Implement -I option: ignore changes matching a set of regexes. From

Jared Yanovich, with twists from millert@ and me. Testing by brad@,
sturm@ and pval@. ok millert@

4 files changed, 144 insertions, 28 deletions

diff --git a/usr.bin/diff/diff.1 b/usr.bin/diff/diff.1
index 633c2f435c0..48668dfd229 100644
--- a/usr.bin/diff/diff.1
+++ b/usr.bin/diff/diff.1
@@ -1,4 +1,4 @@
-.\" $OpenBSD: diff.1,v 1.25 2004/03/16 00:40:34 millert Exp $
+.\" $OpenBSD: diff.1,v 1.26 2004/06/20 18:47:45 otto Exp $
 .\"
 .\" Copyright (c) 1980, 1990, 1993
 .\"	The Regents of the University of California.  All rights reserved.
@@ -38,31 +38,36 @@
 .Sh SYNOPSIS
 .Nm diff
 .Op Fl abdilpqtTw
+.Op Fl I Ar pattern
 .Oo
-.Fl c | Fl e | Fl f |
-.Fl n | Fl u
+.Fl c | e | f |
+.Fl n | u
 .Oc
 .Op Fl L Ar label
 .Ar file1 file2
 .Nm diff
 .Op Fl abdilpqtTw
+.Op Fl I Ar pattern
 .Op Fl L Ar label
 .Fl C Ar number
 .Ar file1 file2
 .Nm diff
 .Op Fl abdilqtw
+.Op Fl I Ar pattern
 .Fl D Ar string
 .Ar file1 file2
 .Nm diff
 .Op Fl abdilpqtTw
+.Op Fl I Ar pattern
 .Op Fl L Ar label
 .Fl U Ar number
 .Ar file1 file2
 .Nm diff
 .Op Fl abdilNPpqtTw
+.Op Fl I Ar pattern
 .Oo
-.Fl c | Fl e | Fl f |
-.Fl n | Fl u
+.Fl c | e | f |
+.Fl n | u
 .Oc
 .Bk -words
 .Op Fl L Ar label
@@ -93,11 +98,12 @@ With
 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 *'s.
+by a line with fifteen
+.Li * Ns 's .
 The lines removed from
 .Ar file1
 are marked with
-.Sq \-\ \& ;
+.Sq \&-\ \& ;
 those added to
 .Ar file2
 are marked
@@ -179,7 +185,8 @@ lines of context.
 Comparison options:
 .Bl -tag -width Ds
 .It Fl a
-Treat all files as ASCII.
+Treat all files as
+.Tn ASCII .
 .It Fl b
 Causes trailing blanks (spaces and tabs) to be ignored, and other
 strings of blanks to compare equal.
@@ -187,6 +194,18 @@ strings of blanks to compare equal.
 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.,
@@ -285,8 +304,9 @@ 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.
+If a non-regular file such as a device special file or
+.Tn FIFO
+is encountered, a diagnostic message is printed.
 .Pp
 If only one of
 .Ar file1
@@ -372,11 +392,11 @@ 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 YY
+.Va XX , Ns Va YY
 from
 .Ar file1
 with the range
-.Va ZZ , Ns QQ
+.Va ZZ , Ns Va QQ
 from
 .Ar file2 .
 .El
@@ -418,7 +438,7 @@ as the temporary directory.
 .El
 .Sh FILES
 .Bl -tag -width /tmp/diff.XXXXXXXX -compact
-.It Pa /tmp/diff.XXXXXXXX
+.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.
@@ -442,7 +462,8 @@ An error occurred.
 .Xr diff3 1 ,
 .Xr ed 1 ,
 .Xr pr 1 ,
-.Xr fnmatch 3
+.Xr fnmatch 3 ,
+.Xr re_format 7
 .Sh STANDARDS
 The
 .Nm
diff --git a/usr.bin/diff/diff.c b/usr.bin/diff/diff.c
index 134cb37c0f3..b32ee897861 100644
--- a/usr.bin/diff/diff.c
+++ b/usr.bin/diff/diff.c
@@ -1,4 +1,4 @@
-/*	$OpenBSD: diff.c,v 1.45 2004/03/16 00:40:34 millert Exp $	*/
+/*	$OpenBSD: diff.c,v 1.46 2004/06/20 18:47:45 otto Exp $	*/
 
 /*
  * Copyright (c) 2003 Todd C. Miller <Todd.Miller@courtesan.com>
@@ -21,7 +21,7 @@
  */
 
 #ifndef lint
-static const char rcsid[] = "$OpenBSD: diff.c,v 1.45 2004/03/16 00:40:34 millert Exp $";
+static const char rcsid[] = "$OpenBSD: diff.c,v 1.46 2004/06/20 18:47:45 otto Exp $";
 #endif /* not lint */
 
 #include <sys/param.h>
@@ -43,11 +43,12 @@ static const char rcsid[] = "$OpenBSD: diff.c,v 1.45 2004/03/16 00:40:34 millert
 int	 aflag, bflag, dflag, iflag, lflag, Nflag, Pflag, pflag, rflag;
 int	 sflag, tflag, Tflag, wflag;
 int	 format, context, status;
-char	*start, *ifdefname, *diffargs, *label;
+char	*start, *ifdefname, *diffargs, *label, *ignore_pats;
 struct stat stb1, stb2;
 struct excludes *excludes_list;
+regex_t	 ignore_re;
 
-#define	OPTIONS	"0123456789abC:cdD:efhiL:lnNPpqrS:sTtU:uwX:x:"
+#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' },
@@ -56,6 +57,7 @@ static struct option longopts[] = {
 	{ "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' },
 	{ "paginate",			no_argument,		0,	'l' },
 	{ "label",			required_argument,	0,	'L' },
@@ -78,6 +80,7 @@ static struct option longopts[] = {
 
 __dead void usage(void);
 void push_excludes(char *);
+void push_ignore_pats(char *);
 void read_excludes_file(char *file);
 void set_argstr(char **, char **);
 
@@ -139,6 +142,9 @@ main(int argc, char **argv)
 		case 'h':
 			/* silently ignore for backwards compatibility */
 			break;
+		case 'I':
+			push_ignore_pats(optarg);
+			break;
 		case 'i':
 			iflag = 1;
 			break;
@@ -216,6 +222,19 @@ main(int argc, char **argv)
 	 */
 	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;
@@ -343,6 +362,25 @@ push_excludes(char *pattern)
 }
 
 void
+push_ignore_pats(char *pattern)
+{
+	size_t len;
+
+	if (ignore_pats == NULL) {
+		/* XXX: estrdup */
+		len = strlen(pattern) + 1;
+		ignore_pats = emalloc(len);
+		strlcpy(ignore_pats, pattern, len);
+	} else {
+		/* old + "|" + new + NUL */
+		len = strlen(ignore_pats) + strlen(pattern) + 2;
+		ignore_pats = erealloc(ignore_pats, 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)
@@ -400,12 +438,14 @@ __dead void
 usage(void)
 {
 	(void)fprintf(stderr,
-	    "usage: diff [-abdilpqtTw] [-c | -e | -f | -n | -u] [-L label] file1 file2\n"
-	    "       diff [-abdilpqtTw] [-L label] -C number file1 file2\n"
-	    "       diff [-abdilqtw] -D string file1 file2\n"
-	    "       diff [-abdilpqtTw] [-L label] -U number file1 file2\n"
-	    "       diff [-abdilNPpqtTw] [-c | -e | -f | -n | -u ] [-L label] [-r] [-s]\n"
-	    "            [-S name] [-X file] [-x pattern] dir1 dir2\n");
+	    "usage: diff [-abdilpqtTw] [-I pattern] [-c | -e | -f | -n | -u]\n"
+	    "            [-L label] file1 file2\n"
+	    "       diff [-abdilpqtTw] [-I pattern] [-L label] -C number file1 file2\n"
+	    "       diff [-abdilqtw] [-I pattern] -D string file1 file2\n"
+	    "       diff [-abdilpqtTw] [-I pattern] [-L label] -U number file1 file2\n"
+	    "       diff [-abdilNPpqtTw] [-I pattern] [-c | -e | -f | -n | -u]\n"
+	    "            [-L label] [-r] [-s] [-S name] [-X file] [-x pattern] dir1\n"
+	    "            dir2\n");
 
 	exit(2);
 }
diff --git a/usr.bin/diff/diff.h b/usr.bin/diff/diff.h
index d9af1a94c0e..8d5e70813dd 100644
--- a/usr.bin/diff/diff.h
+++ b/usr.bin/diff/diff.h
@@ -1,4 +1,4 @@
-/*	$OpenBSD: diff.h,v 1.27 2004/03/16 00:40:34 millert Exp $	*/
+/*	$OpenBSD: diff.h,v 1.28 2004/06/20 18:47:45 otto Exp $	*/
 
 /*-
  * Copyright (c) 1991, 1993
@@ -31,6 +31,9 @@
  *	@(#)diff.h	8.1 (Berkeley) 6/6/93
  */
 
+#include <sys/types.h>
+#include <regex.h>
+
 /*
  * Output format options
  */
@@ -73,9 +76,10 @@ struct excludes {
 extern int	aflag, bflag, dflag, iflag, lflag, Nflag, Pflag, pflag, rflag,
 		sflag, tflag, Tflag, wflag;
 extern int	format, context, status;
-extern char	*start, *ifdefname, *diffargs, *label;
+extern char	*start, *ifdefname, *diffargs, *label, *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);
diff --git a/usr.bin/diff/diffreg.c b/usr.bin/diff/diffreg.c
index 6c5b36ce3ae..cccd4305353 100644
--- a/usr.bin/diff/diffreg.c
+++ b/usr.bin/diff/diffreg.c
@@ -1,4 +1,4 @@
-/*	$OpenBSD: diffreg.c,v 1.56 2004/06/18 07:23:50 otto Exp $	*/
+/*	$OpenBSD: diffreg.c,v 1.57 2004/06/20 18:47:45 otto Exp $	*/
 
 /*
  * Copyright (C) Caldera International Inc.  2001-2002.
@@ -65,7 +65,7 @@
  */
 
 #ifndef lint
-static const char rcsid[] = "$OpenBSD: diffreg.c,v 1.56 2004/06/18 07:23:50 otto Exp $";
+static const char rcsid[] = "$OpenBSD: diffreg.c,v 1.57 2004/06/20 18:47:45 otto Exp $";
 #endif /* not lint */
 
 #include <sys/param.h>
@@ -214,6 +214,7 @@ static void unravel(int);
 static void unsort(struct line *, int, int *);
 static void change(char *, FILE *, char *, FILE *, int, int, int, int);
 static void sort(struct line *, int);
+static int  ignoreline(char *);
 static int  asciifile(FILE *);
 static int  fetch(long *, int, int, FILE *, int, int);
 static int  newcand(int, int, int);
@@ -226,6 +227,7 @@ static int  files_differ(FILE *, FILE *, int);
 static __inline int min(int, int);
 static __inline int max(int, int);
 static char *match_function(const long *, int, FILE *);
+static char *preadline(int, size_t, off_t);
 
 
 /*
@@ -965,6 +967,29 @@ uni_range(int a, int b)
 		printf("%d,0", b);
 }
 
+static char *
+preadline(int fd, size_t len, off_t off)
+{
+	char *line;
+	ssize_t nr;
+
+	line = emalloc(len + 1);
+	if ((nr = pread(fd, line, len, off)) < 0)
+		err(1, "preadline");
+	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
@@ -981,6 +1006,32 @@ change(char *file1, FILE *f1, char *file2, FILE *f2, int a, int b, int c, int d)
 restart:
 	if (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 (format == D_CONTEXT || format == D_UNIFIED) {
 		/*
 		 * Allocate change records as needed.