POSIX doesn't permit an unescaped '/' in an extended regular expression.

Unlike upstream awk, ours has historically allowed unescaped '/'
inside a bracket expression for compatibility with other awk
implementations but the check was too simple-minded.  This improves
the matching to allow things like /[]/]/, /[^]// and '/[abc[:digit:]/@#]/'
To enable strict POSIX compliance, set POSIXLY_CORRECT.

5 files changed, 35 insertions, 21 deletions

diff --git a/usr.bin/awk/awk.1 b/usr.bin/awk/awk.1
index 84e4f057e73..4d320b705f2 100644
--- a/usr.bin/awk/awk.1
+++ b/usr.bin/awk/awk.1
@@ -1,4 +1,4 @@
-.\"	$OpenBSD: awk.1,v 1.50 2020/06/10 21:05:02 millert Exp $
+.\"	$OpenBSD: awk.1,v 1.51 2020/06/13 01:19:55 millert Exp $
 .\"
 .\" Copyright (C) Lucent Technologies 1997
 .\" All Rights Reserved
@@ -22,7 +22,7 @@
 .\" ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE OF
 .\" THIS SOFTWARE.
 .\"
-.Dd $Mdocdate: June 10 2020 $
+.Dd $Mdocdate: June 13 2020 $
 .Dt AWK 1
 .Os
 .Sh NAME
@@ -805,7 +805,9 @@ string argument for
 .Fn sub
 and
 .Fn gsub
-are not collapsed.
+are not collapsed and a slash
+.Pq Ql /
+does not need to be escaped in a bracket expression.
 .Pp
 The flags
 .Op Fl \&dV
diff --git a/usr.bin/awk/awk.h b/usr.bin/awk/awk.h
index e55af318b9a..080bf94cc22 100644
--- a/usr.bin/awk/awk.h
+++ b/usr.bin/awk/awk.h
@@ -1,4 +1,4 @@
-/*	$OpenBSD: awk.h,v 1.23 2020/06/10 21:06:09 millert Exp $	*/
+/*	$OpenBSD: awk.h,v 1.24 2020/06/13 01:19:55 millert Exp $	*/
 /****************************************************************
 Copyright (C) Lucent Technologies 1997
 All Rights Reserved
@@ -62,6 +62,7 @@ extern enum compile_states {
 } compile_time;
 
 extern bool	safe;		/* false => unsafe, true => safe */
+extern bool	do_posix;	/* true if POSIXLY_CORRECT set */
 
 #define	RECSIZE	(8 * 1024)	/* sets limit on records, fields, etc., etc. */
 extern int	recsize;	/* size of current record, orig RECSIZE */
diff --git a/usr.bin/awk/lex.c b/usr.bin/awk/lex.c
index f9ee05c7a5f..18a5ecac785 100644
--- a/usr.bin/awk/lex.c
+++ b/usr.bin/awk/lex.c
@@ -1,4 +1,4 @@
-/*	$OpenBSD: lex.c,v 1.19 2020/06/10 21:05:50 millert Exp $	*/
+/*	$OpenBSD: lex.c,v 1.20 2020/06/13 01:19:55 millert Exp $	*/
 /****************************************************************
 Copyright (C) Lucent Technologies 1997
 All Rights Reserved
@@ -524,12 +524,12 @@ int regexpr(void)
 	int c, openclass = 0;
 	static char *buf = NULL;
 	static int bufsz = 500;
-	char *bp;
+	char *bp, *cstart;
 
 	if (buf == NULL && (buf = malloc(bufsz)) == NULL)
 		FATAL("out of space for rex expr");
 	bp = buf;
-	for ( ; ((c = input()) != '/' || openclass == 1) && c != 0; ) {
+	for ( ; ((c = input()) != '/' || openclass > 0) && c != 0; ) {
 		if (!adjbuf(&buf, &bufsz, bp-buf+3, 500, &bp, "regexpr"))
 			FATAL("out of space for reg expr %.10s...", buf);
 		if (c == '\n') {
@@ -541,10 +541,25 @@ int regexpr(void)
 			*bp++ = '\\';
 			*bp++ = input();
 		} else {
-			if (c == '[')
-				openclass = 1;
-			else if (c == ']')
-				openclass = 0;
+			/*
+			 * POSIX requires a slash in a regexp to be escaped,
+			 * other awks don't require it to be escaped inside
+			 * a character class.
+			 */
+			if (!do_posix) {
+				if (c == '[') {
+					if (++openclass == 1)
+						cstart = bp;
+				} else if (c == ']' && openclass > 0) {
+					/*
+					 * A ']' as the first char in a
+					 * class is treated literally.
+					 */
+					if (cstart != bp - 1 &&
+					    (cstart != bp - 2 || bp[-1] != '^'))
+						openclass--;
+				}
+			}
 			*bp++ = c;
 		}
 	}
diff --git a/usr.bin/awk/main.c b/usr.bin/awk/main.c
index 75feccc0aa5..62b0d4c33dc 100644
--- a/usr.bin/awk/main.c
+++ b/usr.bin/awk/main.c
@@ -1,4 +1,4 @@
-/*	$OpenBSD: main.c,v 1.39 2020/06/11 13:51:18 millert Exp $	*/
+/*	$OpenBSD: main.c,v 1.40 2020/06/13 01:19:55 millert Exp $	*/
 /****************************************************************
 Copyright (C) Lucent Technologies 1997
 All Rights Reserved
@@ -53,7 +53,8 @@ static size_t	maxpfile;	/* max program filename */
 static size_t	npfile;		/* number of filenames */
 static size_t	curpfile;	/* current filename */
 
-bool	safe = false;	/* true => "safe" mode */
+bool	safe = false;		/* true => "safe" mode */
+bool	do_posix = false;	/* true => POSIX mode */
 
 static noreturn void fpecatch(int n
 #ifdef SA_SIGINFO
@@ -162,6 +163,8 @@ int main(int argc, char *argv[])
 	(void)signal(SIGFPE, fpecatch);
 #endif
 
+	do_posix = (getenv("POSIXLY_CORRECT") != NULL);
+
 	yyin = NULL;
 	symtab = makesymtab(NSYMTAB);
 	while (argc > 1 && argv[1][0] == '-' && argv[1][1] != '\0') {
diff --git a/usr.bin/awk/run.c b/usr.bin/awk/run.c
index 7bdb607b8d0..b150d90a55a 100644
--- a/usr.bin/awk/run.c
+++ b/usr.bin/awk/run.c
@@ -1,4 +1,4 @@
-/*	$OpenBSD: run.c,v 1.57 2020/06/10 21:05:50 millert Exp $	*/
+/*	$OpenBSD: run.c,v 1.58 2020/06/13 01:19:55 millert Exp $	*/
 /****************************************************************
 Copyright (C) Lucent Technologies 1997
 All Rights Reserved
@@ -2122,13 +2122,6 @@ void backsub(char **pb_ptr, const char **sptr_ptr)	/* handle \\& variations */
 {						/* sptr[0] == '\\' */
 	char *pb = *pb_ptr;
 	const char *sptr = *sptr_ptr;
-	static bool first = true;
-	static bool do_posix = false;
-
-	if (first) {
-		first = false;
-		do_posix = (getenv("POSIXLY_CORRECT") != NULL);
-	}
 
 	if (sptr[1] == '\\') {
 		if (sptr[2] == '\\' && sptr[3] == '&') { /* \\\& -> \& */