todd miller's change to disable utf-8 for non-multibyte locales

such as C or POSIX

4 files changed, 11 insertions, 6 deletions

diff --git a/awk.h b/awk.h
index 49b5dfc..217319c 100644
--- a/awk.h
+++ b/awk.h
@@ -64,6 +64,8 @@ extern bool	safe;		/* false => unsafe, true => safe */
 #define	RECSIZE	(8 * 1024)	/* sets limit on records, fields, etc., etc. */
 extern int	recsize;	/* size of current record, orig RECSIZE */
 
+extern size_t	awk_mb_cur_max;	/* max size of a multi-byte character */
+
 extern char	EMPTY[];	/* this avoid -Wwritable-strings issues */
 extern char	**FS;
 extern char	**RS;
diff --git a/b.c b/b.c
index 55b320e..9792a97 100644
--- a/b.c
+++ b/b.c
@@ -778,7 +778,7 @@ struct runedata getrune(FILE *fp)
 	c = getc(fp);
 	if (c == EOF)
 		return result;	// result.rune == 0 --> EOF
-	else if (c < 128) {
+	else if (c < 128 || awk_mb_cur_max == 1) {
 		result.bytes[0] = c;
 		result.len = 1;
 		result.rune = c;
diff --git a/main.c b/main.c
index d500729..8f03963 100644
--- a/main.c
+++ b/main.c
@@ -53,6 +53,8 @@ bool	CSV = false;	/* true for csv input */
 
 bool	safe = false;	/* true => "safe" mode */
 
+size_t	awk_mb_cur_max = 1;
+
 static noreturn void fpecatch(int n
 #ifdef SA_SIGINFO
 	, siginfo_t *si, void *uc
@@ -116,6 +118,7 @@ int main(int argc, char *argv[])
 
 	setlocale(LC_CTYPE, "");
 	setlocale(LC_NUMERIC, "C"); /* for parsing cmdline & prog */
+	awk_mb_cur_max = MB_CUR_MAX;
 	cmdname = argv[0];
 	if (argc == 1) {
 		fprintf(stderr,
diff --git a/run.c b/run.c
index bb64e51..bbf149c 100644
--- a/run.c
+++ b/run.c
@@ -605,7 +605,7 @@ int u8_isutf(const char *s)
 	unsigned char c;
 
 	c = s[0];
-	if (c < 128)
+	if (c < 128 || awk_mb_cur_max == 1)
 		return 1; /* what if it's 0? */
 
 	n = strlen(s);
@@ -632,7 +632,7 @@ int u8_rune(int *rune, const char *s)
 	unsigned char c;
 
 	c = s[0];
-	if (c < 128) {
+	if (c < 128 || awk_mb_cur_max == 1) {
 		*rune = c;
 		return 1;
 	}
@@ -679,7 +679,7 @@ int u8_strlen(const char *s)
 	totlen = 0;
 	for (i = 0; i < n; i += len) {
 		c = s[i];
-		if (c < 128) {
+		if (c < 128 || awk_mb_cur_max == 1) {
 			len = 1;
 		} else {
 			len = u8_nextlen(&s[i]);
@@ -1289,7 +1289,7 @@ int format(char **pbuf, int *pbufsize, const char *s, Node *a)	/* printf-like co
 				int charval = (int) getfval(x);
 
 				if (charval != 0) {
-					if (charval < 128)
+					if (charval < 128 || awk_mb_cur_max == 1)
 						snprintf(p, BUFSZ(p), fmt, charval);
 					else {
 						// possible unicode character
@@ -1977,7 +1977,7 @@ static char *nawk_convert(const char *s, int (*fun_c)(int),
 	const char *ps = NULL;
 	size_t n       = 0;
 	wchar_t wc;
-	size_t sz = MB_CUR_MAX;
+	const size_t sz = awk_mb_cur_max;
 	int unused;
 
 	if (sz == 1) {