3 files changed, 123 insertions, 150 deletions

diff --git a/usr.bin/diff/diff.c b/usr.bin/diff/diff.c
index 1bad6226f49d..bfd391a4e3dd 100644
--- a/usr.bin/diff/diff.c
+++ b/usr.bin/diff/diff.c
@@ -362,9 +362,9 @@ main(int argc, char **argv)
 	}
 
 	if (dflags & D_EMPTY1 && dflags & D_EMPTY2){
-		warn("%s", argv[0]);	
+		warn("%s", argv[0]);
 		warn("%s", argv[1]);
-		exit(2);	
+		exit(2);
 	}
 
 	if (stb1.st_mode == 0)
@@ -477,8 +477,10 @@ print_only(const char *path, size_t dirlen, const char *entry)
 void
 print_status(int val, char *path1, char *path2, const char *entry)
 {
-	if (label[0] != NULL) path1 = label[0];
-	if (label[1] != NULL) path2 = label[1];
+	if (label[0] != NULL)
+		path1 = label[0];
+	if (label[1] != NULL)
+		path2 = label[1];
 
 	switch (val) {
 	case D_BINARY:
@@ -535,10 +537,10 @@ usage(void)
 	    "            [--no-ignore-case] [--normal] [--tabsize] [-I pattern] [-L label]\n"
 	    "            [-S name] [-X file] [-x pattern] dir1 dir2\n"
 	    "       diff [-aBbditwW] [--expand-tabs] [--ignore-all-blanks]\n"
-            "            [--ignore-blank-lines] [--ignore-case] [--minimal]\n"
-            "            [--no-ignore-file-name-case] [--strip-trailing-cr]\n"
-            "            [--suppress-common-lines] [--tabsize] [--text] [--width]\n"
-            "            -y | --side-by-side file1 file2\n");
+	    "            [--ignore-blank-lines] [--ignore-case] [--minimal]\n"
+	    "            [--no-ignore-file-name-case] [--strip-trailing-cr]\n"
+	    "            [--suppress-common-lines] [--tabsize] [--text] [--width]\n"
+	    "            -y | --side-by-side file1 file2\n");
 
 	exit(2);
 }
diff --git a/usr.bin/diff/diff.h b/usr.bin/diff/diff.h
index b5536bd7bf77..28d08288e1e9 100644
--- a/usr.bin/diff/diff.h
+++ b/usr.bin/diff/diff.h
@@ -1,6 +1,6 @@
+/*	$OpenBSD: diff.h,v 1.34 2020/11/01 18:16:08 jcs Exp $	*/
 
-
-/*ROR
+/*-
  * Copyright (c) 1991, 1993
  *	The Regents of the University of California.  All rights reserved.
  *
diff --git a/usr.bin/diff/diffreg.c b/usr.bin/diff/diffreg.c
index e728441c2cb2..94026007a1bf 100644
--- a/usr.bin/diff/diffreg.c
+++ b/usr.bin/diff/diffreg.c
@@ -95,66 +95,54 @@ __FBSDID("$FreeBSD$");
  */
 
 /*
- *	Uses an algorithm due to Harold Stone, which finds
- *	a pair of longest identical subsequences in the 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
+ *	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.
+ *	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.
+ *	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.
+ *	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
+ *	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.
+ *	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.
+ *	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 {
@@ -213,24 +201,24 @@ 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 int   hw, padding;	/* half width and padding */
-static int   edoffset;
-static long *ixnew;		/* will be overlaid on file[1] */
-static long *ixold;		/* will be overlaid on klist */
+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 int	 hw, padding;	/* half width and padding */
+static int	 edoffset;
+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 int	 clistlen;		/* the length of clist */
 static struct line *sfile[2];	/* shortened by pruning common prefix/suffix */
-static int (*chrtran)(int);	/* translation table for case-folding */
+static int	(*chrtran)(int);	/* 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;
@@ -251,7 +239,7 @@ static int
 cup2low(int c)
 {
 
-	return tolower(c);
+	return (tolower(c));
 }
 
 int
@@ -278,11 +266,11 @@ diffreg(char *file1, char *file2, int flags, int capsicum)
 		padding = tabsize - (hw % tabsize);
 		if ((flags & D_EXPANDTABS) != 0 || (padding % tabsize == 0))
 			padding = MIN_PAD;
-	
+
 		hw = (width >> 1) -
 		    ((padding == MIN_PAD) ? (padding << 1) : padding) - 1;
 	}
-	
+
 
 	if (flags & D_IGNORECASE)
 		chrtran = cup2low;
@@ -622,7 +610,7 @@ 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;
+	unsigned numtries, bound;
 
 	if (flags & D_MINIMAL)
 		bound = UINT_MAX;
@@ -720,9 +708,9 @@ unravel(int p)
 
 /*
  * 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
+ *  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)
@@ -745,7 +733,7 @@ check(FILE *f1, FILE *f2, int flags)
 			ixnew[j] = ctnew += skipline(f2);
 			j++;
 		}
-		if (flags & (D_FOLDBLANKS|D_IGNOREBLANKS|D_IGNORECASE|D_STRIPCR)) {
+		if (flags & (D_FOLDBLANKS | D_IGNOREBLANKS | D_IGNORECASE | D_STRIPCR)) {
 			for (;;) {
 				c = getc(f1);
 				d = getc(f2);
@@ -753,7 +741,7 @@ check(FILE *f1, FILE *f2, int flags)
 				 * GNU diff ignores a missing newline
 				 * in one file for -b or -w.
 				 */
-				if (flags & (D_FOLDBLANKS|D_IGNOREBLANKS)) {
+				if (flags & (D_FOLDBLANKS | D_IGNOREBLANKS)) {
 					if (c == EOF && d == '\n') {
 						ctnew++;
 						break;
@@ -793,7 +781,7 @@ check(FILE *f1, FILE *f2, int flags)
 							break;
 						ctnew++;
 					} while (isspace(d = getc(f2)));
-				} else if ((flags & D_IGNOREBLANKS)) {
+				} else if (flags & D_IGNOREBLANKS) {
 					while (isspace(c) && c != '\n') {
 						c = getc(f1);
 						ctold++;
@@ -913,16 +901,11 @@ output(char *file1, FILE *f1, char *file2, FILE *f2, int flags)
 	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){
-				if (diff_format == D_SIDEBYSIDE &&
-				    suppress_common != 1) {
-					nc = fetch(ixold, i0, i0, f1, '\0',
-					    1, flags);
-					print_space(nc,
-					    (hw - nc) + (padding << 1) + 1,
-					    flags);
-					fetch(ixnew, J[i0], J[i0], f2, '\0',
-					    0, flags);
+			while (i0 <= m && J[i0] == J[i0 - 1] + 1) {
+				if (diff_format == D_SIDEBYSIDE && suppress_common != 1) {
+					nc = fetch(ixold, i0, i0, f1, '\0', 1, flags);
+					print_space(nc, (hw - nc) + (padding << 1) + 1, flags);
+					fetch(ixnew, J[i0], J[i0], f2, '\0', 0, flags);
 					printf("\n");
 				}
 				i0++;
@@ -935,33 +918,28 @@ output(char *file1, FILE *f1, char *file2, FILE *f2, int flags)
 			J[i1] = j1;
 
 			/*
-			 * When using side-by-side, lines from both of the
-			 * files are printed. The algorithm used by diff(1)
-			 * identifies the ranges in which two files differ.
+			 * When using side-by-side, lines from both of the files are
+			 * printed. The algorithm used by diff(1) identifies the ranges
+			 * in which two files differ.
 			 * See the change() function below.
-			 * The for loop below consumes the shorter range,
-			 * whereas one of the while loops deals with the
-			 * longer one.
+			 * The for loop below consumes the shorter range, whereas one of
+			 * the while loops deals with the longer one.
 			 */
 			if (diff_format == D_SIDEBYSIDE) {
-				for (i=i0, j=j0; i<=i1 && j<=j1; i++, j++)
-					change(file1, f1, file2, f2, i, i,
-					    j, j, &flags);
+				for (i = i0, j = j0; i <= i1 && j <= j1; i++, j++)
+					change(file1, f1, file2, f2, i, i, j, j, &flags);
 
 				while (i <= i1) {
-					change(file1, f1, file2, f2,
-					    i, i, j+1, j, &flags);
+					change(file1, f1, file2, f2, i, i, j + 1, j, &flags);
 					i++;
 				}
 
 				while (j <= j1) {
-					change(file1, f1, file2, f2,
-					    i+1, i, j, j, &flags);
+					change(file1, f1, file2, f2, i + 1, i, j, j, &flags);
 					j++;
 				}
 			} else
-				change(file1, f1, file2, f2, i0, i1, j0,
-				    j1, &flags);
+				change(file1, f1, file2, f2, i0, i1, j0, j1, &flags);
 		}
 	} else {
 		for (i0 = m; i0 >= 1; i0 = i1 - 1) {
@@ -1078,9 +1056,8 @@ restart:
 	if (ignore_pats != NULL || skip_blanks) {
 		char *line;
 		/*
-		 * All lines in the change, insert, or delete must
-		 * match an ignore pattern for the change to be
-		 * ignored.
+		 * 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++) {
@@ -1202,7 +1179,7 @@ proceed:
 			nc = fetch(ixold, a, b, f1, '\0', 1, *pflags);
 			print_space(nc, hw - nc + padding, *pflags);
 		}
-		printf("%c", (a>b)? '>' : ((c>d)? '<' : '|'));
+		printf("%c", (a > b) ? '>' : ((c > d) ? '<' : '|'));
 		print_space(hw + padding + 1 , padding, *pflags);
 		fetch(ixnew, c, d, f2, '\0', 0, *pflags);
 		printf("\n");
@@ -1216,11 +1193,11 @@ proceed:
 		fetch(ixnew, c, d, f2, diff_format == D_NORMAL ? '>' : '\0', 0, *pflags);
 	if (edoffset != 0 && diff_format == D_EDIT) {
 		/*
-		 * A non-zero edoffset 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.
+		 * A non-zero edoffset 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.
 		 */
 		printf(".\n");
 		printf("%ds/.//\n", a + edoffset - 1);
@@ -1271,10 +1248,10 @@ fetch(long *f, int a, int b, FILE *lb, int ch, int oldfile, int flags)
 	}
 	for (i = a; i <= b; i++) {
 		fseek(lb, f[i - 1], SEEK_SET);
-		nc = (f[i] - f[i - 1]);
+		nc = f[i] - f[i - 1];
 		if (diff_format == D_SIDEBYSIDE && hw < nc)
 			nc = hw;
-		if ((diff_format != D_IFDEF && diff_format != D_GFORMAT) &&
+		if (diff_format != D_IFDEF && diff_format != D_GFORMAT &&
 		    ch != '\0') {
 			printf("%c", ch);
 			if (Tflag && (diff_format == D_NORMAL ||
@@ -1301,9 +1278,8 @@ fetch(long *f, int a, int b, FILE *lb, int ch, int oldfile, int flags)
 				    diff_format == D_NREVERSE)
 					warnx("No newline at end of file");
 				else
-					printf("\n\\ No newline at end of "
-					    "file\n");
-				return col;
+					printf("\n\\ No newline at end of file\n");
+				return (col);
 			}
 			/*
 			 * when using --side-by-side, col needs to be increased
@@ -1311,8 +1287,8 @@ fetch(long *f, int a, int b, FILE *lb, int ch, int oldfile, int flags)
 			 */
 			if (c == '\t') {
 				if (flags & D_EXPANDTABS) {
-					newcol = ((col/tabsize)+1)*tabsize;
-					do {	
+					newcol = ((col / tabsize) + 1) * tabsize;
+					do {
 						if (diff_format == D_SIDEBYSIDE)
 							j++;
 						printf(" ");
@@ -1320,8 +1296,7 @@ fetch(long *f, int a, int b, FILE *lb, int ch, int oldfile, int flags)
 				} else {
 					if (diff_format == D_SIDEBYSIDE) {
 						if ((j + tabsize) > nc) {
-							printf("%*s",
-							nc - j,"");
+							printf("%*s", nc - j, "");
 							j = col = nc;
 						} else {
 							printf("\t");
@@ -1334,20 +1309,17 @@ fetch(long *f, int a, int b, FILE *lb, int ch, int oldfile, int flags)
 					}
 				}
 			} else {
-				if (diff_format == D_EDIT && j == 1 && c == '\n'
-				    && lastc == '.') {
+				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 set the,
-					 * global variable edoffset to an
-					 * offset from which to restart.
-					 * The caller must check the value
-					 * of edoffset
+					 * Don't print a bare "." line since that will confuse
+					 * ed(1). Print ".." instead and set the, global variable
+					 * edoffset to an offset from which to restart. The
+					 * caller must check the value of edoffset
 					 */
 					printf(".\n");
 					edoffset = i - a + 1;
-					return edoffset;
+					return (edoffset);
 				}
 				/* when side-by-side, do not print a newline */
 				if (diff_format != D_SIDEBYSIDE || c != '\n') {
@@ -1357,7 +1329,7 @@ fetch(long *f, int a, int b, FILE *lb, int ch, int oldfile, int flags)
 			}
 		}
 	}
-	return col;
+	return (col);
 }
 
 /*
@@ -1428,7 +1400,7 @@ asciifile(FILE *f)
 	return (memchr(buf, '\0', cnt) == NULL);
 }
 
-#define begins_with(s, pre) (strncmp(s, pre, sizeof(pre)-1) == 0)
+#define begins_with(s, pre) (strncmp(s, pre, sizeof(pre) - 1) == 0)
 
 static char *
 match_function(const long *f, int pos, FILE *fp)
@@ -1459,18 +1431,17 @@ match_function(const long *f, int pos, FILE *fp)
 					if (!state)
 						state = " (public)";
 				} else {
-					strlcpy(lastbuf, buf, sizeof lastbuf);
+					strlcpy(lastbuf, buf, sizeof(lastbuf));
 					if (state)
-						strlcat(lastbuf, state,
-						    sizeof lastbuf);
+						strlcat(lastbuf, state, sizeof(lastbuf));
 					lastmatchline = pos;
-					return lastbuf;
+					return (lastbuf);
 				}
 			}
 		}
 		pos--;
 	}
-	return lastmatchline > 0 ? lastbuf : NULL;
+	return (lastmatchline > 0 ? lastbuf : NULL);
 }
 
 /* dump accumulated "context" diff changes */
@@ -1493,7 +1464,7 @@ dump_context_vec(FILE *f1, FILE *f2, int flags)
 
 	printf("***************");
 	if ((flags & D_PROTOTYPE)) {
-		f = match_function(ixold, lowa-1, f1);
+		f = match_function(ixold, lowa - 1, f1);
 		if (f != NULL)
 			printf(" %s", f);
 	}
@@ -1600,7 +1571,7 @@ dump_unified_vec(FILE *f1, FILE *f2, int flags)
 	uni_range(lowc, upd);
 	printf(" @@");
 	if ((flags & D_PROTOTYPE)) {
-		f = match_function(ixold, lowa-1, f1);
+		f = match_function(ixold, lowa - 1, f1);
 		if (f != NULL)
 			printf(" %s", f);
 	}