Use locale for character classes instead of hardcoded values

Misc 8bit cleanup

4 files changed, 89 insertions, 37 deletions

diff --git a/lib/libc/regex/cclass.h b/lib/libc/regex/cclass.h
index cabe7fa..581909c 100644
--- a/lib/libc/regex/cclass.h
+++ b/lib/libc/regex/cclass.h
@@ -37,34 +37,26 @@
  *	@(#)cclass.h	8.3 (Berkeley) 3/20/94
  */
 
+
+typedef enum {CALNUM, CALPHA, CBLANK, CCNTRL, CDIGIT, CGRAPH,
+	      CLOWER, CPRINT, CPUNCT, CSPACE, CUPPER, CXDIGIT} citype;
+
 /* character-class table */
 static struct cclass {
 	char *name;
-	char *chars;
-	char *multis;
+	citype fidx;
 } cclasses[] = {
-	{"alnum",	"ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz\
-0123456789",				""},
-	{"alpha",	"ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz",
-					""},
-	{"blank",	" \t",		""},
-	{"cntrl",	"\007\b\t\n\v\f\r\1\2\3\4\5\6\16\17\20\21\22\23\24\
-\25\26\27\30\31\32\33\34\35\36\37\177",	""},
-	{"digit",	"0123456789",	""},
-	{"graph",	"ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz\
-0123456789!\"#$%&'()*+,-./:;<=>?@[\\]^_`{|}~",
-					""},
-	{"lower",	"abcdefghijklmnopqrstuvwxyz",
-					""},
-	{"print",	"ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz\
-0123456789!\"#$%&'()*+,-./:;<=>?@[\\]^_`{|}~ ",
-					""},
-	{"punct",	"!\"#$%&'()*+,-./:;<=>?@[\\]^_`{|}~",
-					""},
-	{"space",	"\t\n\v\f\r ",	""},
-	{"upper",	"ABCDEFGHIJKLMNOPQRSTUVWXYZ",
-					""},
-	{"xdigit",	"0123456789ABCDEFabcdef",
-					""},
-	{NULL,		0,		""}
+	{"alnum",       CALNUM},
+	{"alpha",       CALPHA},
+	{"blank",       CBLANK},
+	{"cntrl",       CCNTRL},
+	{"digit",       CDIGIT},
+	{"graph",       CGRAPH},
+	{"lower",       CLOWER},
+	{"print",       CPRINT},
+	{"punct",       CPUNCT},
+	{"space",       CSPACE},
+	{"upper",       CUPPER},
+	{"xdigit",      CXDIGIT},
+	{NULL,          }
 };
diff --git a/lib/libc/regex/engine.c b/lib/libc/regex/engine.c
index e7917b9..be569b1 100644
--- a/lib/libc/regex/engine.c
+++ b/lib/libc/regex/engine.c
@@ -1072,7 +1072,7 @@ int ch;
 {
 	static char pbuf[10];
 
-	if (isprint(ch) || ch == ' ')
+	if (isprint((uch)ch) || ch == ' ')
 		sprintf(pbuf, "%c", ch);
 	else
 		sprintf(pbuf, "\\%o", ch);
diff --git a/lib/libc/regex/regcomp.c b/lib/libc/regex/regcomp.c
index f7962ea..1828a3b 100644
--- a/lib/libc/regex/regcomp.c
+++ b/lib/libc/regex/regcomp.c
@@ -816,13 +816,12 @@ p_b_cclass(p, cs)
 register struct parse *p;
 register cset *cs;
 {
+	register int c;
 	register char *sp = p->next;
 	register struct cclass *cp;
 	register size_t len;
-	register char *u;
-	register char c;
 
-	while (MORE() && isalpha((unsigned char)PEEK()))
+	while (MORE() && isalpha((uch)PEEK()))
 		NEXT();
 	len = p->next - sp;
 	for (cp = cclasses; cp->name != NULL; cp++)
@@ -834,11 +833,72 @@ register cset *cs;
 		return;
 	}
 
-	u = cp->chars;
-	while ((c = *u++) != '\0')
-		CHadd(cs, c);
+	switch (cp->fidx) {
+	case CALNUM:
+		for (c = CHAR_MIN; c <= CHAR_MAX; c++)
+			if (isalnum((uch)c))
+				CHadd(cs, c);
+		break;
+	case CALPHA:
+		for (c = CHAR_MIN; c <= CHAR_MAX; c++)
+			if (isalpha((uch)c))
+				CHadd(cs, c);
+		break;
+	case CBLANK:
+		for (c = CHAR_MIN; c <= CHAR_MAX; c++)
+			if (isblank((uch)c))
+				CHadd(cs, c);
+		break;
+	case CCNTRL:
+		for (c = CHAR_MIN; c <= CHAR_MAX; c++)
+			if (iscntrl((uch)c))
+				CHadd(cs, c);
+		break;
+	case CDIGIT:
+		for (c = CHAR_MIN; c <= CHAR_MAX; c++)
+			if (isdigit((uch)c))
+				CHadd(cs, c);
+		break;
+	case CGRAPH:
+		for (c = CHAR_MIN; c <= CHAR_MAX; c++)
+			if (isgraph((uch)c))
+				CHadd(cs, c);
+		break;
+	case CLOWER:
+		for (c = CHAR_MIN; c <= CHAR_MAX; c++)
+			if (islower((uch)c))
+				CHadd(cs, c);
+		break;
+	case CPRINT:
+		for (c = CHAR_MIN; c <= CHAR_MAX; c++)
+			if (isprint((uch)c))
+				CHadd(cs, c);
+		break;
+	case CPUNCT:
+		for (c = CHAR_MIN; c <= CHAR_MAX; c++)
+			if (ispunct((uch)c))
+				CHadd(cs, c);
+		break;
+	case CSPACE:
+		for (c = CHAR_MIN; c <= CHAR_MAX; c++)
+			if (isspace((uch)c))
+				CHadd(cs, c);
+		break;
+	case CUPPER:
+		for (c = CHAR_MIN; c <= CHAR_MAX; c++)
+			if (isupper((uch)c))
+				CHadd(cs, c);
+		break;
+	case CXDIGIT:
+		for (c = CHAR_MIN; c <= CHAR_MAX; c++)
+			if (isxdigit((uch)c))
+				CHadd(cs, c);
+		break;
+	}
+#if 0
 	for (u = cp->multis; *u != '\0'; u += strlen(u) + 1)
 		MCadd(p, cs, u);
+#endif
 }
 
 /*
diff --git a/lib/libc/regex/regex2.h b/lib/libc/regex/regex2.h
index 64b6212..9560e05 100644
--- a/lib/libc/regex/regex2.h
+++ b/lib/libc/regex/regex2.h
@@ -121,13 +121,13 @@ typedef long sopno;
 typedef struct {
 	uch *ptr;		/* -> uch [csetsize] */
 	uch mask;		/* bit within array */
-	uch hash;		/* hash code */
+	short hash;             /* hash code */
 	size_t smultis;
 	char *multis;		/* -> char[smulti]  ab\0cd\0ef\0\0 */
 } cset;
 /* note that CHadd and CHsub are unsafe, and CHIN doesn't yield 0/1 */
-#define	CHadd(cs, c)	((cs)->ptr[(uch)(c)] |= (cs)->mask, (cs)->hash += (c))
-#define	CHsub(cs, c)	((cs)->ptr[(uch)(c)] &= ~(cs)->mask, (cs)->hash -= (c))
+#define CHadd(cs, c)    ((cs)->ptr[(uch)(c)] |= (cs)->mask, (cs)->hash += (uch)(c))
+#define CHsub(cs, c)    ((cs)->ptr[(uch)(c)] &= ~(cs)->mask, (cs)->hash -= (uch)(c))
 #define	CHIN(cs, c)	((cs)->ptr[(uch)(c)] & (cs)->mask)
 #define	MCadd(p, cs, cp)	mcadd(p, cs, cp)	/* regcomp() internal fns */
 #define	MCsub(p, cs, cp)	mcsub(p, cs, cp)
@@ -170,4 +170,4 @@ struct re_guts {
 
 /* misc utilities */
 #define	OUT	(CHAR_MAX+1)	/* a non-character value */
-#define	ISWORD(c)	(isalnum(c) || (c) == '_')
+#define ISWORD(c)       (isalnum((uch)(c)) || (c) == '_')