xmalloc.h"
+
+int Nflag, Pflag, rflag, sflag, Tflag;
+int diff_format, diff_context, status;
+char *start, *ifdefname, *diffargs, *label[2], *ignore_pats;
+struct stat stb1, stb2;
+struct excludes *excludes_list;
+regex_t ignore_re;
+
+#define OPTIONS "0123456789abC:cdD:efhI:iL:lnNPpqrS:sTtU:uwX:x:"
+static struct option longopts[] = {
+ { "text", no_argument, 0, 'a' },
+ { "ignore-space-change", no_argument, 0, 'b' },
+ { "context", optional_argument, 0, 'C' },
+ { "ifdef", required_argument, 0, 'D' },
+ { "minimal", no_argument, 0, 'd' },
+ { "ed", no_argument, 0, 'e' },
+ { "forward-ed", no_argument, 0, 'f' },
+ { "ignore-matching-lines", required_argument, 0, 'I' },
+ { "ignore-case", no_argument, 0, 'i' },
+ { "label", required_argument, 0, 'L' },
+ { "new-file", no_argument, 0, 'N' },
+ { "rcs", no_argument, 0, 'n' },
+ { "unidirectional-new-file", no_argument, 0, 'P' },
+ { "show-c-function", no_argument, 0, 'p' },
+ { "brief", no_argument, 0, 'q' },
+ { "recursive", no_argument, 0, 'r' },
+ { "report-identical-files", no_argument, 0, 's' },
+ { "starting-file", required_argument, 0, 'S' },
+ { "expand-tabs", no_argument, 0, 't' },
+ { "initial-tab", no_argument, 0, 'T' },
+ { "unified", optional_argument, 0, 'U' },
+ { "ignore-all-space", no_argument, 0, 'w' },
+ { "exclude", required_argument, 0, 'x' },
+ { "exclude-from", required_argument, 0, 'X' },
+ { NULL, 0, 0, '\0'}
+};
+
+__dead void usage(void);
+void push_excludes(char *);
+void push_ignore_pats(char *);
+void read_excludes_file(char *file);
+void set_argstr(char **, char **);
+
+int
+main(int argc, char **argv)
+{
+ char *ep, **oargv;
+ long l;
+ int ch, dflags, lastch, gotstdin, prevoptind, newarg;
+
+ oargv = argv;
+ gotstdin = 0;
+ dflags = 0;
+ lastch = '\0';
+ prevoptind = 1;
+ newarg = 1;
+ while ((ch = getopt_long(argc, argv, OPTIONS, longopts, NULL)) != -1) {
+ switch (ch) {
+ case '0': case '1': case '2': case '3': case '4':
+ case '5': case '6': case '7': case '8': case '9':
+ if (newarg)
+ usage(); /* disallow -[0-9]+ */
+ else if (lastch == 'c' || lastch == 'u')
+ diff_context = 0;
+ else if (!isdigit(lastch) || diff_context > INT_MAX / 10)
+ usage();
+ diff_context = (diff_context * 10) + (ch - '0');
+ break;
+ case 'a':
+ dflags |= D_FORCEASCII;
+ break;
+ case 'b':
+ dflags |= D_FOLDBLANKS;
+ break;
+ case 'C':
+ case 'c':
+ diff_format = D_CONTEXT;
+ if (optarg != NULL) {
+ l = strtol(optarg, &ep, 10);
+ if (*ep != '\0' || l < 0 || l >= INT_MAX)
+ usage();
+ diff_context = (int)l;
+ } else
+ diff_context = 3;
+ break;
+ case 'd':
+ dflags |= D_MINIMAL;
+ break;
+ case 'D':
+ diff_format = D_IFDEF;
+ ifdefname = optarg;
+ break;
+ case 'e':
+ diff_format = D_EDIT;
+ break;
+ case 'f':
+ diff_format = D_REVERSE;
+ break;
+ case 'h':
+ /* silently ignore for backwards compatibility */
+ break;
+ case 'I':
+ push_ignore_pats(optarg);
+ break;
+ case 'i':
+ dflags |= D_IGNORECASE;
+ break;
+ case 'L':
+ if (label[0] == NULL)
+ label[0] = optarg;
+ else if (label[1] == NULL)
+ label[1] = optarg;
+ else
+ usage();
+ break;
+ case 'N':
+ Nflag = 1;
+ break;
+ case 'n':
+ diff_format = D_NREVERSE;
+ break;
+ case 'p':
+ dflags |= D_PROTOTYPE;
+ break;
+ case 'P':
+ Pflag = 1;
+ break;
+ case 'r':
+ rflag = 1;
+ break;
+ case 'q':
+ diff_format = D_BRIEF;
+ break;
+ case 'S':
+ start = optarg;
+ break;
+ case 's':
+ sflag = 1;
+ break;
+ case 'T':
+ Tflag = 1;
+ break;
+ case 't':
+ dflags |= D_EXPANDTABS;
+ break;
+ case 'U':
+ case 'u':
+ diff_format = D_UNIFIED;
+ if (optarg != NULL) {
+ l = strtol(optarg, &ep, 10);
+ if (*ep != '\0' || l < 0 || l >= INT_MAX)
+ usage();
+ diff_context = (int)l;
+ } else
+ diff_context = 3;
+ break;
+ case 'w':
+ dflags |= D_IGNOREBLANKS;
+ break;
+ case 'X':
+ read_excludes_file(optarg);
+ break;
+ case 'x':
+ push_excludes(optarg);
+ break;
+ default:
+ usage();
+ break;
+ }
+ lastch = ch;
+ newarg = optind != prevoptind;
+ prevoptind = optind;
+ }
+ argc -= optind;
+ argv += optind;
+
+ if (pledge("stdio rpath tmppath", NULL) == -1)
+ err(2, "pledge");
+
+ /*
+ * Do sanity checks, fill in stb1 and stb2 and call the appropriate
+ * driver routine. Both drivers use the contents of stb1 and stb2.
+ */
+ if (argc != 2)
+ usage();
+ if (ignore_pats != NULL) {
+ char buf[BUFSIZ];
+ int error;
+
+ if ((error = regcomp(&ignore_re, ignore_pats,
+ REG_NEWLINE | REG_EXTENDED)) != 0) {
+ regerror(error, &ignore_re, buf, sizeof(buf));
+ if (*ignore_pats != '\0')
+ errx(2, "%s: %s", ignore_pats, buf);
+ else
+ errx(2, "%s", buf);
+ }
+ }
+ if (strcmp(argv[0], "-") == 0) {
+ fstat(STDIN_FILENO, &stb1);
+ gotstdin = 1;
+ } else if (stat(argv[0], &stb1) != 0)
+ err(2, "%s", argv[0]);
+ if (strcmp(argv[1], "-") == 0) {
+ fstat(STDIN_FILENO, &stb2);
+ gotstdin = 1;
+ } else if (stat(argv[1], &stb2) != 0)
+ err(2, "%s", argv[1]);
+ if (gotstdin && (S_ISDIR(stb1.st_mode) || S_ISDIR(stb2.st_mode)))
+ errx(2, "can't compare - to a directory");
+ set_argstr(oargv, argv);
+ if (S_ISDIR(stb1.st_mode) && S_ISDIR(stb2.st_mode)) {
+ if (diff_format == D_IFDEF)
+ errx(2, "-D option not supported with directories");
+ diffdir(argv[0], argv[1], dflags);
+ } else {
+ if (S_ISDIR(stb1.st_mode)) {
+ argv[0] = splice(argv[0], argv[1]);
+ if (stat(argv[0], &stb1) == -1)
+ err(2, "%s", argv[0]);
+ }
+ if (S_ISDIR(stb2.st_mode)) {
+ argv[1] = splice(argv[1], argv[0]);
+ if (stat(argv[1], &stb2) == -1)
+ err(2, "%s", argv[1]);
+ }
+ print_status(diffreg(argv[0], argv[1], dflags), argv[0], argv[1],
+ "");
+ }
+ exit(status);
+}
+
+void
+set_argstr(char **av, char **ave)
+{
+ size_t argsize;
+ char **ap;
+
+ argsize = 4 + *ave - *av + 1;
+ diffargs = xmalloc(argsize);
+ strlcpy(diffargs, "diff", argsize);
+ for (ap = av + 1; ap < ave; ap++) {
+ if (strcmp(*ap, "--") != 0) {
+ strlcat(diffargs, " ", argsize);
+ strlcat(diffargs, *ap, argsize);
+ }
+ }
+}
+
+/*
+ * Read in an excludes file and push each line.
+ */
+void
+read_excludes_file(char *file)
+{
+ FILE *fp;
+ char *buf, *pattern;
+ size_t len;
+
+ if (strcmp(file, "-") == 0)
+ fp = stdin;
+ else if ((fp = fopen(file, "r")) == NULL)
+ err(2, "%s", file);
+ while ((buf = fgetln(fp, &len)) != NULL) {
+ if (buf[len - 1] == '\n')
+ len--;
+ pattern = xmalloc(len + 1);
+ memcpy(pattern, buf, len);
+ pattern[len] = '\0';
+ push_excludes(pattern);
+ }
+ if (strcmp(file, "-") != 0)
+ fclose(fp);
+}
+
+/*
+ * Push a pattern onto the excludes list.
+ */
+void
+push_excludes(char *pattern)
+{
+ struct excludes *entry;
+
+ entry = xmalloc(sizeof(*entry));
+ entry->pattern = pattern;
+ entry->next = excludes_list;
+ excludes_list = entry;
+}
+
+void
+push_ignore_pats(char *pattern)
+{
+ size_t len;
+
+ if (ignore_pats == NULL)
+ ignore_pats = xstrdup(pattern);
+ else {
+ /* old + "|" + new + NUL */
+ len = strlen(ignore_pats) + strlen(pattern) + 2;
+ ignore_pats = xreallocarray(ignore_pats, 1, len);
+ strlcat(ignore_pats, "|", len);
+ strlcat(ignore_pats, pattern, len);
+ }
+}
+
+void
+print_only(const char *path, size_t dirlen, const char *entry)
+{
+ if (dirlen > 1)
+ dirlen--;
+ printf("Only in %.*s: %s\n", (int)dirlen, path, entry);
+}
+
+void
+print_status(int val, char *path1, char *path2, char *entry)
+{
+ switch (val) {
+ case D_BINARY:
+ printf("Binary files %s%s and %s%s differ\n",
+ path1, entry, path2, entry);
+ break;
+ case D_DIFFER:
+ if (diff_format == D_BRIEF)
+ printf("Files %s%s and %s%s differ\n",
+ path1, entry, path2, entry);
+ break;
+ case D_SAME:
+ if (sflag)
+ printf("Files %s%s and %s%s are identical\n",
+ path1, entry, path2, entry);
+ break;
+ case D_MISMATCH1:
+ printf("File %s%s is a directory while file %s%s is a regular file\n",
+ path1, entry, path2, entry);
+ break;
+ case D_MISMATCH2:
+ printf("File %s%s is a regular file while file %s%s is a directory\n",
+ path1, entry, path2, entry);
+ break;
+ case D_SKIPPED1:
+ printf("File %s%s is not a regular file or directory and was skipped\n",
+ path1, entry);
+ break;
+ case D_SKIPPED2:
+ printf("File %s%s is not a regular file or directory and was skipped\n",
+ path2, entry);
+ break;
+ }
+}
+
+__dead void
+usage(void)
+{
+ (void)fprintf(stderr,
+ "usage: diff [-abdipTtw] [-c | -e | -f | -n | -q | -u] [-I pattern] [-L label]\n"
+ " file1 file2\n"
+ " diff [-abdipTtw] [-I pattern] [-L label] -C number file1 file2\n"
+ " diff [-abditw] [-I pattern] -D string file1 file2\n"
+ " diff [-abdipTtw] [-I pattern] [-L label] -U number file1 file2\n"
+ " diff [-abdiNPprsTtw] [-c | -e | -f | -n | -q | -u] [-I pattern]\n"
+ " [-L label] [-S name] [-X file] [-x pattern] dir1 dir2\n");
+
+ exit(2);
+}
diff --git a/usr.bin/diff/diff.h b/usr.bin/diff/diff.h
new file mode 100644
index 0000000..3a36222
--- /dev/null
+++ b/usr.bin/diff/diff.h
@@ -0,0 +1,98 @@
+
+
+/*ROR
+ * 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.
+ *
+ * @(#)diff.h 8.1 (Berkeley) 6/6/93
+ */
+
+#include <sys/types.h>
+#include <regex.h>
+
+/*
+ * Output format options
+ */
+#define D_NORMAL 0 /* Normal output */
+#define D_EDIT -1 /* Editor script out */
+#define D_REVERSE 1 /* Reverse editor script */
+#define D_CONTEXT 2 /* Diff with context */
+#define D_UNIFIED 3 /* Unified context diff */
+#define D_IFDEF 4 /* Diff with merged #ifdef's */
+#define D_NREVERSE 5 /* Reverse ed script with numbered
+ lines and no trailing . */
+#define D_BRIEF 6 /* Say if the files differ */
+
+/*
+ * Output flags
+ */
+#define D_HEADER 0x001 /* Print a header/footer between files */
+#define D_EMPTY1 0x002 /* Treat first file as empty (/dev/null) */
+#define D_EMPTY2 0x004 /* Treat second file as empty (/dev/null) */
+
+/*
+ * Command line flags
+ */
+#define D_FORCEASCII 0x008 /* Treat file as ascii regardless of content */
+#define D_FOLDBLANKS 0x010 /* Treat all white space as equal */
+#define D_MINIMAL 0x020 /* Make diff as small as possible */
+#define D_IGNORECASE 0x040 /* Case-insensitive matching */
+#define D_PROTOTYPE 0x080 /* Display C function prototype */
+#define D_EXPANDTABS 0x100 /* Expand tabs to spaces */
+#define D_IGNOREBLANKS 0x200 /* Ignore white space changes */
+
+/*
+ * Status values for print_status() and diffreg() return values
+ */
+#define D_SAME 0 /* Files are the same */
+#define D_DIFFER 1 /* Files are different */
+#define D_BINARY 2 /* Binary files are different */
+#define D_MISMATCH1 3 /* path1 was a dir, path2 a file */
+#define D_MISMATCH2 4 /* path1 was a file, path2 a dir */
+#define D_SKIPPED1 5 /* path1 was a special file */
+#define D_SKIPPED2 6 /* path2 was a special file */
+
+struct excludes {
+ char *pattern;
+ struct excludes *next;
+};
+
+extern int Nflag, Pflag, rflag, sflag, Tflag;
+extern int diff_format, diff_context, status;
+extern char *start, *ifdefname, *diffargs, *label[2], *ignore_pats;
+extern struct stat stb1, stb2;
+extern struct excludes *excludes_list;
+extern regex_t ignore_re;
+
+char *splice(char *, char *);
+int diffreg(char *, char *, int);
+int easprintf(char **, const char *, ...);
+void *emalloc(size_t);
+void *erealloc(void *, size_t);
+void diffdir(char *, char *, int);
+void print_only(const char *, size_t, const char *);
+void print_status(int, char *, char *, char *);
diff --git a/usr.bin/diff/diffdir.c b/usr.bin/diff/diffdir.c
new file mode 100644
index 0000000..ad1acdb
--- /dev/null
+++ b/usr.bin/diff/diffdir.c
@@ -0,0 +1,237 @@
+/* $OpenBSD: diffdir.c,v 1.47 2019/01/25 00:19:26 millert Exp $ */
+
+/*
+ * Copyright (c) 2003, 2010 Todd C. Miller <millert@openbsd.org>
+ *
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose with or without fee is hereby granted, provided that the above
+ * copyright notice and this permission notice appear in all copies.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
+ * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
+ * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
+ * ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
+ * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
+ * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
+ * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
+ *
+ * Sponsored in part by the Defense Advanced Research Projects
+ * Agency (DARPA) and Air Force Research Laboratory, Air Force
+ * Materiel Command, USAF, under agreement number F39502-99-1-0512.
+ */
+
+#include <sys/stat.h>
+
+#include <dirent.h>
+#include <err.h>
+#include <errno.h>
+#include <fcntl.h>
+#include <fnmatch.h>
+#include <paths.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <unistd.h>
+#include <limits.h>
+
+#include "diff.h"
+#include "xmalloc.h"
+
+static int selectfile(const struct dirent *);
+static void diffit(struct dirent *, char *, size_t, char *, size_t, int);
+
+#define d_status d_type /* we need to store status for -l */
+
+/*
+ * Diff directory traversal. Will be called recursively if -r was specified.
+ */
+void
+diffdir(char *p1, char *p2, int flags)
+{
+ struct dirent *dent1, **dp1, **edp1, **dirp1 = NULL;
+ struct dirent *dent2, **dp2, **edp2, **dirp2 = NULL;
+ size_t dirlen1, dirlen2;
+ char path1[PATH_MAX], path2[PATH_MAX];
+ int pos;
+
+ dirlen1 = strlcpy(path1, *p1 ? p1 : ".", sizeof(path1));
+ if (dirlen1 >= sizeof(path1) - 1) {
+ warnc(ENAMETOOLONG, "%s", p1);
+ status |= 2;
+ return;
+ }
+ if (path1[dirlen1 - 1] != '/') {
+ path1[dirlen1++] = '/';
+ path1[dirlen1] = '\0';
+ }
+ dirlen2 = strlcpy(path2, *p2 ? p2 : ".", sizeof(path2));
+ if (dirlen2 >= sizeof(path2) - 1) {
+ warnc(ENAMETOOLONG, "%s", p2);
+ status |= 2;
+ return;
+ }
+ if (path2[dirlen2 - 1] != '/') {
+ path2[dirlen2++] = '/';
+ path2[dirlen2] = '\0';
+ }
+
+ /*
+ * Get a list of entries in each directory, skipping "excluded" files
+ * and sorting alphabetically.
+ */
+ pos = scandir(path1, &dirp1, selectfile, alphasort);
+ if (pos == -1) {
+ if (errno == ENOENT && (Nflag || Pflag)) {
+ pos = 0;
+ } else {
+ warn("%s", path1);
+ goto closem;
+ }
+ }
+ dp1 = dirp1;
+ edp1 = dirp1 + pos;
+
+ pos = scandir(path2, &dirp2, selectfile, alphasort);
+ if (pos == -1) {
+ if (errno == ENOENT && Nflag) {
+ pos = 0;
+ } else {
+ warn("%s", path2);
+ goto closem;
+ }
+ }
+ dp2 = dirp2;
+ edp2 = dirp2 + pos;
+
+ /*
+ * If we were given a starting point, find it.
+ */
+ if (start != NULL) {
+ while (dp1 != edp1 && strcmp((*dp1)->d_name, start) < 0)
+ dp1++;
+ while (dp2 != edp2 && strcmp((*dp2)->d_name, start) < 0)
+ dp2++;
+ }
+
+ /*
+ * Iterate through the two directory lists, diffing as we go.
+ */
+ while (dp1 != edp1 || dp2 != edp2) {
+ dent1 = dp1 != edp1 ? *dp1 : NULL;
+ dent2 = dp2 != edp2 ? *dp2 : NULL;
+
+ pos = dent1 == NULL ? 1 : dent2 == NULL ? -1 :
+ strcmp(dent1->d_name, dent2->d_name);
+ if (pos == 0) {
+ /* file exists in both dirs, diff it */
+ diffit(dent1, path1, dirlen1, path2, dirlen2, flags);
+ dp1++;
+ dp2++;
+ } else if (pos < 0) {
+ /* file only in first dir, only diff if -N */
+ if (Nflag) {
+ diffit(dent1, path1, dirlen1, path2, dirlen2,
+ flags);
+ } else {
+ print_only(path1, dirlen1, dent1->d_name);
+ status |= 1;
+ }
+ dp1++;
+ } else {
+ /* file only in second dir, only diff if -N or -P */
+ if (Nflag || Pflag) {
+ diffit(dent2, path1, dirlen1, path2, dirlen2,
+ flags);
+ } else {
+ print_only(path2, dirlen2, dent2->d_name);
+ status |= 1;
+ }
+ dp2++;
+ }
+ }
+
+closem:
+ if (dirp1 != NULL) {
+ for (dp1 = dirp1; dp1 < edp1; dp1++)
+ free(*dp1);
+ free(dirp1);
+ }
+ if (dirp2 != NULL) {
+ for (dp2 = dirp2; dp2 < edp2; dp2++)
+ free(*dp2);
+ free(dirp2);
+ }
+}
+
+/*
+ * Do the actual diff by calling either diffreg() or diffdir().
+ */
+static void
+diffit(struct dirent *dp, char *path1, size_t plen1, char *path2, size_t plen2,
+ int flags)
+{
+ flags |= D_HEADER;
+ strlcpy(path1 + plen1, dp->d_name, PATH_MAX - plen1);
+ if (stat(path1, &stb1) != 0) {
+ if (!(Nflag || Pflag) || errno != ENOENT) {
+ warn("%s", path1);
+ return;
+ }
+ flags |= D_EMPTY1;
+ memset(&stb1, 0, sizeof(stb1));
+ }
+
+ strlcpy(path2 + plen2, dp->d_name, PATH_MAX - plen2);
+ if (stat(path2, &stb2) != 0) {
+ if (!Nflag || errno != ENOENT) {
+ warn("%s", path2);
+ return;
+ }
+ flags |= D_EMPTY2;
+ memset(&stb2, 0, sizeof(stb2));
+ stb2.st_mode = stb1.st_mode;
+ }
+ if (stb1.st_mode == 0)
+ stb1.st_mode = stb2.st_mode;
+
+ if (S_ISDIR(stb1.st_mode) && S_ISDIR(stb2.st_mode)) {
+ if (rflag)
+ diffdir(path1, path2, flags);
+ else
+ printf("Common subdirectories: %s and %s\n",
+ path1, path2);
+ return;
+ }
+ if (!S_ISREG(stb1.st_mode) && !S_ISDIR(stb1.st_mode))
+ dp->d_status = D_SKIPPED1;
+ else if (!S_ISREG(stb2.st_mode) && !S_ISDIR(stb2.st_mode))
+ dp->d_status = D_SKIPPED2;
+ else
+ dp->d_status = diffreg(path1, path2, flags);
+ print_status(dp->d_status, path1, path2, "");
+}
+
+/*
+ * Returns 1 if the directory entry should be included in the
+ * diff, else 0. Checks the excludes list.
+ */
+static int
+selectfile(const struct dirent *dp)
+{
+ struct excludes *excl;
+
+ if (dp->d_fileno == 0)
+ return (0);
+
+ /* always skip "." and ".." */
+ if (dp->d_name[0] == '.' && (dp->d_name[1] == '\0' ||
+ (dp->d_name[1] == '.' && dp->d_name[2] == '\0')))
+ return (0);
+
+ /* check excludes list */
+ for (excl = excludes_list; excl != NULL; excl = excl->next)
+ if (fnmatch(excl->pattern, dp->d_name, FNM_PATHNAME) == 0)
+ return (0);
+
+ return (1);
+}
diff --git a/usr.bin/diff/diffreg.c b/usr.bin/diff/diffreg.c
new file mode 100644
index 0000000..fc0029d
--- /dev/null
+++ b/usr.bin/diff/diffreg.c
@@ -0,0 +1,1485 @@
+/* $OpenBSD: diffreg.c,v 1.93 2019/06/28 13:35:00 deraadt 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.
+ *
+ * @(#)diffreg.c 8.1 (Berkeley) 6/6/93
+ */
+
+#include <sys/stat.h>
+#include <sys/wait.h>
+
+#include <ctype.h>
+#include <err.h>
+#include <errno.h>
+#include <fcntl.h>
+#include <paths.h>
+#include <stddef.h>
+#include <stdint.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <time.h>
+#include <unistd.h>
+#include <limits.h>
+
+#include "diff.h"
+#include "xmalloc.h"
+
+#define MINIMUM(a, b) (((a) < (b)) ? (a) : (b))
+#define MAXIMUM(a, b) (((a) > (b)) ? (a) : (b))
+
+/*
+ * diff - compare two files.
+ */
+
+/*
+ * Uses an algorithm due to Harold Stone, which finds
+ * a pair of longest identical subsequences in the two
+ * files.
+ *
+ * The major goal is to generate the match vector J.
+ * J[i] is the index of the line in file1 corresponding
+ * to line i file0. J[i] = 0 if there is no
+ * such line in file1.
+ *
+ * Lines are hashed so as to work in core. All potential
+ * matches are located by sorting the lines of each file
+ * on the hash (called ``value''). In particular, this
+ * collects the equivalence classes in file1 together.
+ * Subroutine equiv replaces the value of each line in
+ * file0 by the index of the first element of its
+ * matching equivalence in (the reordered) file1.
+ * To save space equiv squeezes file1 into a single
+ * array member in which the equivalence classes
+ * are simply concatenated, except that their first
+ * members are flagged by changing sign.
+ *
+ * Next the indices that point into member are unsorted into
+ * array class according to the original order of file0.
+ *
+ * The cleverness lies in routine stone. This marches
+ * through the lines of file0, developing a vector klist
+ * of "k-candidates". At step i a k-candidate is a matched
+ * pair of lines x,y (x in file0 y in file1) such that
+ * there is a common subsequence of length k
+ * between the first i lines of file0 and the first y
+ * lines of file1, but there is no such subsequence for
+ * any smaller y. x is the earliest possible mate to y
+ * that occurs in such a subsequence.
+ *
+ * Whenever any of the members of the equivalence class of
+ * lines in file1 matable to a line in file0 has serial number
+ * less than the y of some k-candidate, that k-candidate
+ * with the smallest such y is replaced. The new
+ * k-candidate is chained (via pred) to the current
+ * k-1 candidate so that the actual subsequence can
+ * be recovered. When a member has serial number greater
+ * that the y of all k-candidates, the klist is extended.
+ * At the end, the longest subsequence is pulled out
+ * and placed in the array J by unravel
+ *
+ * With J in hand, the matches there recorded are
+ * check'ed against reality to assure that no spurious
+ * matches have crept in due to hashing. If they have,
+ * they are broken, and "jackpot" is recorded--a harmless
+ * matter except that a true match for a spuriously
+ * mated line may now be unnecessarily reported as a change.
+ *
+ * Much of the complexity of the program comes simply
+ * from trying to minimize core utilization and
+ * maximize the range of doable problems by dynamically
+ * allocating what is needed and reusing what is not.
+ * The core requirements for problems larger than somewhat
+ * are (in words) 2*length(file0) + length(file1) +
+ * 3*(number of k-candidates installed), typically about
+ * 6n words for files of length n.
+ */
+
+struct cand {
+ int x;
+ int y;
+ int pred;
+};
+
+struct line {
+ int serial;
+ int value;
+} *file[2];
+
+/*
+ * The following struct is used to record change information when
+ * doing a "context" or "unified" diff. (see routine "change" to
+ * understand the highly mnemonic field names)
+ */
+struct context_vec {
+ int a; /* start line in old file */
+ int b; /* end line in old file */
+ int c; /* start line in new file */
+ int d; /* end line in new file */
+};
+
+#define diff_output printf
+static FILE *opentemp(const char *);
+static void output(char *, FILE *, char *, FILE *, int);
+static void check(FILE *, FILE *, int);
+static void range(int, int, char *);
+static void uni_range(int, int);
+static void dump_context_vec(FILE *, FILE *, int);
+static void dump_unified_vec(FILE *, FILE *, int);
+static void prepare(int, FILE *, off_t, int);
+static void prune(void);
+static void equiv(struct line *, int, struct line *, int, int *);
+static void unravel(int);
+static void unsort(struct line *, int, int *);
+static void change(char *, FILE *, char *, FILE *, int, int, int, int, int *);
+static void sort(struct line *, int);
+static void print_header(const char *, const char *);
+static int ignoreline(char *);
+static int asciifile(FILE *);
+static int fetch(long *, int, int, FILE *, int, int, int);
+static int newcand(int, int, int);
+static int search(int *, int, int);
+static int skipline(FILE *);
+static int isqrt(int);
+static int stone(int *, int, int *, int *, int);
+static int readhash(FILE *, int);
+static int files_differ(FILE *, FILE *, int);
+static char *match_function(const long *, int, FILE *);
+static char *preadline(int, size_t, off_t);
+
+static int *J; /* will be overlaid on class */
+static int *class; /* will be overlaid on file[0] */
+static int *klist; /* will be overlaid on file[0] after class */
+static int *member; /* will be overlaid on file[1] */
+static int clen;
+static int inifdef; /* whether or not we are in a #ifdef block */
+static int len[2];
+static int pref, suff; /* length of prefix and suffix */
+static int slen[2];
+static int anychange;
+static long *ixnew; /* will be overlaid on file[1] */
+static long *ixold; /* will be overlaid on klist */
+static struct cand *clist; /* merely a free storage pot for candidates */
+static int clistlen; /* the length of clist */
+static struct line *sfile[2]; /* shortened by pruning common prefix/suffix */
+static u_char *chrtran; /* translation table for case-folding */
+static struct context_vec *context_vec_start;
+static struct context_vec *context_vec_end;
+static struct context_vec *context_vec_ptr;
+
+#define FUNCTION_CONTEXT_SIZE 55
+static char lastbuf[FUNCTION_CONTEXT_SIZE];
+static int lastline;
+static int lastmatchline;
+
+
+/*
+ * chrtran points to one of 2 translation tables: cup2low if folding upper to
+ * lower case clow2low if not folding case
+ */
+u_char clow2low[256] = {
+ 0x00, 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07, 0x08, 0x09, 0x0a,
+ 0x0b, 0x0c, 0x0d, 0x0e, 0x0f, 0x10, 0x11, 0x12, 0x13, 0x14, 0x15,
+ 0x16, 0x17, 0x18, 0x19, 0x1a, 0x1b, 0x1c, 0x1d, 0x1e, 0x1f, 0x20,
+ 0x21, 0x22, 0x23, 0x24, 0x25, 0x26, 0x27, 0x28, 0x29, 0x2a, 0x2b,
+ 0x2c, 0x2d, 0x2e, 0x2f, 0x30, 0x31, 0x32, 0x33, 0x34, 0x35, 0x36,
+ 0x37, 0x38, 0x39, 0x3a, 0x3b, 0x3c, 0x3d, 0x3e, 0x3f, 0x40, 0x41,
+ 0x42, 0x43, 0x44, 0x45, 0x46, 0x47, 0x48, 0x49, 0x4a, 0x4b, 0x4c,
+ 0x4d, 0x4e, 0x4f, 0x50, 0x51, 0x52, 0x53, 0x54, 0x55, 0x56, 0x57,
+ 0x58, 0x59, 0x5a, 0x5b, 0x5c, 0x5d, 0x5e, 0x5f, 0x60, 0x61, 0x62,
+ 0x63, 0x64, 0x65, 0x66, 0x67, 0x68, 0x69, 0x6a, 0x6b, 0x6c, 0x6d,
+ 0x6e, 0x6f, 0x70, 0x71, 0x72, 0x73, 0x74, 0x75, 0x76, 0x77, 0x78,
+ 0x79, 0x7a, 0x7b, 0x7c, 0x7d, 0x7e, 0x7f, 0x80, 0x81, 0x82, 0x83,
+ 0x84, 0x85, 0x86, 0x87, 0x88, 0x89, 0x8a, 0x8b, 0x8c, 0x8d, 0x8e,
+ 0x8f, 0x90, 0x91, 0x92, 0x93, 0x94, 0x95, 0x96, 0x97, 0x98, 0x99,
+ 0x9a, 0x9b, 0x9c, 0x9d, 0x9e, 0x9f, 0xa0, 0xa1, 0xa2, 0xa3, 0xa4,
+ 0xa5, 0xa6, 0xa7, 0xa8, 0xa9, 0xaa, 0xab, 0xac, 0xad, 0xae, 0xaf,
+ 0xb0, 0xb1, 0xb2, 0xb3, 0xb4, 0xb5, 0xb6, 0xb7, 0xb8, 0xb9, 0xba,
+ 0xbb, 0xbc, 0xbd, 0xbe, 0xbf, 0xc0, 0xc1, 0xc2, 0xc3, 0xc4, 0xc5,
+ 0xc6, 0xc7, 0xc8, 0xc9, 0xca, 0xcb, 0xcc, 0xcd, 0xce, 0xcf, 0xd0,
+ 0xd1, 0xd2, 0xd3, 0xd4, 0xd5, 0xd6, 0xd7, 0xd8, 0xd9, 0xda, 0xdb,
+ 0xdc, 0xdd, 0xde, 0xdf, 0xe0, 0xe1, 0xe2, 0xe3, 0xe4, 0xe5, 0xe6,
+ 0xe7, 0xe8, 0xe9, 0xea, 0xeb, 0xec, 0xed, 0xee, 0xef, 0xf0, 0xf1,
+ 0xf2, 0xf3, 0xf4, 0xf5, 0xf6, 0xf7, 0xf8, 0xf9, 0xfa, 0xfb, 0xfc,
+ 0xfd, 0xfe, 0xff
+};
+
+u_char cup2low[256] = {
+ 0x00, 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07, 0x08, 0x09, 0x0a,
+ 0x0b, 0x0c, 0x0d, 0x0e, 0x0f, 0x10, 0x11, 0x12, 0x13, 0x14, 0x15,
+ 0x16, 0x17, 0x18, 0x19, 0x1a, 0x1b, 0x1c, 0x1d, 0x1e, 0x1f, 0x20,
+ 0x21, 0x22, 0x23, 0x24, 0x25, 0x26, 0x27, 0x28, 0x29, 0x2a, 0x2b,
+ 0x2c, 0x2d, 0x2e, 0x2f, 0x30, 0x31, 0x32, 0x33, 0x34, 0x35, 0x36,
+ 0x37, 0x38, 0x39, 0x3a, 0x3b, 0x3c, 0x3d, 0x3e, 0x3f, 0x60, 0x61,
+ 0x62, 0x63, 0x64, 0x65, 0x66, 0x67, 0x68, 0x69, 0x6a, 0x6b, 0x6c,
+ 0x6d, 0x6e, 0x6f, 0x70, 0x71, 0x72, 0x73, 0x74, 0x75, 0x76, 0x77,
+ 0x78, 0x79, 0x7a, 0x7b, 0x7c, 0x7d, 0x7e, 0x7f, 0x60, 0x61, 0x62,
+ 0x63, 0x64, 0x65, 0x66, 0x67, 0x68, 0x69, 0x6a, 0x6b, 0x6c, 0x6d,
+ 0x6e, 0x6f, 0x70, 0x71, 0x72, 0x73, 0x74, 0x75, 0x76, 0x77, 0x78,
+ 0x79, 0x7a, 0x7b, 0x7c, 0x7d, 0x7e, 0x7f, 0x80, 0x81, 0x82, 0x83,
+ 0x84, 0x85, 0x86, 0x87, 0x88, 0x89, 0x8a, 0x8b, 0x8c, 0x8d, 0x8e,
+ 0x8f, 0x90, 0x91, 0x92, 0x93, 0x94, 0x95, 0x96, 0x97, 0x98, 0x99,
+ 0x9a, 0x9b, 0x9c, 0x9d, 0x9e, 0x9f, 0xa0, 0xa1, 0xa2, 0xa3, 0xa4,
+ 0xa5, 0xa6, 0xa7, 0xa8, 0xa9, 0xaa, 0xab, 0xac, 0xad, 0xae, 0xaf,
+ 0xb0, 0xb1, 0xb2, 0xb3, 0xb4, 0xb5, 0xb6, 0xb7, 0xb8, 0xb9, 0xba,
+ 0xbb, 0xbc, 0xbd, 0xbe, 0xbf, 0xc0, 0xc1, 0xc2, 0xc3, 0xc4, 0xc5,
+ 0xc6, 0xc7, 0xc8, 0xc9, 0xca, 0xcb, 0xcc, 0xcd, 0xce, 0xcf, 0xd0,
+ 0xd1, 0xd2, 0xd3, 0xd4, 0xd5, 0xd6, 0xd7, 0xd8, 0xd9, 0xda, 0xdb,
+ 0xdc, 0xdd, 0xde, 0xdf, 0xe0, 0xe1, 0xe2, 0xe3, 0xe4, 0xe5, 0xe6,
+ 0xe7, 0xe8, 0xe9, 0xea, 0xeb, 0xec, 0xed, 0xee, 0xef, 0xf0, 0xf1,
+ 0xf2, 0xf3, 0xf4, 0xf5, 0xf6, 0xf7, 0xf8, 0xf9, 0xfa, 0xfb, 0xfc,
+ 0xfd, 0xfe, 0xff
+};
+
+int
+diffreg(char *file1, char *file2, int flags)
+{
+ FILE *f1, *f2;
+ int i, rval;
+
+ f1 = f2 = NULL;
+ rval = D_SAME;
+ anychange = 0;
+ lastline = 0;
+ lastmatchline = 0;
+ context_vec_ptr = context_vec_start - 1;
+ if (flags & D_IGNORECASE)
+ chrtran = cup2low;
+ else
+ chrtran = clow2low;
+ if (S_ISDIR(stb1.st_mode) != S_ISDIR(stb2.st_mode))
+ return (S_ISDIR(stb1.st_mode) ? D_MISMATCH1 : D_MISMATCH2);
+ if (strcmp(file1, "-") == 0 && strcmp(file2, "-") == 0)
+ goto closem;
+
+ if (flags & D_EMPTY1)
+ f1 = fopen(_PATH_DEVNULL, "r");
+ else {
+ if (!S_ISREG(stb1.st_mode)) {
+ if ((f1 = opentemp(file1)) == NULL ||
+ fstat(fileno(f1), &stb1) == -1) {
+ warn("%s", file1);
+ status |= 2;
+ goto closem;
+ }
+ } else if (strcmp(file1, "-") == 0)
+ f1 = stdin;
+ else
+ f1 = fopen(file1, "r");
+ }
+ if (f1 == NULL) {
+ warn("%s", file1);
+ status |= 2;
+ goto closem;
+ }
+
+ if (flags & D_EMPTY2)
+ f2 = fopen(_PATH_DEVNULL, "r");
+ else {
+ if (!S_ISREG(stb2.st_mode)) {
+ if ((f2 = opentemp(file2)) == NULL ||
+ fstat(fileno(f2), &stb2) == -1) {
+ warn("%s", file2);
+ status |= 2;
+ goto closem;
+ }
+ } else if (strcmp(file2, "-") == 0)
+ f2 = stdin;
+ else
+ f2 = fopen(file2, "r");
+ }
+ if (f2 == NULL) {
+ warn("%s", file2);
+ status |= 2;
+ goto closem;
+ }
+
+ switch (files_differ(f1, f2, flags)) {
+ case 0:
+ goto closem;
+ case 1:
+ break;
+ default:
+ /* error */
+ status |= 2;
+ goto closem;
+ }
+
+ if ((flags & D_FORCEASCII) == 0 &&
+ (!asciifile(f1) || !asciifile(f2))) {
+ rval = D_BINARY;
+ status |= 1;
+ goto closem;
+ }
+ prepare(0, f1, stb1.st_size, flags);
+ prepare(1, f2, stb2.st_size, flags);
+
+ prune();
+ sort(sfile[0], slen[0]);
+ sort(sfile[1], slen[1]);
+
+ member = (int *)file[1];
+ equiv(sfile[0], slen[0], sfile[1], slen[1], member);
+ member = xreallocarray(member, slen[1] + 2, sizeof(*member));
+
+ class = (int *)file[0];
+ unsort(sfile[0], slen[0], class);
+ class = xreallocarray(class, slen[0] + 2, sizeof(*class));
+
+ klist = xcalloc(slen[0] + 2, sizeof(*klist));
+ clen = 0;
+ clistlen = 100;
+ clist = xcalloc(clistlen, sizeof(*clist));
+ i = stone(class, slen[0], member, klist, flags);
+ free(member);
+ free(class);
+
+ J = xreallocarray(J, len[0] + 2, sizeof(*J));
+ unravel(klist[i]);
+ free(clist);
+ free(klist);
+
+ ixold = xreallocarray(ixold, len[0] + 2, sizeof(*ixold));
+ ixnew = xreallocarray(ixnew, len[1] + 2, sizeof(*ixnew));
+ check(f1, f2, flags);
+ output(file1, f1, file2, f2, flags);
+closem:
+ if (anychange) {
+ status |= 1;
+ if (rval == D_SAME)
+ rval = D_DIFFER;
+ }
+ if (f1 != NULL)
+ fclose(f1);
+ if (f2 != NULL)
+ fclose(f2);
+
+ return (rval);
+}
+
+/*
+ * Check to see if the given files differ.
+ * Returns 0 if they are the same, 1 if different, and -1 on error.
+ * XXX - could use code from cmp(1) [faster]
+ */
+static int
+files_differ(FILE *f1, FILE *f2, int flags)
+{
+ char buf1[BUFSIZ], buf2[BUFSIZ];
+ size_t i, j;
+
+ if ((flags & (D_EMPTY1|D_EMPTY2)) || stb1.st_size != stb2.st_size ||
+ (stb1.st_mode & S_IFMT) != (stb2.st_mode & S_IFMT))
+ return (1);
+ for (;;) {
+ i = fread(buf1, 1, sizeof(buf1), f1);
+ j = fread(buf2, 1, sizeof(buf2), f2);
+ if ((!i && ferror(f1)) || (!j && ferror(f2)))
+ return (-1);
+ if (i != j)
+ return (1);
+ if (i == 0)
+ return (0);
+ if (memcmp(buf1, buf2, i) != 0)
+ return (1);
+ }
+}
+
+static FILE *
+opentemp(const char *file)
+{
+ char buf[BUFSIZ], tempfile[PATH_MAX];
+ ssize_t nread;
+ int ifd, ofd;
+
+ if (strcmp(file, "-") == 0)
+ ifd = STDIN_FILENO;
+ else if ((ifd = open(file, O_RDONLY, 0644)) == -1)
+ return (NULL);
+
+ (void)strlcpy(tempfile, _PATH_TMP "/diff.XXXXXXXX", sizeof(tempfile));
+
+ if ((ofd = mkstemp(tempfile)) == -1) {
+ close(ifd);
+ return (NULL);
+ }
+ unlink(tempfile);
+ while ((nread = read(ifd, buf, BUFSIZ)) > 0) {
+ if (write(ofd, buf, nread) != nread) {
+ close(ifd);
+ close(ofd);
+ return (NULL);
+ }
+ }
+ close(ifd);
+ lseek(ofd, (off_t)0, SEEK_SET);
+ return (fdopen(ofd, "r"));
+}
+
+char *
+splice(char *dir, char *file)
+{
+ char *tail, *buf;
+ size_t dirlen;
+
+ dirlen = strlen(dir);
+ while (dirlen != 0 && dir[dirlen - 1] == '/')
+ dirlen--;
+ if ((tail = strrchr(file, '/')) == NULL)
+ tail = file;
+ else
+ tail++;
+ xasprintf(&buf, "%.*s/%s", (int)dirlen, dir, tail);
+ return (buf);
+}
+
+static void
+prepare(int i, FILE *fd, off_t filesize, int flags)
+{
+ struct line *p;
+ int j, h;
+ size_t sz;
+
+ rewind(fd);
+
+ sz = (filesize <= SIZE_MAX ? filesize : SIZE_MAX) / 25;
+ if (sz < 100)
+ sz = 100;
+
+ p = xcalloc(sz + 3, sizeof(*p));
+ for (j = 0; (h = readhash(fd, flags));) {
+ if (j == sz) {
+ sz = sz * 3 / 2;
+ p = xreallocarray(p, sz + 3, sizeof(*p));
+ }
+ p[++j].value = h;
+ }
+ len[i] = j;
+ file[i] = p;
+}
+
+static void
+prune(void)
+{
+ int i, j;
+
+ for (pref = 0; pref < len[0] && pref < len[1] &&
+ file[0][pref + 1].value == file[1][pref + 1].value;
+ pref++)
+ ;
+ for (suff = 0; suff < len[0] - pref && suff < len[1] - pref &&
+ file[0][len[0] - suff].value == file[1][len[1] - suff].value;
+ suff++)
+ ;
+ for (j = 0; j < 2; j++) {
+ sfile[j] = file[j] + pref;
+ slen[j] = len[j] - pref - suff;
+ for (i = 0; i <= slen[j]; i++)
+ sfile[j][i].serial = i;
+ }
+}
+
+static void
+equiv(struct line *a, int n, struct line *b, int m, int *c)
+{
+ int i, j;
+
+ i = j = 1;
+ while (i <= n && j <= m) {
+ if (a[i].value < b[j].value)
+ a[i++].value = 0;
+ else if (a[i].value == b[j].value)
+ a[i++].value = j;
+ else
+ j++;
+ }
+ while (i <= n)
+ a[i++].value = 0;
+ b[m + 1].value = 0;
+ j = 0;
+ while (++j <= m) {
+ c[j] = -b[j].serial;
+ while (b[j + 1].value == b[j].value) {
+ j++;
+ c[j] = b[j].serial;
+ }
+ }
+ c[j] = -1;
+}
+
+/* Code taken from ping.c */
+static int
+isqrt(int n)
+{
+ int y, x = 1;
+
+ if (n == 0)
+ return (0);
+
+ do { /* newton was a stinker */
+ y = x;
+ x = n / x;
+ x += y;
+ x /= 2;
+ } while ((x - y) > 1 || (x - y) < -1);
+
+ return (x);
+}
+
+static int
+stone(int *a, int n, int *b, int *c, int flags)
+{
+ int i, k, y, j, l;
+ int oldc, tc, oldl, sq;
+ u_int numtries, bound;
+
+ if (flags & D_MINIMAL)
+ bound = UINT_MAX;
+ else {
+ sq = isqrt(n);
+ bound = MAXIMUM(256, sq);
+ }
+
+ k = 0;
+ c[0] = newcand(0, 0, 0);
+ for (i = 1; i <= n; i++) {
+ j = a[i];
+ if (j == 0)
+ continue;
+ y = -b[j];
+ oldl = 0;
+ oldc = c[0];
+ numtries = 0;
+ do {
+ if (y <= clist[oldc].y)
+ continue;
+ l = search(c, k, y);
+ if (l != oldl + 1)
+ oldc = c[l - 1];
+ if (l <= k) {
+ if (clist[c[l]].y <= y)
+ continue;
+ tc = c[l];
+ c[l] = newcand(i, y, oldc);
+ oldc = tc;
+ oldl = l;
+ numtries++;
+ } else {
+ c[l] = newcand(i, y, oldc);
+ k++;
+ break;
+ }
+ } while ((y = b[++j]) > 0 && numtries < bound);
+ }
+ return (k);
+}
+
+static int
+newcand(int x, int y, int pred)
+{
+ struct cand *q;
+
+ if (clen == clistlen) {
+ clistlen = clistlen * 11 / 10;
+ clist = xreallocarray(clist, clistlen, sizeof(*clist));
+ }
+ q = clist + clen;
+ q->x = x;
+ q->y = y;
+ q->pred = pred;
+ return (clen++);
+}
+
+static int
+search(int *c, int k, int y)
+{
+ int i, j, l, t;
+
+ if (clist[c[k]].y < y) /* quick look for typical case */
+ return (k + 1);
+ i = 0;
+ j = k + 1;
+ for (;;) {
+ l = (i + j) / 2;
+ if (l <= i)
+ break;
+ t = clist[c[l]].y;
+ if (t > y)
+ j = l;
+ else if (t < y)
+ i = l;
+ else
+ return (l);
+ }
+ return (l + 1);
+}
+
+static void
+unravel(int p)
+{
+ struct cand *q;
+ int i;
+
+ for (i = 0; i <= len[0]; i++)
+ J[i] = i <= pref ? i :
+ i > len[0] - suff ? i + len[1] - len[0] : 0;
+ for (q = clist + p; q->y != 0; q = clist + q->pred)
+ J[q->x + pref] = q->y + pref;
+}
+
+/*
+ * Check does double duty:
+ * 1. ferret out any fortuitous correspondences due
+ * to confounding by hashing (which result in "jackpot")
+ * 2. collect random access indexes to the two files
+ */
+static void
+check(FILE *f1, FILE *f2, int flags)
+{
+ int i, j, jackpot, c, d;
+ long ctold, ctnew;
+
+ rewind(f1);
+ rewind(f2);
+ j = 1;
+ ixold[0] = ixnew[0] = 0;
+ jackpot = 0;
+ ctold = ctnew = 0;
+ for (i = 1; i <= len[0]; i++) {
+ if (J[i] == 0) {
+ ixold[i] = ctold += skipline(f1);
+ continue;
+ }
+ while (j < J[i]) {
+ ixnew[j] = ctnew += skipline(f2);
+ j++;
+ }
+ if (flags & (D_FOLDBLANKS|D_IGNOREBLANKS|D_IGNORECASE)) {
+ for (;;) {
+ c = getc(f1);
+ d = getc(f2);
+ /*
+ * GNU diff ignores a missing newline
+ * in one file for -b or -w.
+ */
+ if (flags & (D_FOLDBLANKS|D_IGNOREBLANKS)) {
+ if (c == EOF && d == '\n') {
+ ctnew++;
+ break;
+ } else if (c == '\n' && d == EOF) {
+ ctold++;
+ break;
+ }
+ }
+ ctold++;
+ ctnew++;
+ if ((flags & D_FOLDBLANKS) && isspace(c) &&
+ isspace(d)) {
+ do {
+ if (c == '\n')
+ break;
+ ctold++;
+ } while (isspace(c = getc(f1)));
+ do {
+ if (d == '\n')
+ break;
+ ctnew++;
+ } while (isspace(d = getc(f2)));
+ } else if ((flags & D_IGNOREBLANKS)) {
+ while (isspace(c) && c != '\n') {
+ c = getc(f1);
+ ctold++;
+ }
+ while (isspace(d) && d != '\n') {
+ d = getc(f2);
+ ctnew++;
+ }
+ }
+ if (chrtran[c] != chrtran[d]) {
+ jackpot++;
+ J[i] = 0;
+ if (c != '\n' && c != EOF)
+ ctold += skipline(f1);
+ if (d != '\n' && c != EOF)
+ ctnew += skipline(f2);
+ break;
+ }
+ if (c == '\n' || c == EOF)
+ break;
+ }
+ } else {
+ for (;;) {
+ ctold++;
+ ctnew++;
+ if ((c = getc(f1)) != (d = getc(f2))) {
+ /* jackpot++; */
+ J[i] = 0;
+ if (c != '\n' && c != EOF)
+ ctold += skipline(f1);
+ if (d != '\n' && c != EOF)
+ ctnew += skipline(f2);
+ break;
+ }
+ if (c == '\n' || c == EOF)
+ break;
+ }
+ }
+ ixold[i] = ctold;
+ ixnew[j] = ctnew;
+ j++;
+ }
+ for (; j <= len[1]; j++)
+ ixnew[j] = ctnew += skipline(f2);
+ /*
+ * if (jackpot)
+ * fprintf(stderr, "jackpot\n");
+ */
+}
+
+/* shellsort CACM #201 */
+static void
+sort(struct line *a, int n)
+{
+ struct line *ai, *aim, w;
+ int j, m = 0, k;
+
+ if (n == 0)
+ return;
+ for (j = 1; j <= n; j *= 2)
+ m = 2 * j - 1;
+ for (m /= 2; m != 0; m /= 2) {
+ k = n - m;
+ for (j = 1; j <= k; j++) {
+ for (ai = &a[j]; ai > a; ai -= m) {
+ aim = &ai[m];
+ if (aim < ai)
+ break; /* wraparound */
+ if (aim->value > ai[0].value ||
+ (aim->value == ai[0].value &&
+ aim->serial > ai[0].serial))
+ break;
+ w.value = ai[0].value;
+ ai[0].value = aim->value;
+ aim->value = w.value;
+ w.serial = ai[0].serial;
+ ai[0].serial = aim->serial;
+ aim->serial = w.serial;
+ }
+ }
+ }
+}
+
+static void
+unsort(struct line *f, int l, int *b)
+{
+ int *a, i;
+
+ a = xcalloc(l + 1, sizeof(*a));
+ for (i = 1; i <= l; i++)
+ a[f[i].serial] = f[i].value;
+ for (i = 1; i <= l; i++)
+ b[i] = a[i];
+ free(a);
+}
+
+static int
+skipline(FILE *f)
+{
+ int i, c;
+
+ for (i = 1; (c = getc(f)) != '\n' && c != EOF; i++)
+ continue;
+ return (i);
+}
+
+static void
+output(char *file1, FILE *f1, char *file2, FILE *f2, int flags)
+{
+ int m, i0, i1, j0, j1;
+
+ rewind(f1);
+ rewind(f2);
+ m = len[0];
+ J[0] = 0;
+ J[m + 1] = len[1] + 1;
+ if (diff_format != D_EDIT) {
+ for (i0 = 1; i0 <= m; i0 = i1 + 1) {
+ while (i0 <= m && J[i0] == J[i0 - 1] + 1)
+ i0++;
+ j0 = J[i0 - 1] + 1;
+ i1 = i0 - 1;
+ while (i1 < m && J[i1 + 1] == 0)
+ i1++;
+ j1 = J[i1 + 1] - 1;
+ J[i1] = j1;
+ change(file1, f1, file2, f2, i0, i1, j0, j1, &flags);
+ }
+ } else {
+ for (i0 = m; i0 >= 1; i0 = i1 - 1) {
+ while (i0 >= 1 && J[i0] == J[i0 + 1] - 1 && J[i0] != 0)
+ i0--;
+ j0 = J[i0 + 1] - 1;
+ i1 = i0 + 1;
+ while (i1 > 1 && J[i1 - 1] == 0)
+ i1--;
+ j1 = J[i1 - 1] + 1;
+ J[i1] = j1;
+ change(file1, f1, file2, f2, i1, i0, j1, j0, &flags);
+ }
+ }
+ if (m == 0)
+ change(file1, f1, file2, f2, 1, 0, 1, len[1], &flags);
+ if (diff_format == D_IFDEF) {
+ for (;;) {
+#define c i0
+ if ((c = getc(f1)) == EOF)
+ return;
+ diff_output("%c", c);
+ }
+#undef c
+ }
+ if (anychange != 0) {
+ if (diff_format == D_CONTEXT)
+ dump_context_vec(f1, f2, flags);
+ else if (diff_format == D_UNIFIED)
+ dump_unified_vec(f1, f2, flags);
+ }
+}
+
+static void
+range(int a, int b, char *separator)
+{
+ diff_output("%d", a > b ? b : a);
+ if (a < b)
+ diff_output("%s%d", separator, b);
+}
+
+static void
+uni_range(int a, int b)
+{
+ if (a < b)
+ diff_output("%d,%d", a, b - a + 1);
+ else if (a == b)
+ diff_output("%d", b);
+ else
+ diff_output("%d,0", b);
+}
+
+static char *
+preadline(int fd, size_t rlen, off_t off)
+{
+ char *line;
+ ssize_t nr;
+
+ line = xmalloc(rlen + 1);
+ if ((nr = pread(fd, line, rlen, off)) == -1)
+ err(2, "preadline");
+ if (nr > 0 && line[nr-1] == '\n')
+ nr--;
+ line[nr] = '\0';
+ return (line);
+}
+
+static int
+ignoreline(char *line)
+{
+ int ret;
+
+ ret = regexec(&ignore_re, line, 0, NULL, 0);
+ free(line);
+ return (ret == 0); /* if it matched, it should be ignored. */
+}
+
+/*
+ * Indicate that there is a difference between lines a and b of the from file
+ * to get to lines c to d of the to file. If a is greater then b then there
+ * are no lines in the from file involved and this means that there were
+ * lines appended (beginning at b). If c is greater than d then there are
+ * lines missing from the to file.
+ */
+static void
+change(char *file1, FILE *f1, char *file2, FILE *f2, int a, int b, int c, int d,
+ int *pflags)
+{
+ static size_t max_context = 64;
+ int i;
+
+restart:
+ if (diff_format != D_IFDEF && a > b && c > d)
+ return;
+ if (ignore_pats != NULL) {
+ char *line;
+ /*
+ * All lines in the change, insert, or delete must
+ * match an ignore pattern for the change to be
+ * ignored.
+ */
+ if (a <= b) { /* Changes and deletes. */
+ for (i = a; i <= b; i++) {
+ line = preadline(fileno(f1),
+ ixold[i] - ixold[i - 1], ixold[i - 1]);
+ if (!ignoreline(line))
+ goto proceed;
+ }
+ }
+ if (a > b || c <= d) { /* Changes and inserts. */
+ for (i = c; i <= d; i++) {
+ line = preadline(fileno(f2),
+ ixnew[i] - ixnew[i - 1], ixnew[i - 1]);
+ if (!ignoreline(line))
+ goto proceed;
+ }
+ }
+ return;
+ }
+proceed:
+ if (*pflags & D_HEADER) {
+ diff_output("%s %s %s\n", diffargs, file1, file2);
+ *pflags &= ~D_HEADER;
+ }
+ if (diff_format == D_CONTEXT || diff_format == D_UNIFIED) {
+ /*
+ * Allocate change records as needed.
+ */
+ if (context_vec_ptr == context_vec_end - 1) {
+ ptrdiff_t offset = context_vec_ptr - context_vec_start;
+ max_context <<= 1;
+ context_vec_start = xreallocarray(context_vec_start,
+ max_context, sizeof(*context_vec_start));
+ context_vec_end = context_vec_start + max_context;
+ context_vec_ptr = context_vec_start + offset;
+ }
+ if (anychange == 0) {
+ /*
+ * Print the context/unidiff header first time through.
+ */
+ print_header(file1, file2);
+ anychange = 1;
+ } else if (a > context_vec_ptr->b + (2 * diff_context) + 1 &&
+ c > context_vec_ptr->d + (2 * diff_context) + 1) {
+ /*
+ * If this change is more than 'diff_context' lines from the
+ * previous change, dump the record and reset it.
+ */
+ if (diff_format == D_CONTEXT)
+ dump_context_vec(f1, f2, *pflags);
+ else
+ dump_unified_vec(f1, f2, *pflags);
+ }
+ context_vec_ptr++;
+ context_vec_ptr->a = a;
+ context_vec_ptr->b = b;
+ context_vec_ptr->c = c;
+ context_vec_ptr->d = d;
+ return;
+ }
+ if (anychange == 0)
+ anychange = 1;
+ switch (diff_format) {
+ case D_BRIEF:
+ return;
+ case D_NORMAL:
+ case D_EDIT:
+ range(a, b, ",");
+ diff_output("%c", a > b ? 'a' : c > d ? 'd' : 'c');
+ if (diff_format == D_NORMAL)
+ range(c, d, ",");
+ diff_output("\n");
+ break;
+ case D_REVERSE:
+ diff_output("%c", a > b ? 'a' : c > d ? 'd' : 'c');
+ range(a, b, " ");
+ diff_output("\n");
+ break;
+ case D_NREVERSE:
+ if (a > b)
+ diff_output("a%d %d\n", b, d - c + 1);
+ else {
+ diff_output("d%d %d\n", a, b - a + 1);
+ if (!(c > d))
+ /* add changed lines */
+ diff_output("a%d %d\n", b, d - c + 1);
+ }
+ break;
+ }
+ if (diff_format == D_NORMAL || diff_format == D_IFDEF) {
+ fetch(ixold, a, b, f1, '<', 1, *pflags);
+ if (a <= b && c <= d && diff_format == D_NORMAL)
+ diff_output("---\n");
+ }
+ i = fetch(ixnew, c, d, f2, diff_format == D_NORMAL ? '>' : '\0', 0, *pflags);
+ if (i != 0 && diff_format == D_EDIT) {
+ /*
+ * A non-zero return value for D_EDIT indicates that the
+ * last line printed was a bare dot (".") that has been
+ * escaped as ".." to prevent ed(1) from misinterpreting
+ * it. We have to add a substitute command to change this
+ * back and restart where we left off.
+ */
+ diff_output(".\n");
+ diff_output("%ds/.//\n", a + i - 1);
+ b = a + i - 1;
+ a = b + 1;
+ c += i;
+ goto restart;
+ }
+ if ((diff_format == D_EDIT || diff_format == D_REVERSE) && c <= d)
+ diff_output(".\n");
+ if (inifdef) {
+ diff_output("#endif /* %s */\n", ifdefname);
+ inifdef = 0;
+ }
+}
+
+static int
+fetch(long *f, int a, int b, FILE *lb, int ch, int oldfile, int flags)
+{
+ int i, j, c, lastc, col, nc;
+
+ /*
+ * When doing #ifdef's, copy down to current line
+ * if this is the first file, so that stuff makes it to output.
+ */
+ if (diff_format == D_IFDEF && oldfile) {
+ long curpos = ftell(lb);
+ /* print through if append (a>b), else to (nb: 0 vs 1 orig) */
+ nc = f[a > b ? b : a - 1] - curpos;
+ for (i = 0; i < nc; i++)
+ diff_output("%c", getc(lb));
+ }
+ if (a > b)
+ return (0);
+ if (diff_format == D_IFDEF) {
+ if (inifdef) {
+ diff_output("#else /* %s%s */\n",
+ oldfile == 1 ? "!" : "", ifdefname);
+ } else {
+ if (oldfile)
+ diff_output("#ifndef %s\n", ifdefname);
+ else
+ diff_output("#ifdef %s\n", ifdefname);
+ }
+ inifdef = 1 + oldfile;
+ }
+ for (i = a; i <= b; i++) {
+ fseek(lb, f[i - 1], SEEK_SET);
+ nc = f[i] - f[i - 1];
+ if (diff_format != D_IFDEF && ch != '\0') {
+ diff_output("%c", ch);
+ if (Tflag && (diff_format == D_NORMAL || diff_format == D_CONTEXT
+ || diff_format == D_UNIFIED))
+ diff_output("\t");
+ else if (diff_format != D_UNIFIED)
+ diff_output(" ");
+ }
+ col = 0;
+ for (j = 0, lastc = '\0'; j < nc; j++, lastc = c) {
+ if ((c = getc(lb)) == EOF) {
+ if (diff_format == D_EDIT || diff_format == D_REVERSE ||
+ diff_format == D_NREVERSE)
+ warnx("No newline at end of file");
+ else
+ diff_output("\n\\ No newline at end of "
+ "file\n");
+ return (0);
+ }
+ if (c == '\t' && (flags & D_EXPANDTABS)) {
+ do {
+ diff_output(" ");
+ } while (++col & 7);
+ } else {
+ if (diff_format == D_EDIT && j == 1 && c == '\n'
+ && lastc == '.') {
+ /*
+ * Don't print a bare "." line
+ * since that will confuse ed(1).
+ * Print ".." instead and return,
+ * giving the caller an offset
+ * from which to restart.
+ */
+ diff_output(".\n");
+ return (i - a + 1);
+ }
+ diff_output("%c", c);
+ col++;
+ }
+ }
+ }
+ return (0);
+}
+
+/*
+ * Hash function taken from Robert Sedgewick, Algorithms in C, 3d ed., p 578.
+ */
+static int
+readhash(FILE *f, int flags)
+{
+ int i, t, space;
+ int sum;
+
+ sum = 1;
+ space = 0;
+ if ((flags & (D_FOLDBLANKS|D_IGNOREBLANKS)) == 0) {
+ if (flags & D_IGNORECASE)
+ for (i = 0; (t = getc(f)) != '\n'; i++) {
+ if (t == EOF) {
+ if (i == 0)
+ return (0);
+ break;
+ }
+ sum = sum * 127 + chrtran[t];
+ }
+ else
+ for (i = 0; (t = getc(f)) != '\n'; i++) {
+ if (t == EOF) {
+ if (i == 0)
+ return (0);
+ break;
+ }
+ sum = sum * 127 + t;
+ }
+ } else {
+ for (i = 0;;) {
+ switch (t = getc(f)) {
+ case '\t':
+ case '\r':
+ case '\v':
+ case '\f':
+ case ' ':
+ space++;
+ continue;
+ default:
+ if (space && (flags & D_IGNOREBLANKS) == 0) {
+ i++;
+ space = 0;
+ }
+ sum = sum * 127 + chrtran[t];
+ i++;
+ continue;
+ case EOF:
+ if (i == 0)
+ return (0);
+ /* FALLTHROUGH */
+ case '\n':
+ break;
+ }
+ break;
+ }
+ }
+ /*
+ * There is a remote possibility that we end up with a zero sum.
+ * Zero is used as an EOF marker, so return 1 instead.
+ */
+ return (sum == 0 ? 1 : sum);
+}
+
+static int
+asciifile(FILE *f)
+{
+ unsigned char buf[BUFSIZ];
+ size_t cnt;
+
+ if (f == NULL)
+ return (1);
+
+ rewind(f);
+ cnt = fread(buf, 1, sizeof(buf), f);
+ return (memchr(buf, '\0', cnt) == NULL);
+}
+
+#define begins_with(s, pre) (strncmp(s, pre, sizeof(pre)-1) == 0)
+
+static char *
+match_function(const long *f, int pos, FILE *fp)
+{
+ unsigned char buf[FUNCTION_CONTEXT_SIZE];
+ size_t nc;
+ int last = lastline;
+ char *state = NULL;
+
+ lastline = pos;
+ while (pos > last) {
+ fseek(fp, f[pos - 1], SEEK_SET);
+ nc = f[pos] - f[pos - 1];
+ if (nc >= sizeof(buf))
+ nc = sizeof(buf) - 1;
+ nc = fread(buf, 1, nc, fp);
+ if (nc > 0) {
+ buf[nc] = '\0';
+ buf[strcspn(buf, "\n")] = '\0';
+ if (isalpha(buf[0]) || buf[0] == '_' || buf[0] == '$') {
+ if (begins_with(buf, "private:")) {
+ if (!state)
+ state = " (private)";
+ } else if (begins_with(buf, "protected:")) {
+ if (!state)
+ state = " (protected)";
+ } else if (begins_with(buf, "public:")) {
+ if (!state)
+ state = " (public)";
+ } else {
+ strlcpy(lastbuf, buf, sizeof lastbuf);
+ if (state)
+ strlcat(lastbuf, state,
+ sizeof lastbuf);
+ lastmatchline = pos;
+ return lastbuf;
+ }
+ }
+ }
+ pos--;
+ }
+ return lastmatchline > 0 ? lastbuf : NULL;
+}
+
+/* dump accumulated "context" diff changes */
+static void
+dump_context_vec(FILE *f1, FILE *f2, int flags)
+{
+ struct context_vec *cvp = context_vec_start;
+ int lowa, upb, lowc, upd, do_output;
+ int a, b, c, d;
+ char ch, *f;
+
+ if (context_vec_start > context_vec_ptr)
+ return;
+
+ b = d = 0; /* gcc */
+ lowa = MAXIMUM(1, cvp->a - diff_context);
+ upb = MINIMUM(len[0], context_vec_ptr->b + diff_context);
+ lowc = MAXIMUM(1, cvp->c - diff_context);
+ upd = MINIMUM(len[1], context_vec_ptr->d + diff_context);
+
+ diff_output("***************");
+ if ((flags & D_PROTOTYPE)) {
+ f = match_function(ixold, lowa-1, f1);
+ if (f != NULL)
+ diff_output(" %s", f);
+ }
+ diff_output("\n*** ");
+ range(lowa, upb, ",");
+ diff_output(" ****\n");
+
+ /*
+ * Output changes to the "old" file. The first loop suppresses
+ * output if there were no changes to the "old" file (we'll see
+ * the "old" lines as context in the "new" list).
+ */
+ do_output = 0;
+ for (; cvp <= context_vec_ptr; cvp++)
+ if (cvp->a <= cvp->b) {
+ cvp = context_vec_start;
+ do_output++;
+ break;
+ }
+ if (do_output) {
+ while (cvp <= context_vec_ptr) {
+ a = cvp->a;
+ b = cvp->b;
+ c = cvp->c;
+ d = cvp->d;
+
+ if (a <= b && c <= d)
+ ch = 'c';
+ else
+ ch = (a <= b) ? 'd' : 'a';
+
+ if (ch == 'a')
+ fetch(ixold, lowa, b, f1, ' ', 0, flags);
+ else {
+ fetch(ixold, lowa, a - 1, f1, ' ', 0, flags);
+ fetch(ixold, a, b, f1,
+ ch == 'c' ? '!' : '-', 0, flags);
+ }
+ lowa = b + 1;
+ cvp++;
+ }
+ fetch(ixold, b + 1, upb, f1, ' ', 0, flags);
+ }
+ /* output changes to the "new" file */
+ diff_output("--- ");
+ range(lowc, upd, ",");
+ diff_output(" ----\n");
+
+ do_output = 0;
+ for (cvp = context_vec_start; cvp <= context_vec_ptr; cvp++)
+ if (cvp->c <= cvp->d) {
+ cvp = context_vec_start;
+ do_output++;
+ break;
+ }
+ if (do_output) {
+ while (cvp <= context_vec_ptr) {
+ a = cvp->a;
+ b = cvp->b;
+ c = cvp->c;
+ d = cvp->d;
+
+ if (a <= b && c <= d)
+ ch = 'c';
+ else
+ ch = (a <= b) ? 'd' : 'a';
+
+ if (ch == 'd')
+ fetch(ixnew, lowc, d, f2, ' ', 0, flags);
+ else {
+ fetch(ixnew, lowc, c - 1, f2, ' ', 0, flags);
+ fetch(ixnew, c, d, f2,
+ ch == 'c' ? '!' : '+', 0, flags);
+ }
+ lowc = d + 1;
+ cvp++;
+ }
+ fetch(ixnew, d + 1, upd, f2, ' ', 0, flags);
+ }
+ context_vec_ptr = context_vec_start - 1;
+}
+
+/* dump accumulated "unified" diff changes */
+static void
+dump_unified_vec(FILE *f1, FILE *f2, int flags)
+{
+ struct context_vec *cvp = context_vec_start;
+ int lowa, upb, lowc, upd;
+ int a, b, c, d;
+ char ch, *f;
+
+ if (context_vec_start > context_vec_ptr)
+ return;
+
+ b = d = 0; /* gcc */
+ lowa = MAXIMUM(1, cvp->a - diff_context);
+ upb = MINIMUM(len[0], context_vec_ptr->b + diff_context);
+ lowc = MAXIMUM(1, cvp->c - diff_context);
+ upd = MINIMUM(len[1], context_vec_ptr->d + diff_context);
+
+ diff_output("@@ -");
+ uni_range(lowa, upb);
+ diff_output(" +");
+ uni_range(lowc, upd);
+ diff_output(" @@");
+ if ((flags & D_PROTOTYPE)) {
+ f = match_function(ixold, lowa-1, f1);
+ if (f != NULL)
+ diff_output(" %s", f);
+ }
+ diff_output("\n");
+
+ /*
+ * Output changes in "unified" diff format--the old and new lines
+ * are printed together.
+ */
+ for (; cvp <= context_vec_ptr; cvp++) {
+ a = cvp->a;
+ b = cvp->b;
+ c = cvp->c;
+ d = cvp->d;
+
+ /*
+ * c: both new and old changes
+ * d: only changes in the old file
+ * a: only changes in the new file
+ */
+ if (a <= b && c <= d)
+ ch = 'c';
+ else
+ ch = (a <= b) ? 'd' : 'a';
+
+ switch (ch) {
+ case 'c':
+ fetch(ixold, lowa, a - 1, f1, ' ', 0, flags);
+ fetch(ixold, a, b, f1, '-', 0, flags);
+ fetch(ixnew, c, d, f2, '+', 0, flags);
+ break;
+ case 'd':
+ fetch(ixold, lowa, a - 1, f1, ' ', 0, flags);
+ fetch(ixold, a, b, f1, '-', 0, flags);
+ break;
+ case 'a':
+ fetch(ixnew, lowc, c - 1, f2, ' ', 0, flags);
+ fetch(ixnew, c, d, f2, '+', 0, flags);
+ break;
+ }
+ lowa = b + 1;
+ lowc = d + 1;
+ }
+ fetch(ixnew, d + 1, upd, f2, ' ', 0, flags);
+
+ context_vec_ptr = context_vec_start - 1;
+}
+
+static void
+print_header(const char *file1, const char *file2)
+{
+ if (label[0] != NULL)
+ diff_output("%s %s\n", diff_format == D_CONTEXT ? "***" : "---",
+ label[0]);
+ else
+ diff_output("%s %s\t%s", diff_format == D_CONTEXT ? "***" : "---",
+ file1, ctime(&stb1.st_mtime));
+ if (label[1] != NULL)
+ diff_output("%s %s\n", diff_format == D_CONTEXT ? "---" : "+++",
+ label[1]);
+ else
+ diff_output("%s %s\t%s", diff_format == D_CONTEXT ? "---" : "+++",
+ file2, ctime(&stb2.st_mtime));
+}
diff --git a/usr.bin/diff/xmalloc.c b/usr.bin/diff/xmalloc.c
new file mode 100644
index 0000000..ce0f454
--- /dev/null
+++ b/usr.bin/diff/xmalloc.c
@@ -0,0 +1,85 @@
+/* $OpenBSD: xmalloc.c,v 1.10 2019/06/28 05:44:09 deraadt Exp $ */
+/*
+ * Author: Tatu Ylonen <ylo@cs.hut.fi>
+ * Copyright (c) 1995 Tatu Ylonen <ylo@cs.hut.fi>, Espoo, Finland
+ * All rights reserved
+ * Versions of malloc and friends that check their results, and never return
+ * failure (they call fatal if they encounter an error).
+ *
+ * As far as I am concerned, the code I have written for this software
+ * can be used freely for any purpose. Any derived versions of this
+ * software must be clearly marked as such, and if the derived work is
+ * incompatible with the protocol description in the RFC file, it must be
+ * called by a name other than "ssh" or "Secure Shell".
+ */
+
+#include <err.h>
+#include <stdarg.h>
+#include <stdint.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+
+#include "xmalloc.h"
+
+void *
+xmalloc(size_t size)
+{
+ void *ptr;
+
+ if (size == 0)
+ errx(2, "xmalloc: zero size");
+ ptr = malloc(size);
+ if (ptr == NULL)
+ err(2, "xmalloc: allocating %zu bytes", size);
+ return ptr;
+}
+
+void *
+xcalloc(size_t nmemb, size_t size)
+{
+ void *ptr;
+
+ ptr = calloc(nmemb, size);
+ if (ptr == NULL)
+ err(2, "xcalloc: allocating %zu * %zu bytes", nmemb, size);
+ return ptr;
+}
+
+void *
+xreallocarray(void *ptr, size_t nmemb, size_t size)
+{
+ void *new_ptr;
+
+ new_ptr = reallocarray(ptr, nmemb, size);
+ if (new_ptr == NULL)
+ err(2, "xreallocarray: allocating %zu * %zu bytes",
+ nmemb, size);
+ return new_ptr;
+}
+
+char *
+xstrdup(const char *str)
+{
+ char *cp;
+
+ if ((cp = strdup(str)) == NULL)
+ err(2, "xstrdup");
+ return cp;
+}
+
+int
+xasprintf(char **ret, const char *fmt, ...)
+{
+ va_list ap;
+ int i;
+
+ va_start(ap, fmt);
+ i = vasprintf(ret, fmt, ap);
+ va_end(ap);
+
+ if (i == -1)
+ err(2, "xasprintf");
+
+ return i;
+}
diff --git a/usr.bin/diff/xmalloc.h b/usr.bin/diff/xmalloc.h
new file mode 100644
index 0000000..2139671
--- /dev/null
+++ b/usr.bin/diff/xmalloc.h
@@ -0,0 +1,30 @@
+/* $OpenBSD: xmalloc.h,v 1.4 2015/11/12 16:30:30 mmcc Exp $ */
+
+/*
+ * Author: Tatu Ylonen <ylo@cs.hut.fi>
+ * Copyright (c) 1995 Tatu Ylonen <ylo@cs.hut.fi>, Espoo, Finland
+ * All rights reserved
+ * Created: Mon Mar 20 22:09:17 1995 ylo
+ *
+ * Versions of malloc and friends that check their results, and never return
+ * failure (they call fatal if they encounter an error).
+ *
+ * As far as I am concerned, the code I have written for this software
+ * can be used freely for any purpose. Any derived versions of this
+ * software must be clearly marked as such, and if the derived work is
+ * incompatible with the protocol description in the RFC file, it must be
+ * called by a name other than "ssh" or "Secure Shell".
+ */
+
+#ifndef XMALLOC_H
+#define XMALLOC_H
+
+void *xmalloc(size_t);
+void *xcalloc(size_t, size_t);
+void *xreallocarray(void *, size_t, size_t);
+char *xstrdup(const char *);
+int xasprintf(char **, const char *, ...)
+ __attribute__((__format__ (printf, 2, 3)))
+ __attribute__((__nonnull__ (2)));
+
+#endif /* XMALLOC_H */