[PARSER] Only use signed char for syntax arrays

The existing scheme of using the native char for syntax array indicies
makes cross-compiling difficult.  Therefore it makes sense to choose
one specific sign for everyone.

Since signed chars are native to most platforms and i386, it makes more
sense to use that if we are to choose one type for everyone.

8 files changed, 49 insertions, 127 deletions

diff --git a/src/expand.c b/src/expand.c
index dafb51f..cf64921 100644
--- a/src/expand.c
+++ b/src/expand.c
@@ -171,7 +171,7 @@ STATIC size_t
 esclen(const char *start, const char *p) {
 	size_t esc = 0;
 
-	while (p > start && *--p == CTLESC) {
+	while (p > start && *--p == (char)CTLESC) {
 		esc++;
 	}
 	return esc;
@@ -296,7 +296,7 @@ argstr(char *p, int flag)
 		flag &= ~EXP_TILDE;
 tilde:
 		q = p;
-		if (*q == CTLESC && (flag & EXP_QWORD))
+		if (*q == (char)CTLESC && (flag & EXP_QWORD))
 			q++;
 		if (*q == '~')
 			p = exptilde(p, q, flag);
@@ -305,7 +305,7 @@ start:
 	startloc = expdest - (char *)stackblock();
 	for (;;) {
 		length += strcspn(p + length, reject);
-		c = p[length];
+		c = (signed char)p[length];
 		if (c && (!(c & 0x80) || c == CTLENDARI)) {
 			/* c == '=' || c == ':' || c == CTLENDARI */
 			length++;
@@ -352,9 +352,9 @@ start:
 			if (
 				!inquotes &&
 				!memcmp(p, dolatstr, DOLATSTRLEN) &&
-				(p[4] == CTLQUOTEMARK || (
-					p[4] == CTLENDVAR &&
-					p[5] == CTLQUOTEMARK
+				(p[4] == (char)CTLQUOTEMARK || (
+					p[4] == (char)CTLENDVAR &&
+					p[5] == (char)CTLQUOTEMARK
 				))
 			) {
 				p = evalvar(p + 1, flag) + 1;
@@ -394,7 +394,7 @@ breakloop:
 STATIC char *
 exptilde(char *startp, char *p, int flag)
 {
-	char c;
+	signed char c;
 	char *name;
 	const char *home;
 	int quotes = flag & QUOTES_ESC;
@@ -503,7 +503,7 @@ expari(int quotes)
 	do {
 		int esc;
 
-		while (*p != CTLARI) {
+		while (*p != (char)CTLARI) {
 			p--;
 #ifdef DEBUG
 			if (p < start) {
@@ -626,7 +626,7 @@ scanleft(
 		*loc2 = c;
 		if (match)
 			return loc;
-		if (quotes && *loc == CTLESC)
+		if (quotes && *loc == (char)CTLESC)
 			loc++;
 		loc++;
 		loc2++;
@@ -860,7 +860,7 @@ end:
 	if (subtype != VSNORMAL) {	/* skip to end of alternative */
 		int nesting = 1;
 		for (;;) {
-			if ((c = *p++) == CTLESC)
+			if ((c = (signed char)*p++) == CTLESC)
 				p++;
 			else if (c == CTLBACKQ || c == (CTLBACKQ|CTLQUOTE)) {
 				if (varlen >= 0)
@@ -892,7 +892,7 @@ memtodest(const char *p, size_t len, const char *syntax, int quotes) {
 	q = makestrspace(len * 2, expdest);
 
 	do {
-		int c = (unsigned char)*p++;
+		int c = (signed char)*p++;
 		if (c) {
 			if ((quotes & QUOTES_ESC) &&
 			    (syntax[c] == CCTL || syntax[c] == CBACK))
@@ -1078,7 +1078,7 @@ ifsbreakup(char *string, struct arglist *arglist)
 			ifsspc = 0;
 			while (p < string + ifsp->endoff) {
 				q = p;
-				if (*p == CTLESC)
+				if (*p == (char)CTLESC)
 					p++;
 				if (strchr(ifs, *p)) {
 					if (!nulonly)
@@ -1101,7 +1101,7 @@ ifsbreakup(char *string, struct arglist *arglist)
 								break;
 							}
 							q = p;
-							if (*p == CTLESC)
+							if (*p == (char)CTLESC)
 								p++;
 							if (strchr(ifs, *p) == NULL ) {
 								p = q;
@@ -1658,7 +1658,7 @@ _rmescapes(char *str, int flag)
 	globbing = flag & RMESCAPE_GLOB;
 	notescaped = globbing;
 	while (*p) {
-		if (*p == CTLQUOTEMARK) {
+		if (*p == (char)CTLQUOTEMARK) {
 			inquotes = ~inquotes;
 			p++;
 			notescaped = globbing;
@@ -1669,7 +1669,7 @@ _rmescapes(char *str, int flag)
 			notescaped = 0;
 			goto copy;
 		}
-		if (*p == CTLESC) {
+		if (*p == (char)CTLESC) {
 			p++;
 			if (notescaped && inquotes && *p != '/') {
 				*q++ = '\\';
@@ -1734,7 +1734,7 @@ varunset(const char *end, const char *var, const char *umsg, int varflags)
 	tail = nullstr;
 	msg = "parameter not set";
 	if (umsg) {
-		if (*end == CTLENDVAR) {
+		if (*end == (char)CTLENDVAR) {
 			if (varflags & VSNUL)
 				tail = " or null";
 		} else
diff --git a/src/input.c b/src/input.c
index 639c023..057da71 100644
--- a/src/input.c
+++ b/src/input.c
@@ -263,7 +263,7 @@ preadbuffer(void)
 		}
 		popstring();
 		if (--parsenleft >= 0)
-			return (*parsenextc++);
+			return (signed char)*parsenextc++;
 	}
 	if (unlikely(parsenleft == EOF_NLEFT || parsefile->buf == NULL))
 		return PEOF;
@@ -346,7 +346,7 @@ again:
 
 	*q = savec;
 
-	return *parsenextc++;
+	return (signed char)*parsenextc++;
 }
 
 /*
diff --git a/src/input.h b/src/input.h
index 7675653..1ed9ddf 100644
--- a/src/input.h
+++ b/src/input.h
@@ -64,4 +64,5 @@ void popfile(void);
 void popallfiles(void);
 void closescript(void);
 
-#define pgetc_macro()	(--parsenleft >= 0? *parsenextc++ : preadbuffer())
+#define pgetc_macro() \
+	(--parsenleft >= 0 ? (signed char)*parsenextc++ : preadbuffer())
diff --git a/src/jobs.c b/src/jobs.c
index 0113354..9e28adb 100644
--- a/src/jobs.c
+++ b/src/jobs.c
@@ -1357,8 +1357,9 @@ STATIC void
 cmdputs(const char *s)
 {
 	const char *p, *str;
-	char c, cc[2] = " ";
+	char cc[2] = " ";
 	char *nextc;
+	signed char c;
 	int subtype = 0;
 	int quoted = 0;
 	static const char vstype[VSTYPE + 1][4] = {
diff --git a/src/mksyntax.c b/src/mksyntax.c
index 5e65bdd..7a8a9ae 100644
--- a/src/mksyntax.c
+++ b/src/mksyntax.c
@@ -92,34 +92,20 @@ static char writer[] = "\
 static FILE *cfile;
 static FILE *hfile;
 static char *syntax[513];
-static int base;
-static int size;	/* number of values which a char variable can have */
-static int nbits;	/* number of bits in a character */
-static int digit_contig;/* true if digits are contiguous */
 
 static void filltable(char *);
 static void init(void);
 static void add(char *, char *);
 static void print(char *);
-static void output_type_macros(int);
-static void digit_convert(void);
+static void output_type_macros(void);
 int main(int, char **);
 
 int
 main(int argc, char **argv)
 {
-#ifdef	TARGET_CHAR
-	TARGET_CHAR c;
-	TARGET_CHAR d;
-#else
-	char c;
-	char d;
-#endif
-	int sign;
 	int i;
 	char buf[80];
 	int pos;
-	static char digit[] = "0123456789";
 
 	/* Create output files */
 	if ((cfile = fopen("syntax.c", "w")) == NULL) {
@@ -133,32 +119,6 @@ main(int argc, char **argv)
 	fputs(writer, hfile);
 	fputs(writer, cfile);
 
-	/* Determine the characteristics of chars. */
-	c = -1;
-	if (c <= 0)
-		sign = 1;
-	else
-		sign = 0;
-	for (nbits = 1 ; ; nbits++) {
-		d = (1 << nbits) - 1;
-		if (d == c)
-			break;
-	}
-	printf("%s %d bit chars\n", sign? "signed" : "unsigned", nbits);
-	if (nbits > 9) {
-		fputs("Characters can't have more than 9 bits\n", stderr);
-		exit(2);
-	}
-	size = (1 << nbits) + 1;
-	base = 2;
-	if (sign)
-		base += 1 << (nbits - 1);
-	digit_contig = 1;
-	for (i = 0 ; i < 10 ; i++) {
-		if (digit[i] != '0' + i)
-			digit_contig = 0;
-	}
-
 	fputs("#include <ctype.h>\n", hfile);
 	fputs("\n", hfile);
 	fputs("#ifdef CEOF\n", hfile);
@@ -185,16 +145,16 @@ main(int argc, char **argv)
 		fprintf(hfile, "/* %s */\n", is_entry[i].comment);
 	}
 	putc('\n', hfile);
-	fprintf(hfile, "#define SYNBASE %d\n", base);
-	fprintf(hfile, "#define PEOF %d\n\n", -base);
-	fprintf(hfile, "#define PEOA %d\n\n", -base + 1);
+	fprintf(hfile, "#define SYNBASE %d\n", 130);
+	fprintf(hfile, "#define PEOF %d\n\n", -130);
+	fprintf(hfile, "#define PEOA %d\n\n", -129);
 	putc('\n', hfile);
 	fputs("#define BASESYNTAX (basesyntax + SYNBASE)\n", hfile);
 	fputs("#define DQSYNTAX (dqsyntax + SYNBASE)\n", hfile);
 	fputs("#define SQSYNTAX (sqsyntax + SYNBASE)\n", hfile);
 	fputs("#define ARISYNTAX (arisyntax + SYNBASE)\n", hfile);
 	putc('\n', hfile);
-	output_type_macros(sign);		/* is_digit, etc. */
+	output_type_macros();		/* is_digit, etc. */
 	putc('\n', hfile);
 
 	/* Generate the syntax tables. */
@@ -248,8 +208,6 @@ main(int argc, char **argv)
 	add("_", "ISUNDER");
 	add("#?$!-*@", "ISSPECL");
 	print("is_type");
-	if (! digit_contig)
-		digit_convert();
 	exit(0);
 	/* NOTREACHED */
 }
@@ -265,7 +223,7 @@ filltable(char *dftval)
 {
 	int i;
 
-	for (i = 0 ; i < size ; i++)
+	for (i = 0 ; i < 257; i++)
 		syntax[i] = dftval;
 }
 
@@ -283,11 +241,7 @@ init(void)
 	syntax[0] = "CEOF";
 	syntax[1] = "CIGN";
 	for (ctl = CTL_FIRST; ctl <= CTL_LAST; ctl++ )
-#ifdef TARGET_CHAR
-		syntax[base + (TARGET_CHAR)ctl ] = "CCTL";
-#else
-		syntax[base + ctl] = "CCTL";
-#endif /* TARGET_CHAR */
+		syntax[130 + ctl] = "CCTL";
 }
 
 
@@ -299,7 +253,7 @@ static void
 add(char *p, char *type)
 {
 	while (*p)
-		syntax[*p++ + base] = type;
+		syntax[(signed char)*p++ + 130] = type;
 }
 
 
@@ -315,9 +269,9 @@ print(char *name)
 	int col;
 
 	fprintf(hfile, "extern const char %s[];\n", name);
-	fprintf(cfile, "const char %s[%d] = {\n", name, size);
+	fprintf(cfile, "const char %s[%d] = {\n", name, 257);
 	col = 0;
-	for (i = 0 ; i < size ; i++) {
+	for (i = 0 ; i < 257; i++) {
 		if (i == 0) {
 			fputs("      ", cfile);
 		} else if ((i & 03) == 0) {
@@ -342,7 +296,7 @@ print(char *name)
  */
 
 static char *macro[] = {
-	"#define is_digit(c)\t((is_type+SYNBASE)[(signed char)(c)] & ISDIGIT)\n",
+	"#define is_digit(c)\t((unsigned)((c) - '0') <= 9)\n",
 	"#define is_alpha(c)\tisalpha((unsigned char)(c))\n",
 	"#define is_name(c)\t((c) == '_' || isalpha((unsigned char)(c)))\n",
 	"#define is_in_name(c)\t((c) == '_' || isalnum((unsigned char)(c)))\n",
@@ -351,45 +305,11 @@ static char *macro[] = {
 };
 
 static void
-output_type_macros(int sign)
+output_type_macros(void)
 {
 	char **pp;
 
-	if (digit_contig)
-		macro[0] = "#define is_digit(c)\t((unsigned)((c) - '0') <= 9)\n";
 	for (pp = macro ; *pp ; pp++)
-		fprintf(hfile, *pp, sign ? "char" : "unsigned char");
-	if (digit_contig)
-		fputs("#define digit_val(c)\t((c) - '0')\n", hfile);
-	else
-		fputs("#define digit_val(c)\t(digit_value[(unsigned char)(c)])\n", hfile);
-}
-
-
-
-/*
- * Output digit conversion table (if digits are not contiguous).
- */
-
-static void
-digit_convert(void)
-{
-	int maxdigit;
-	static char digit[] = "0123456789";
-	char *p;
-	int i;
-
-	maxdigit = 0;
-	for (p = digit ; *p ; p++)
-		if (*p > maxdigit)
-			maxdigit = *p;
-	fputs("extern const char digit_value[];\n", hfile);
-	fputs("\n\nconst char digit_value[] = {\n", cfile);
-	for (i = 0 ; i <= maxdigit ; i++) {
-		for (p = digit ; *p && *p != i ; p++);
-		if (*p == '\0')
-			p = digit;
-		fprintf(cfile, "      %ld,\n", (long)(p - digit));
-	}
-	fputs("};\n", cfile);
+		fputs(*pp, hfile);
+	fputs("#define digit_val(c)\t((c) - '0')\n", hfile);
 }
diff --git a/src/parser.c b/src/parser.c
index 4362f6a..deea702 100644
--- a/src/parser.c
+++ b/src/parser.c
@@ -1414,7 +1414,7 @@ STATIC int
 noexpand(char *text)
 {
 	char *p;
-	char c;
+	signed char c;
 
 	p = text;
 	while ((c = *p++) != '\0') {
diff --git a/src/parser.h b/src/parser.h
index fa58ed7..d0cf440 100644
--- a/src/parser.h
+++ b/src/parser.h
@@ -35,17 +35,17 @@
  */
 
 /* control characters in argument strings */
-#define CTL_FIRST '\201'	/* first 'special' character */
-#define CTLESC '\201'		/* escape next character */
-#define CTLVAR '\202'		/* variable defn */
-#define CTLENDVAR '\203'
-#define CTLBACKQ '\204'
+#define CTL_FIRST -127		/* first 'special' character */
+#define CTLESC -127		/* escape next character */
+#define CTLVAR -126		/* variable defn */
+#define CTLENDVAR -125
+#define CTLBACKQ -124
 #define CTLQUOTE 01		/* ored with CTLBACKQ code if in quotes */
-/*	CTLBACKQ | CTLQUOTE == '\205' */
-#define	CTLARI	'\206'		/* arithmetic expression */
-#define	CTLENDARI '\207'
-#define	CTLQUOTEMARK '\210'
-#define	CTL_LAST '\210'		/* last 'special' character */
+/*	CTLBACKQ | CTLQUOTE == 133 */
+#define	CTLARI -122		/* arithmetic expression */
+#define	CTLENDARI -121
+#define	CTLQUOTEMARK -120
+#define	CTL_LAST -120		/* last 'special' character */
 
 /* variable substitution byte (follows CTLVAR) */
 #define VSTYPE	0x0f		/* type of variable substitution */
diff --git a/src/show.c b/src/show.c
index 05af328..1b58de1 100644
--- a/src/show.c
+++ b/src/show.c
@@ -165,7 +165,7 @@ sharg(union node *arg, FILE *fp)
 	}
 	bqlist = arg->narg.backquote;
 	for (p = arg->narg.text ; *p ; p++) {
-		switch (*p) {
+		switch ((signed char)*p) {
 		case CTLESC:
 			putc(*++p, fp);
 			break;
@@ -306,7 +306,7 @@ trstring(char *s)
 		return;
 	putc('"', tracefile);
 	for (p = s ; *p ; p++) {
-		switch (*p) {
+		switch ((signed char)*p) {
 		case '\n':  c = 'n';  goto backslash;
 		case '\t':  c = 't';  goto backslash;
 		case '\r':  c = 'r';  goto backslash;