• (re)implement shebang execution file parsing; this makes the “and which

do not start with a "#!shell" sequence” part of mksh(1) true again; this
  was probably lost in mksh R21 or so when I decided/saw that our kernel
  always parses shebang lines (code written myself, not taken from pdksh
  again or MirBSD kernel, but verified against both)
• bom+shebang execution now works, no need setting EXECSHELL in the test
• bump version

check.t

@@ -1,4 +1,4 @@
-# $MirOS: src/bin/mksh/check.t,v 1.96 2007/04/15 10:58:55 tg Exp $
+# $MirOS: src/bin/mksh/check.t,v 1.97 2007/04/15 12:09:56 tg Exp $
 # $OpenBSD: bksl-nl.t,v 1.2 2001/01/28 23:04:56 niklas Exp $
 # $OpenBSD: history.t,v 1.5 2001/01/28 23:04:56 niklas Exp $
 # $OpenBSD: read.t,v 1.3 2003/03/10 03:48:16 david Exp $
@@ -7,7 +7,7 @@
 # http://www.research.att.com/~gsf/public/ifs.sh
 
 expected-stdout:
-	@(#)MIRBSD KSH R29 2007/03/14
+	@(#)MIRBSD KSH R29 2007/04/15
 description:
 	Check version of shell.
 category: pdksh
@@ -3931,6 +3931,7 @@ expected-stdout:
 name: utf8bom-2
 description:
 	Check that we can execute BOM-shebangs
+	XXX if the OS can already execute them, we lose
 category: pdksh
 env-setup: !FOO=BAR!
 stdin:
@@ -3939,8 +3940,6 @@ stdin:
 	print '#!/usr/bin/env perl\nprint "a=$ENV{FOO}\n";' >t3
 	print '#!/usr/bin/env perl\nprint "a=$ENV{FOO}\n";' >t4
 	chmod +x t?
-	EXECSHELL=$0
-	export EXECSHELL
 	./t1
 	./t2
 	./t3

exec.c

@@ -2,7 +2,7 @@
 
 #include "sh.h"
 
-__RCSID("$MirOS: src/bin/mksh/exec.c,v 1.26 2007/03/04 03:04:24 tg Exp $");
+__RCSID("$MirOS: src/bin/mksh/exec.c,v 1.27 2007/04/15 12:09:57 tg Exp $");
 
 static int comexec(struct op *, struct tbl *volatile, const char **,
     int volatile);
@@ -674,6 +674,9 @@ static void
 scriptexec(struct op *tp, const char **ap)
 {
 	const char *sh;
+	unsigned char *cp;
+	char buf[64];		/* 64 == MAXINTERP in MirBSD <sys/param.h> */
+	int fd;
 	union mksh_ccphack args, cap;
 
 	sh = str_val(global("EXECSHELL"));
@@ -683,6 +686,55 @@ scriptexec(struct op *tp, const char **ap)
 		sh = "/bin/sh";
 
 	*tp->args-- = tp->str;
+
+	if ((fd = open(tp->str, O_RDONLY)) >= 0) {
+		/* read first MAXINTERP octets from file */
+		if (read(fd, buf, sizeof (buf)) <= 0)
+			/* read error -> no good */
+			buf[0] = '\0';
+		close(fd);
+		/* scan for newline or NUL _before_ end of buffer */
+		cp = (unsigned char *)buf;
+		while ((char *)cp < (buf + sizeof (buf)))
+			if (*cp == '\0' || *cp == '\n') {
+				*cp = '\0';
+				break;
+			} else
+				++cp;
+		/* if the shebang line is longer than MAXINTERP, bail out */
+		if ((char *)cp >= (buf + sizeof (buf)))
+			goto noshebang;
+		/* skip UTF-8 Byte Order Mark, if present */
+		cp = (unsigned char *)buf;
+		if ((cp[0] == 0xEF) && (cp[1] == 0xBB) && (cp[2] == 0xBF))
+			cp += 3;
+		/* bail out if read error (above) or no shebang */
+		if ((cp[0] != '#') || (cp[1] != '!'))
+			goto noshebang;
+		cp += 2;
+		/* skip whitespace before shell name */
+		while (*cp == ' ' || *cp == '\t')
+			++cp;
+		/* just whitespace on the line? */
+		if (*cp == '\0')
+			goto noshebang;
+		/* no, we actually found an interpreter name */
+		sh = (char *)cp;
+		/* look for end of shell/interpreter name */
+		while (*cp != ' ' && *cp != '\t' && *cp != '\0')
+			++cp;
+		/* any arguments? */
+		if (*cp) {
+			*cp++ = '\0';
+			/* skip spaces before arguments */
+			while (*cp == ' ' || *cp == '\t')
+				++cp;
+			/* pass it all in ONE argument (historic reasons) */
+			if (*cp)
+				*tp->args-- = (char *)cp;
+		}
+	}
+ noshebang:
 	args.ro = tp->args;
 	*args.ro = sh;
 

sh.h

@@ -8,8 +8,8 @@
 /*	$OpenBSD: c_test.h,v 1.4 2004/12/20 11:34:26 otto Exp $	*/
 /*	$OpenBSD: tty.h,v 1.5 2004/12/20 11:34:26 otto Exp $	*/
 
-#define MKSH_SH_H_ID "$MirOS: src/bin/mksh/sh.h,v 1.122 2007/04/15 10:45:59 tg Exp $"
-#define MKSH_VERSION "R29 2007/03/14"
+#define MKSH_SH_H_ID "$MirOS: src/bin/mksh/sh.h,v 1.123 2007/04/15 12:09:57 tg Exp $"
+#define MKSH_VERSION "R29 2007/04/15"
 
 #if HAVE_SYS_PARAM_H
 #include <sys/param.h>