Don't recognize built-ins in contexts where they don't make sense.

Namely, it doesn't help to try and expand include if it's not followed
by parenthesis and a filename.

This should make applications like sendmail m4 scripts more sturdy for
unquoted machine names that happen to collide with built-ins.

The only drawback is that our m4 may now do intelligent things with scripts
that don't work on other systems.

4 files changed, 57 insertions, 41 deletions

diff --git a/usr.bin/m4/extern.h b/usr.bin/m4/extern.h
index aae0fabde18..5d44b105031 100644
--- a/usr.bin/m4/extern.h
+++ b/usr.bin/m4/extern.h
@@ -1,4 +1,4 @@
-/*	$OpenBSD: extern.h,v 1.13 1999/11/30 22:19:50 espie Exp $	*/
+/*	$OpenBSD: extern.h,v 1.14 2000/01/11 14:06:11 espie Exp $	*/
 /*	$NetBSD: extern.h,v 1.3 1996/01/13 23:25:24 pk Exp $	*/
 
 /*-
@@ -57,6 +57,9 @@ extern unsigned	hash __P((const char *));
 extern ndptr	lookup __P((const char *));
 extern void	remhash __P((const char *, int));
 
+/* main.c */
+extern void outputstr __P((const char *));
+
 /* misc.c */
 extern void	chrsave __P((int));
 extern char 	*compute_prevep __P((void));
diff --git a/usr.bin/m4/m4.1 b/usr.bin/m4/m4.1
index 646e9820f6e..25f4db51b4d 100644
--- a/usr.bin/m4/m4.1
+++ b/usr.bin/m4/m4.1
@@ -1,4 +1,4 @@
-.\"	@(#) $OpenBSD: m4.1,v 1.8 1999/11/17 15:31:53 espie Exp $
+.\"	@(#) $OpenBSD: m4.1,v 1.9 2000/01/11 14:06:11 espie Exp $
 .\"
 .\"
 .Dd January 26, 1993
@@ -30,16 +30,19 @@ parenthesis '('.  If the macro name is not followed by an open
 parenthesis it is processed with no arguments.
 .Pp
 Macro names consist of a leading alphabetic or underscore
-possibly followed by alphanumeric or underscore characters, therefore
-valid macro names match this pattern [a-zA-Z_][a-zA-Z0-9_]*.
+possibly followed by alphanumeric or underscore characters, e.g.,
+valid macro names match the pattern [a-zA-Z_][a-zA-Z0-9_]*.
 .Pp
 In arguments to macros, leading unquoted space, tab and newline
-characters are ignored.  To quote strings use left and right single
+characters are ignored.  To quote strings, use left and right single
 quotes (e.g., ` this is a string with a leading space').  You can change
 the quote characters with the
 .Ic changequote
 built-in macro.
 .Pp
+Some built-ins don't make any sense without arguments, and hence are not
+recognized as special when not followed by an open parenthesis.
+.Pp
 The options are as follows:
 .Bl -tag -width "-Dname[=value]xxx"
 .It Fl D Ns Ar name Ns Oo
diff --git a/usr.bin/m4/main.c b/usr.bin/m4/main.c
index b906b8ec873..13c8395a7ba 100644
--- a/usr.bin/m4/main.c
+++ b/usr.bin/m4/main.c
@@ -1,4 +1,4 @@
-/*	$OpenBSD: main.c,v 1.23 2000/01/05 16:06:14 espie Exp $	*/
+/*	$OpenBSD: main.c,v 1.24 2000/01/11 14:06:11 espie Exp $	*/
 /*	$NetBSD: main.c,v 1.12 1997/02/08 23:54:49 cgd Exp $	*/
 
 /*-
@@ -47,7 +47,7 @@ static char copyright[] =
 #if 0
 static char sccsid[] = "@(#)main.c	8.1 (Berkeley) 6/6/93";
 #else
-static char rcsid[] = "$OpenBSD: main.c,v 1.23 2000/01/05 16:06:14 espie Exp $";
+static char rcsid[] = "$OpenBSD: main.c,v 1.24 2000/01/11 14:06:11 espie Exp $";
 #endif
 #endif /* not lint */
 
@@ -93,7 +93,7 @@ struct keyblk keywrds[] = {	/* m4 keywords to be installed */
 	{ "sinclude",     SINCTYPE },
 	{ "define",       DEFITYPE },
 	{ "defn",         DEFNTYPE },
-	{ "divert",       DIVRTYPE },
+	{ "divert",       DIVRTYPE | NOARGS },
 	{ "expr",         EXPRTYPE },
 	{ "eval",         EXPRTYPE },
 	{ "substr",       SUBSTYPE },
@@ -102,9 +102,9 @@ struct keyblk keywrds[] = {	/* m4 keywords to be installed */
 	{ "len",          LENGTYPE },
 	{ "incr",         INCRTYPE },
 	{ "decr",         DECRTYPE },
-	{ "dnl",          DNLNTYPE },
-	{ "changequote",  CHNQTYPE },
-	{ "changecom",    CHNCTYPE },
+	{ "dnl",          DNLNTYPE | NOARGS },
+	{ "changequote",  CHNQTYPE | NOARGS },
+	{ "changecom",    CHNCTYPE | NOARGS },
 	{ "index",        INDXTYPE },
 #ifdef EXTENDED
 	{ "paste",        PASTTYPE },
@@ -112,24 +112,24 @@ struct keyblk keywrds[] = {	/* m4 keywords to be installed */
 #endif
 	{ "popdef",       POPDTYPE },
 	{ "pushdef",      PUSDTYPE },
-	{ "dumpdef",      DUMPTYPE },
-	{ "shift",        SHIFTYPE },
+	{ "dumpdef",      DUMPTYPE | NOARGS },
+	{ "shift",        SHIFTYPE | NOARGS },
 	{ "translit",     TRNLTYPE },
 	{ "undefine",     UNDFTYPE },
-	{ "undivert",     UNDVTYPE },
-	{ "divnum",       DIVNTYPE },
+	{ "undivert",     UNDVTYPE | NOARGS },
+	{ "divnum",       DIVNTYPE | NOARGS },
 	{ "maketemp",     MKTMTYPE },
-	{ "errprint",     ERRPTYPE },
-	{ "m4wrap",       M4WRTYPE },
-	{ "m4exit",       EXITTYPE },
+	{ "errprint",     ERRPTYPE | NOARGS },
+	{ "m4wrap",       M4WRTYPE | NOARGS },
+	{ "m4exit",       EXITTYPE | NOARGS },
 	{ "syscmd",       SYSCTYPE },
-	{ "sysval",       SYSVTYPE },
+	{ "sysval",       SYSVTYPE | NOARGS },
 
-#if defined(unix) || defined(__unix__)
-	{ "unix",         SELFTYPE },
+#if defined(unix) || defined(__unix__) 
+	{ "unix",         SELFTYPE | NOARGS },
 #else
 #ifdef vms
-	{ "vms",          SELFTYPE },
+	{ "vms",          SELFTYPE | NOARGS },
 #endif
 #endif
 };
@@ -278,14 +278,12 @@ macro()
 		if (t == '_' || isalpha(t)) {
 			putback(t);
 			s = token;
-			if ((p = inspect(s)) == nil) {
-				if (sp < 0)
-					while (*s)
-						putc(*s++, active);
-				else
-					while (*s)
-						chrsave(*s++);
-			}
+			p = inspect(s);
+			if (p != nil)
+				putback(l = gpbc());
+			if (p == nil || (l != LPAREN && 
+			    (p->type & NEEDARGS) != 0))
+				outputstr(s);
 			else {
 		/*
 		 * real thing.. First build a call frame:
@@ -301,7 +299,6 @@ macro()
 				pushs(p->name);	      /* macro name  */
 				pushs(ep);	      /* start next..*/
 
-				putback(l = gpbc());
 				if (l != LPAREN)  {   /* add bracks  */
 					putback(RPAREN);
 					putback(LPAREN);
@@ -343,14 +340,8 @@ macro()
 					chars[1] = EOS;
 					s = chars;
 				}
-				if (nlpar > 0) {
-					if (sp < 0)
-						while (*s)
-							putc(*s++, active);
-					else
-						while (*s)
-							chrsave(*s++);
-				}
+				if (nlpar > 0)
+					outputstr(s);
 			}
 			while (nlpar != 0);
 		}
@@ -439,6 +430,21 @@ macro()
 	}
 }
 
+/* 
+ * output string directly, without pushing it for reparses. 
+ */
+void
+outputstr(s)
+	const char *s;
+{
+	if (sp < 0)
+		while (*s)
+			putc(*s++, active);
+	else
+		while (*s)
+			chrsave(*s++);
+}
+
 /*
  * build an input token..
  * consider only those starting with _ or A-Za-z. This is a
@@ -490,7 +496,9 @@ initkwds()
 		p->name = keywrds[i].knam;
 		p->defn = null;
 		p->hv = h;
-		p->type = keywrds[i].ktyp | STATIC;
+		p->type = (keywrds[i].ktyp & TYPEMASK) | STATIC;
+		if ((keywrds[i].ktyp & NOARGS) == 0)
+			p->type |= NEEDARGS;
 	}
 }
 
diff --git a/usr.bin/m4/mdef.h b/usr.bin/m4/mdef.h
index d06da732c90..77bbaa3ab28 100644
--- a/usr.bin/m4/mdef.h
+++ b/usr.bin/m4/mdef.h
@@ -1,4 +1,4 @@
-/*	$OpenBSD: mdef.h,v 1.10 2000/01/05 16:06:14 espie Exp $	*/
+/*	$OpenBSD: mdef.h,v 1.11 2000/01/11 14:06:12 espie Exp $	*/
 /*	$NetBSD: mdef.h,v 1.7 1996/01/13 23:25:27 pk Exp $	*/
 
 /*
@@ -77,6 +77,8 @@
 #define TYPEMASK	63	/* Keep bits really corresponding to a type. */
 #define STATIC          128	/* Name is statically allocated, don't free. */
 #define RECDEF		256	/* Pure recursive def, don't expand it */
+#define NOARGS		512	/* builtin needs no args */
+#define NEEDARGS	1024	/* mark builtin that need args with this */
 
 /*
  * m4 special characters