Release calc version 2.11.0t9.4.5

Release calc version 2.11.0t9.4.4
Release calc version 2.11.0t9.4.3
2025-08-22 01:23:28 +03:00 · 2017-05-21 15:38:33 -07:00 · 2017-05-21 15:38:33 -07:00 · 2017-05-21 15:38:33 -07:00 · 2017-05-21 15:38:32 -07:00
43 changed files with 1570 additions and 506 deletions
--- a/67
+++ b/67
@@ -31,7 +31,9 @@ The calc web site is located at:
 If you have tried all of the above and things still are not right,
 then it may be time to send in a bug report.  You can send bug reports to:

-    calc-tester@postofc.corp.sgi.com
+    calc-bugs at postofc dot corp dot sgi dot com
+
+    [[ Replace 'at' with @, 'dot' is with . and remove the spaces ]]

 When you send your report, please include the following information:

@@ -62,7 +64,12 @@ Stack traces from core dumps are useful to send as well.
 Send any comments, compiler warning messages, suggestions and most
 importantly, fixes (in the form of a context diff patch) to:

-    calc-tester@postofc.corp.sgi.com
+    calc-tester at postofc dot corp dot sgi dot com
+
+    [[ Replace 'at' with @, 'dot' is with . and remove the spaces ]]
+
+You should use the above calc-bugs address for bug reports, if you are
+not currently a member of the calc-tester mailing list.

 =-=

@@ -96,31 +103,36 @@ Known bugs:
      the terminal in a 'bad' state, as if stty -icanon -echo -echoe
      had been executed.

-    * Dec Alpha Linux compiling with gcc-2.95.1 (or gcc-2.95.2) and
-      -O2 fails the version 2.11.0t8.10 regression test with:
+    * Use of 'fmt' in the 2nd arg of printf() calls in c_sysinfo.c
+      cause some compilers to issue warnings.

-	make -s check
-	000: Beginning regression tests
-	001: Some of these tests may take a while ...
-	002: Within each section, output should be numbered sequentially
-	003: parsed global definitions
-	004: parsed vrfy()
-	005: parsed prob(str)
-	006: parsed getglobalvar()
-	007: parsed test_booleans()
-	008: parsed test_variables()
-	make: *** [check] Segmentation fault (core dumped)
+    We are sure some more bugs exist.  When you find them, please let
+    us know!  See the above for details on how to report and were to
+    EMail your bug reports and hopefully patches to fix them.

-      Other programs have reported problems when compiling -O2 with
-      gcc-2.95.1 on the Alpha and Mips.
+=-=

-      One work-a-round is to not compile with -O2 (perhaps just -O).
-      Another work-a-round is not use gcc-2.95.1.
+Other items of note:
+
+    * There is a bug in gcc-2.95 that causes calc, when compiled with -O2,
+      to fail the regression test.  The work-a-round is to compile with -O
+      or to use gcc-2.96 or later.
+
+    * There is a bug in some versions of the Dec/Compaq cc for the Alpha
+      where the following:
+
+	    #include <stdio.h>
+	    #define SVAL(a,b) (unsigned long)(0x ## a ## b ## ULL)
+	    main(){SVAL(b8a8aeb0,8168eadc);}
+
+      fails because it puts a space inside the concatenated hex.  Calc
+      has code that is affected by this bug.  This bug has been reported
+      to Compaq and may be fixed in the future.  A work-a-round is to
+      compile with cc -std0 or to use a later version of their compiler.

    * On a Digital UNIX V4.0F (Rev. 1229) on a 500 Mhz 21264, make check
      dies a horrible death starting in test 600 and 622 gives 100s of
-      messages for calc version 2.11.0t9.2 using the Dec's cc with -O2
-      and without -std0:
+      messages for calc version 2.11.0t9.4 using the Dec's cc with -O2:

 	600: Beginning test_bignums
 	601: muldivcheck 1
@@ -135,6 +147,13 @@ Known bugs:

 	it finally hangs at test 2000.

-    We are sure some more bugs exist.  When you find them, please let
-    us know!  See the above for details on how to report and were to
-    EMail your bug reports and hopefully patches to fix them.
+      The work-a-round is to compile calc without the optimizer.  If this
+      happens to you, try compiling without -O and without -O2.  I.e., in
+      the Makefile, set:
+
+	DEBUG= -g
+
+    * The sparcv9 support for 64 bit Solaris under gcc-2.96 is able
+      to compile calc, but calc dumps core very early on in startup.
+      It is said that sparcv9 support in gcc-2.96 is very unofficial.
+      There is no work-a-round for this compile problem.
--- a/162
+++ b/162
@@ -2,14 +2,14 @@ Following is the change from calc version 2.11.0t8.9.1 to date:

    The config("verbose_quit") will control the printing of the message:

-	Quit or abort executed
+	    Quit or abort executed

-    when a non-interactive ABORT, QUIT or EXIT is encounted.  By default,
-    config("verbose_quit") is TRUE and the message is printed.  If one does:
+	when a non-interactive ABORT, QUIT or EXIT is encounted.  By default,
+	config("verbose_quit") is TRUE and the message is printed.  If one does:

-	config("verbose_quit", 0)
+	    config("verbose_quit", 0)

-    the message is disabled.
+	the message is disabled.

    Added 8400 regression test set and test8400.cal to test the new
    quit and config("verbose_quit") functionality.
@@ -33,34 +33,34 @@ Following is the change from calc version 2.11.0t8.9.1 to date:
    When prompt() occurs while reading a file, it will take input from
    the terminal rather than taking it from a file.  For example:

-	/* This demonstrates the use of prompt() and some other things  */
-	config("verbose_quit", 0);
-	define getnumber() {
-	    local x;
-	    for (;;) {
-		x = eval(prompt(">>> "));
-		if (isnum(x))
-		    return x;
-		print "Not a number! Try again";
+	    /* This demonstrates the use of prompt() and some other things  */
+	    config("verbose_quit", 0);
+	    define getnumber() {
+		local x;
+		for (;;) {
+		    x = eval(prompt(">>> "));
+		    if (isnum(x))
+			return x;
+		    print "Not a number! Try again";
+		}
 	    }
-	}
-	print "This will display the sqrt of each number you enter";
-	print "Enter quit to stop";
-	for (;;) {
-	    print sqrt(getnumber());
-	}
-	print "Good bye";
+	    print "This will display the sqrt of each number you enter";
+	    print "Enter quit to stop";
+	    for (;;) {
+		print sqrt(getnumber());
+	    }
+	    print "Good bye";

-    Comments entered at input terminal level may be spread over several
-    lines.  For example:
+	Comments entered at input terminal level may be spread over several
+	lines.  For example:

-	/*
-	 * Assume that this calc script is called: comment.cal
-	 * Then these commands now work:
-	 *	cat comment.cal | calc
-	 *	calc < comment.cal
-	 */
-	print "Hello";
+	    /*
+	     * Assume that this calc script is called: comment.cal
+	     * Then these commands now work:
+	     *	cat comment.cal | calc
+	     *	calc < comment.cal
+	     */
+	    print "Hello";

    Added:

@@ -95,21 +95,20 @@ Following is the change from calc version 2.11.0t8.9.1 to date:
    GNU-readline.  Note that GNU-readline is not shipped with calc.
    His patch only provides the hooks to use it.  One must comment out:

-	USE_READLINE=
-	READLINE_LIB=
-	READLINE_INCLUDE=
+	    USE_READLINE=
+	    READLINE_LIB=
+	    READLINE_INCLUDE=

-    and comment in:
+	and comment in:

-	USE_READLINE= -DUSE_READLINE
-	READLINE_LIB= -lreadline -lhistory
-	READLINE_INCLUDE= -I/usr/include/readline
+	    USE_READLINE= -DUSE_READLINE
+	    READLINE_LIB= -lreadline -lhistory
+	    READLINE_INCLUDE= -I/usr/include/readline

-    in addition to pre-installing GNU-readline in your system to use
-    this facility.
+	in addition to pre-installing GNU-readline in your system to use
+	this facility.

-    Changed the "object already defined" math_error message to a
-    scanerror message.
+    Changed the "object already defined" math_error to a scanerror message.

    Removed the limit on the number of object types.

@@ -154,6 +153,87 @@ Following is the change from calc version 2.11.0t8.9.1 to date:
    Fixed the URLs found thruout the source and documentation which did
    not and in /, but should for performance and server load reasons.

+    Cleaned up and improved handling of "mat" and "obj".  The comma in:
+
+	    mat A[2], B[3];
+
+	is changed to whatever is appropriate in the context:
+
+	    + comma operator
+	    + separator of arguments in a function call
+	    + separator of arguments in a defintion
+	    etc.
+
+	The expression (mat A[2]), B[3] returns B[3], assuming B already
+	exists as something created by a statement like: global mat B[4].
+
+	What used to be done by the expression:
+
+	    mat A[2], B[3]
+
+	will now require something like:
+
+	    mat A[2], mat B[3]	or 	A = mat[2], B = mat[3]
+
+	For example, if obj point and obj pair are known types, the
+	following is now allowed:
+
+	    L = list(mat[2], mat[3], obj point, obj pair)
+
+	As another example, the following is allowed:
+
+	    define f(a = mat[2] = {3,4}) = 5 * a;
+
+	as well as the following:
+
+	    obj point {x,y}, PP = obj pair {A,B} = {obj point, obj point}
+
+	which creates two object types at compile time and when executed,
+	assigns a pair-object value to a variable PP.
+
+    Fixed a bug whereby a for loop would behave incorrectly.  For example:
+
+	    config("trace", 2),
+	    global x;
+	    define f() {for ( ; x > 0; x--) {print x;}}
+	    x = 5, f()
+
+	will stop after printing 1 instead of looping forever.
+
+    Added values l_format, which when CHECK_L_FORMAT is defined ahead
+    of including longlong.h will help detect when a system can deal with
+    'long long' but not '%lld' in printf.  If a system with 'long long'
+    uses '%ld' to print a 64 bit value, then l_format will be > 0;
+    othewise if "%lld" is required, l_format will be < 0.
+
+    Added HAVE_STRDUP Makefile variable as well as the have_strdup.c
+    program that forms the have_strdup.h file.  The have_strdup.h file
+    will define HAVE_STRDUP is the system has strdup().  If HAVE_STRDUP
+    is not defined, then calc will use calc_strdup() to simulate
+    the real strdup() function.
+
+    Calc no longer makes use of sys_errlist and sys_nerr.  Some systems
+    no longer suppor these values (even though they should from a
+    legacy prospective).  Calc now relies on the fact that strerror()
+    will return NULL of no such system error exists.  System errors >=
+    10000 will be considered calc errors instead.  The Makefile symbol
+    ERRNO_DECL has gone away as well as calc_errno.c and calc_errno.h.
+
+    On system errors are are not known to to the libc strerror() function,
+    strerror() will now print something such as:
+
+	    Unknown error 9999
+
+    Fixed some insure code inspection tool issues that were discovered
+    and investigated by Michel van der List <vanderlistmj@sbphrd.com>.
+
+    Made an effort to ensure that the v_subtype of VALUES are initialized
+    to V_NOSUBTYPE thruout the source code.
+
+    Established a separate calc-bugs address from the calc-tester
+    maining list.  Using anti-spam address forms in order to try and
+    stay under the radar of spammers as much as one can do so.
+

 Following is the change from calc version 2.11.0t8 to 2.11.0t8.9:

--- a/229
+++ b/229
@@ -273,6 +273,18 @@ HAVE_URANDOM=
 HAVE_GETRUSAGE=
 #HAVE_GETRUSAGE= -DHAVE_NO_GETRUSAGE

+# Determine if we have strdup()
+#
+# If HAVE_STRDUP is empty, this makefile will run the have_memmv program
+# to determine if strdup() is supported.  If HAVE_STRDUP is set to
+# -DHAVE_NO_STRDUP, then calc will use internal functions to simulate
+# the memory move function that does correct overlapping memory modes.
+#
+# If in doubt, leave HAVE_STRDUP empty and this Makefile will figure it out.
+#
+HAVE_STRDUP=
+#HAVE_STRDUP= -DHAVE_NO_STRDUP
+
 # Some architectures such as Sparc do not allow one to access 32 bit values
 # that are not alligned on a 32 bit boundary.
 #
@@ -290,39 +302,6 @@ HAVE_GETRUSAGE=
 ALIGN32= -DMUST_ALIGN32
 #ALIGN32= -UMUST_ALIGN32

-# On most machines: errno sys_errlist and sys_nerr are declared
-# by either <stdio.h> and/or <errno.h>.  But some systems declare
-# them somewhere else or do not declare them at all!
-#
-# If the system were doing a proper job in headers, this should declare them:
-#
-#	#include <stdio.h>
-#	#include <errno.h>
-#
-# But one some systems one must explicitly declare them as:
-#
-#	extern int errno;
-#	extern const char *const sys_errlist[];
-#	extern int sys_nerr;
-#
-# and on some old systems they must be explicitly and incorrectly declared as:
-#
-#	extern int errno;
-#	extern char *sys_errlist[];
-#	extern int sys_nerr;
-#
-# ERRNO_DECL=			let calc_errno.c determine how to declare them
-# ERRNO_DECL= -DERRNO_NO_DECL	headers declare them correctly
-# ERRNO_DECL= -DERRNO_STD_DECL	one must explicitly declare then
-# ERRNO_DECL= -DERRNO_OLD_DECL	one must explicitly & incorrectly declare them
-#
-# When in doubt, be safe leave ERRNO_DECL empty.
-#
-ERRNO_DECL=
-#ERRNO_DECL= -DERRNO_NO_DECL
-#ERRNO_DECL= -DERRNO_STD_DECL
-#ERRNO_DECL= -DERRNO_OLD_DECL
-
 # where to install binary files
 #
 BINDIR= /usr/local/bin
@@ -762,7 +741,7 @@ CC= ${PURIFY} ${LCC}
 #
 ###
 #
-# RS6000 set
+# AIX RS/6000 set
 #
 #CCWARN=
 #CCOPT= ${DEBUG} ${NO_SHARED}
@@ -930,9 +909,10 @@ BUILD_H_SRC= align32.h args.h calcerr.h conf.h endian_calc.h \
 	fposval.h have_const.h have_fpos.h have_malloc.h \
 	have_memmv.h have_newstr.h have_offscl.h have_posscl.h \
 	have_stdlib.h have_string.h have_times.h have_uid_t.h \
-	have_unistd.h longbits.h longlong.h terminal.h calc_errno.h \
+	have_unistd.h longbits.h longlong.h terminal.h \
 	have_ustat.h have_getsid.h have_getpgid.h \
-	have_gettime.h have_getprid.h have_urandom.h have_rusage.h
+	have_gettime.h have_getprid.h have_urandom.h have_rusage.h \
+	have_strdup.h

 # we build these .c files during the make
 #
@@ -944,9 +924,9 @@ BUILD_C_SRC= calcerr.c
 #
 UTIL_C_SRC= align32.c endian.c longbits.c have_newstr.c have_uid_t.c \
 	have_const.c have_stdvs.c have_varvs.c fposval.c have_fpos.c \
-	longlong.c have_offscl.c have_posscl.c have_memmv.c calc_errno.c \
+	longlong.c have_offscl.c have_posscl.c have_memmv.c \
 	have_ustat.c have_getsid.c have_getpgid.c \
-	have_gettime.c have_getprid.c have_rusage.c
+	have_gettime.c have_getprid.c have_rusage.c have_strdup.c

 # these awk and sed tools are used in the process of building BUILD_H_SRC
 # and BUILD_C_SRC
@@ -960,22 +940,22 @@ UTIL_MISC_SRC= calcerr_h.sed calcerr_h.awk calcerr_c.sed calcerr_c.awk \
 #
 UTIL_OBJS= endian.o longbits.o have_newstr.o have_uid_t.o \
 	have_const.o fposval.o have_fpos.o longlong.o try_strarg.o \
-	have_stdvs.o have_varvs.o have_posscl.o have_memmv.o calc_errno.o \
+	have_stdvs.o have_varvs.o have_posscl.o have_memmv.o \
 	have_ustat.o have_getsid.o have_getpgid.o \
-	have_gettime.o have_getprid.o ver_calc.o have_rusage.o
+	have_gettime.o have_getprid.o ver_calc.o have_rusage.o have_strdup.o

 # these temp files may be created (and removed) during the build of BUILD_C_SRC
 #
 UTIL_TMP= ll_tmp fpos_tmp fposv_tmp const_tmp uid_tmp newstr_tmp vs_tmp \
-	calc_errno_tmp memmv_tmp offscl_tmp posscl_tmp newstr_tmp \
-	getsid_tmp gettime_tmp getprid_tmp rusage_tmp
+	memmv_tmp offscl_tmp posscl_tmp newstr_tmp \
+	getsid_tmp gettime_tmp getprid_tmp rusage_tmp strdup_tmp

 # these utility progs may be used in the process of building BUILD_H_SRC
 #
 UTIL_PROGS= align32 fposval have_uid_t longlong have_const \
-	endian longbits have_newstr have_stdvs have_varvs calc_errno \
+	endian longbits have_newstr have_stdvs have_varvs \
 	have_ustat have_getsid have_getpgid \
-	have_gettime have_getprid ver_calc
+	have_gettime have_getprid ver_calc have_strdup

 # These files are required by the regress.cal regression test.
 #
@@ -1128,13 +1108,14 @@ calc.1: calc.man ${MAKE_FILE}
 ##

 calc.o: calc.c ${MAKE_FILE}
-	${CC} ${CFLAGS} ${CCOPT} ${ALLOW_CUSTOM} -c calc.c
+	${CC} ${CFLAGS} ${ALLOW_CUSTOM} -c calc.c

 custom.o: custom.c ${MAKE_FILE}
 	${CC} ${CFLAGS} ${ALLOW_CUSTOM} -c custom.c

 hist.o: hist.c ${MAKE_FILE}
-	${CC} ${CFLAGS} ${TERMCONTROL} ${USE_READLINE} ${READLINE_INCLUDE} -c hist.c
+	${CC} ${CFLAGS} ${TERMCONTROL} ${USE_READLINE} ${READLINE_INCLUDE} \
+	    -c hist.c

 func.o: func.c ${MAKE_FILE}
 	${CC} ${CFLAGS} ${ALLOW_CUSTOM} -c func.c
@@ -1535,6 +1516,8 @@ longlong.h: longlong.c have_stdlib.h have_string.h ${MAKE_FILE}
 	else \
 	    echo '#undef HAVE_LONGLONG' >> longlong.h; \
 	    echo '#define LONGLONG_BITS 0  /* no */' >> longlong.h; \
+	    echo '#undef LL_FORMAT' >> longlong.h; \
+	    echo '#undef L_FORMAT' >> longlong.h; \
 	fi
 	${Q}echo '' >> longlong.h
 	${Q}echo '' >> longlong.h
@@ -2174,6 +2157,45 @@ have_rusage.h: have_rusage.c ${MAKE_FILE}
 	    true; \
 	fi

+have_strdup.h: have_strdup.c ${MAKE_FILE}
+	-${Q}rm -f have_strdup have_strdup.o strdup_tmp have_strdup.h
+	${Q}echo 'forming have_strdup.h'
+	${Q}echo '/*' > have_strdup.h
+	${Q}echo ' * DO NOT EDIT -- generated by the Makefile' >> have_strdup.h
+	${Q}echo ' */' >> have_strdup.h
+	${Q}echo '' >> have_strdup.h
+	${Q}echo '' >> have_strdup.h
+	${Q}echo '#if !defined(__HAVE_RUSAGE_H__)' >> have_strdup.h
+	${Q}echo '#define __HAVE_RUSAGE_H__' >> have_strdup.h
+	${Q}echo '' >> have_strdup.h
+	${Q}echo '' >> have_strdup.h
+	${Q}echo '/* do we have or want getstrdup()? */' >> have_strdup.h
+	-${Q}rm -f have_strdup.o have_strdup
+	-${Q}${LCC} ${ICFLAGS} ${HAVE_STRDUP} have_strdup.c -c 2>/dev/null; \
+	     true
+	-${Q}${LCC} ${ILDFLAGS} have_strdup.o -o have_strdup 2>/dev/null; true
+	-${Q}${SHELL} -c "./have_strdup > strdup_tmp 2>/dev/null" \
+	    >/dev/null 2>&1; true
+	-${Q}if [ -s strdup_tmp ]; then \
+	    cat strdup_tmp >> have_strdup.h; \
+	else \
+	    echo '#undef HAVE_STRDUP /* no */' >> have_strdup.h; \
+	fi
+	${Q}echo '' >> have_strdup.h
+	${Q}echo '' >> have_strdup.h
+	${Q}echo '#endif /* !__HAVE_RUSAGE_H__ */' >> have_strdup.h
+	-${Q}rm -f have_strdup have_strdup.o strdup_tmp
+	${Q}echo 'have_strdup.h formed'
+	-@if [ -z "${Q}" ]; then \
+	    echo ''; \
+	    echo '=-=-= start of $@ =-=-='; \
+	    cat $@; \
+	    echo '=-=-= end of $@ =-=-='; \
+	    echo ''; \
+	else \
+	    true; \
+	fi
+
 args.h: have_stdvs.c have_varvs.c have_string.h have_unistd.h have_string.h
 	-${Q}rm -f args.h have_args
 	${Q}echo 'forming args.h'
@@ -2233,98 +2255,6 @@ args.h: have_stdvs.c have_varvs.c have_string.h have_unistd.h have_string.h
 	    true; \
 	fi

-calc_errno.h: calc_errno.c ${MAKE_FILE}
-	-${Q}rm -f calc_errno.h calc_errno calc_errno_tmp
-	${Q}echo 'forming calc_errno.h'
-	${Q}echo '/*' > calc_errno.h
-	${Q}echo ' * DO NOT EDIT -- generated by the Makefile' >> calc_errno.h
-	${Q}echo ' */' >> calc_errno.h
-	${Q}echo '' >> calc_errno.h
-	${Q}echo '' >> calc_errno.h
-	${Q}echo '#if !defined(__CALC_ERRNO_H__)' >> calc_errno.h
-	${Q}echo '#define __CALC_ERRNO_H__' >> calc_errno.h
-	${Q}echo '' >> calc_errno.h
-	${Q}echo '' >> calc_errno.h
-	-${Q}if [ X"${ERRNO_DECL}" = X"-DERRNO_NO_DECL" ]; then \
-	    echo '/*' >> calc_errno.h; \
-	    echo ' * The calc Makefile explicitly told us' >> calc_errno.h; \
-	    echo ' * how to declare errno and friends.' >> calc_errno.h; \
-	    echo ' */' >> calc_errno.h; \
-	    echo '' >> calc_errno.h; \
-	    echo '#include <stdio.h>' >> calc_errno.h; \
-	    echo '#include <errno.h>' >> calc_errno.h; \
-	elif [ X"${ERRNO_DECL}" = X"-DERRNO_STD_DECL" ]; then \
-	    echo '/*' >> calc_errno.h; \
-	    echo ' * The calc Makefile explicitly told us' >> calc_errno.h; \
-	    echo ' * how to declare errno and friends.' >> calc_errno.h; \
-	    echo ' */' >> calc_errno.h; \
-	    echo '' >> calc_errno.h; \
-	    echo 'extern int errno;			' \
-		 '/* last system error */' >> calc_errno.h; \
-	    echo 'extern const char *const sys_errlist[];' \
-		 ' /* system error messages*/' >> calc_errno.h; \
-	    echo 'extern int sys_nerr;			' \
-		 '/* number of system errors */' >> calc_errno.h; \
-	elif [ X"${ERRNO_DECL}" = X"-DERRNO_OLD_DECL" ]; then \
-	    echo '/*' >> calc_errno.h; \
-	    echo ' * The calc Makefile explicitly told us' >> calc_errno.h; \
-	    echo ' * how to declare errno and friends.' >> calc_errno.h; \
-	    echo ' */' >> calc_errno.h; \
-	    echo '' >> calc_errno.h; \
-	    echo 'extern int errno;		' \
-		 '/* last system error */' >> calc_errno.h; \
-	    echo 'extern char *sys_errlist[];' \
-		 '	 /* system error messages*/' >> calc_errno.h; \
-	    echo 'extern int sys_nerr;		' \
-		 '/* number of system errors */' >> calc_errno.h; \
-	else \
-	    ${LCC} ${ICFLAGS} -DTRY_ERRNO_NO_DECL \
-		  calc_errno.c -o calc_errno 2>calc_errno_tmp; \
-	    if [ -x ./calc_errno ]; then \
-		./calc_errno >> calc_errno.h; \
-	    else \
-		${LCC} ${ICFLAGS} -DTRY_ERRNO_STD_DECL \
-		      calc_errno.c -o calc_errno 2>calc_errno_tmp; \
-		if [ -x ./calc_errno ]; then \
-		    ./calc_errno >> calc_errno.h; \
-		else \
-		    ${LCC} ${ICFLAGS} -DTRY_ERRNO_OLD_DECL \
-			  calc_errno.c -o calc_errno 2>calc_errno_tmp; \
-		    if [ -x ./calc_errno ]; then \
-			./calc_errno >> calc_errno.h; \
-		    else \
-	    echo '/*' >> calc_errno.h; \
-	    echo ' * We were unable to to determine' >> calc_errno.h; \
-	    echo ' * how to declare errno and friends.' >> calc_errno.h; \
-	    echo ' * So we will guess this will work' >> calc_errno.h; \
-	    echo ' * and hope for the best.' >> calc_errno.h; \
-	    echo ' */' >> calc_errno.h; \
-	    echo '' >> calc_errno.h; \
-	    echo 'extern int errno;			' \
-		 '/* last system error */' >> calc_errno.h; \
-	    echo 'extern const char *const sys_errlist[];' \
-		 ' /* system error messages*/' >> calc_errno.h; \
-	    echo 'extern int sys_nerr;			' \
-		 '/* number of system errors */' >> calc_errno.h; \
-		    fi; \
-		fi; \
-	    fi; \
-	fi
-	${Q}echo '' >> calc_errno.h
-	${Q}echo '' >> calc_errno.h
-	${Q}echo '#endif /* !__CALC_ERRNO_H__ */' >> calc_errno.h
-	-${Q}rm -f calc_errno calc_errno_tmp
-	${Q}echo 'calc_errno.h formed'
-	-@if [ -z "${Q}" ]; then \
-	    echo ''; \
-	    echo '=-=-= start of $@ =-=-='; \
-	    cat $@; \
-	    echo '=-=-= end of $@ =-=-='; \
-	    echo ''; \
-	else \
-	    true; \
-	fi
-
 calcerr.h: calcerr.tbl calcerr_h.sed calcerr_h.awk ${MAKE_FILE}
 	-${Q}rm -f calerr.h
 	${Q}echo 'forming calcerr.h'
@@ -2700,7 +2630,6 @@ env:
 	@echo "HAVE_GETPRID=${HAVE_GETPRID}"; echo ""
 	@echo "HAVE_URANDOM=${HAVE_URANDOM}"; echo ""
 	@echo "ALIGN32=${ALIGN32}"; echo ""
-	@echo "ERRNO_DECL=${ERRNO_DECL}"; echo ""
 	@echo "BINDIR=${BINDIR}"; echo ""
 	@echo "TOPDIR=${TOPDIR}"; echo ""
 	@echo "LIBDIR=${LIBDIR}"; echo ""
@@ -3027,6 +2956,8 @@ install: calc libcalc.a ${LIB_H_SRC} ${BUILD_H_SRC} calc.1
 	-rm -f ${LIBDIR}/stdarg.h stdarg.h
 	-rm -f ${LIBDIR}/prototype.h prototype.h
 	-rm -f ${LIBDIR}/libcalcerr.a libcalcerr.a
+	-rm -f ${LIBDIR}/calc_errno.h calc_errno.h ${INCDIRCALC}/calc_errno.h
+	-rm -f calc_errno.c calc_errno.o calc_errno
 	${V} echo '=-=-=-=-= end of $@ rule =-=-=-=-='

 ##
@@ -3048,6 +2979,7 @@ addop.o: config.h
 addop.o: endian_calc.h
 addop.o: func.h
 addop.o: hash.h
+addop.o: have_const.h
 addop.o: have_malloc.h
 addop.o: have_memmv.h
 addop.o: have_newstr.h
@@ -3104,6 +3036,7 @@ blkcpy.o: config.h
 blkcpy.o: endian_calc.h
 blkcpy.o: file.h
 blkcpy.o: hash.h
+blkcpy.o: have_const.h
 blkcpy.o: have_fpos.h
 blkcpy.o: have_malloc.h
 blkcpy.o: have_memmv.h
@@ -3192,7 +3125,6 @@ calc.o: symbol.h
 calc.o: token.h
 calc.o: value.h
 calc.o: zmath.h
-calc_errno.o: calc_errno.c
 calcerr.o: calcerr.c
 calcerr.o: calcerr.h
 calcerr.o: have_const.h
@@ -3208,6 +3140,7 @@ codegen.o: config.h
 codegen.o: endian_calc.h
 codegen.o: func.h
 codegen.o: hash.h
+codegen.o: have_const.h
 codegen.o: have_malloc.h
 codegen.o: have_memmv.h
 codegen.o: have_newstr.h
@@ -3292,6 +3225,7 @@ const.o: config.h
 const.o: const.c
 const.o: endian_calc.h
 const.o: hash.h
+const.o: have_const.h
 const.o: have_malloc.h
 const.o: have_memmv.h
 const.o: have_newstr.h
@@ -3346,6 +3280,7 @@ file.o: file.c
 file.o: file.h
 file.o: fposval.h
 file.o: hash.h
+file.o: have_const.h
 file.o: have_fpos.h
 file.o: have_malloc.h
 file.o: have_memmv.h
@@ -3370,7 +3305,6 @@ func.o: alloc.h
 func.o: block.h
 func.o: byteswap.h
 func.o: calc.h
-func.o: calc_errno.h
 func.o: calcerr.h
 func.o: cmath.h
 func.o: config.h
@@ -3447,6 +3381,7 @@ have_rusage.o: have_rusage.c
 have_stdvs.o: have_stdvs.c
 have_stdvs.o: have_string.h
 have_stdvs.o: have_unistd.h
+have_strdup.o: have_strdup.c
 have_uid_t.o: have_uid_t.c
 have_uid_t.o: have_unistd.h
 have_ustat.o: have_ustat.c
@@ -3463,6 +3398,7 @@ help.o: conf.h
 help.o: config.h
 help.o: endian_calc.h
 help.o: hash.h
+help.o: have_const.h
 help.o: have_malloc.h
 help.o: have_memmv.h
 help.o: have_newstr.h
@@ -3488,10 +3424,12 @@ hist.o: cmath.h
 hist.o: config.h
 hist.o: endian_calc.h
 hist.o: hash.h
+hist.o: have_const.h
 hist.o: have_malloc.h
 hist.o: have_memmv.h
 hist.o: have_newstr.h
 hist.o: have_stdlib.h
+hist.o: have_strdup.h
 hist.o: have_string.h
 hist.o: have_unistd.h
 hist.o: hist.c
@@ -3516,6 +3454,7 @@ input.o: conf.h
 input.o: config.h
 input.o: endian_calc.h
 input.o: hash.h
+input.o: have_const.h
 input.o: have_malloc.h
 input.o: have_memmv.h
 input.o: have_newstr.h
@@ -3546,6 +3485,7 @@ label.o: config.h
 label.o: endian_calc.h
 label.o: func.h
 label.o: hash.h
+label.o: have_const.h
 label.o: have_malloc.h
 label.o: have_memmv.h
 label.o: have_newstr.h
@@ -3580,6 +3520,7 @@ lib_calc.o: have_malloc.h
 lib_calc.o: have_memmv.h
 lib_calc.o: have_newstr.h
 lib_calc.o: have_stdlib.h
+lib_calc.o: have_strdup.h
 lib_calc.o: have_string.h
 lib_calc.o: have_unistd.h
 lib_calc.o: label.h
@@ -3675,6 +3616,7 @@ math_error.o: cmath.h
 math_error.o: config.h
 math_error.o: endian_calc.h
 math_error.o: hash.h
+math_error.o: have_const.h
 math_error.o: have_malloc.h
 math_error.o: have_memmv.h
 math_error.o: have_newstr.h
@@ -3725,6 +3667,7 @@ obj.o: config.h
 obj.o: endian_calc.h
 obj.o: func.h
 obj.o: hash.h
+obj.o: have_const.h
 obj.o: have_malloc.h
 obj.o: have_memmv.h
 obj.o: have_newstr.h
@@ -4030,6 +3973,7 @@ string.o: cmath.h
 string.o: config.h
 string.o: endian_calc.h
 string.o: hash.h
+string.o: have_const.h
 string.o: have_malloc.h
 string.o: have_memmv.h
 string.o: have_newstr.h
@@ -4055,6 +3999,7 @@ symbol.o: config.h
 symbol.o: endian_calc.h
 symbol.o: func.h
 symbol.o: hash.h
+symbol.o: have_const.h
 symbol.o: have_malloc.h
 symbol.o: have_memmv.h
 symbol.o: have_newstr.h
@@ -4084,6 +4029,7 @@ token.o: cmath.h
 token.o: config.h
 token.o: endian_calc.h
 token.o: hash.h
+token.o: have_const.h
 token.o: have_malloc.h
 token.o: have_memmv.h
 token.o: have_newstr.h
@@ -4143,6 +4089,7 @@ version.o: cmath.h
 version.o: config.h
 version.o: endian_calc.h
 version.o: hash.h
+version.o: have_const.h
 version.o: have_malloc.h
 version.o: have_memmv.h
 version.o: have_newstr.h
--- a/12
+++ b/12
@@ -74,7 +74,9 @@ for a wish/todo list.  Code contributions are welcome.

 To join the calc-tester mailing list.  Send a request to:

-     calc-tester-request@postofc.corp.sgi.com
+    calc-tester-request at postofc dot corp dot sgi dot com
+
+    [[ Replace 'at' with @, 'dot' is with . and remove the spaces ]]

 Your message body (not the subject) should consist of:

@@ -85,6 +87,14 @@ Your message body (not the subject) should consist of:
 where ``address'' is your EMail address and ``your_full_name'' is
 your full name.

+Calc bug reports, however should be sent to:
+
+    calc-bugs at postofc dot corp dot sgi dot com
+
+    [[ Replace 'at' with @, 'dot' is with . and remove the spaces ]]
+
+but see the BUGS file first.
+
 The calc web site is located at:

 	http://reality.sgi.com/chongo/tech/comp/calc/
--- a/addop.c
+++ b/addop.c
@@ -114,6 +114,7 @@ beginfunc(char *name, BOOL newflag)
 	fp->f_localcount = 0;
 	fp->f_opcodecount = 0;
 	fp->f_savedvalue.v_type = V_NULL;
+	fp->f_savedvalue.v_subtype = V_NOSUBTYPE;
 	fp->f_name = namestr(&funcnames, newindex);
 	curfunc = fp;
 	initlocals();
--- a/assocfunc.c
+++ b/assocfunc.c
@@ -80,6 +80,7 @@ associndex(ASSOC *ap, BOOL create, long dim, VALUE *indices)
 	 */
 	if (!create) {
 		val.v_type = V_NULL;
+		val.v_subtype = V_NOSUBTYPE;
 		return &val;
 	}

@@ -292,6 +293,7 @@ assoccopy(ASSOC *oldap)
 			ep->e_dim = oldep->e_dim;
 			ep->e_hash = oldep->e_hash;
 			ep->e_value.v_type = V_NULL;
+			ep->e_value.v_subtype = V_NOSUBTYPE;
 			for (i = 0; i < ep->e_dim; i++)
 				copyvalue(&oldep->e_indices[i], &ep->e_indices[i]);
 			copyvalue(&oldep->e_value, &ep->e_value);
--- a/blkcpy.c
+++ b/blkcpy.c
@@ -330,6 +330,7 @@ copyblk2mat(BLOCK *blk, long ssi, long num, MATRIX *dmat, long dsi)
 	i = num;
 	while (i-- > 0) {
 		vp->v_type = V_NUM;
+		vp->v_subtype = V_NOSUBTYPE;
 		vp->v_num = itoq((long) *op++);
 		vp++;
 	}
--- a/calc.c
+++ b/calc.c
@@ -39,9 +39,6 @@
 /*
 * static definitions and functions
 */
-static char *usage = "usage: %s [-c] [-C] [-d] [-e] [-h] [-i] [-m mode]\n"
-		     "\t[-D calc_debug[:lib_debug:[user_debug]]]\n"
-		     "\t[-n] [-p] [-q] [-u] [-v] [[--] calc_cmd ...]\n";
 static void intint(int arg);	/* interrupt routine */


@@ -153,7 +150,11 @@ main(int argc, char **argv)
 			 * we are too early in processing to call
 			 * libcalc_call_me_last() - nothing to cleanup
 			 */
-			fprintf(stderr, usage, program);
+			fprintf(stderr,
+			  "usage: %s [-c] [-C] [-d] [-e] [-h] [-i] [-m mode]\n"
+			  "\t[-D calc_debug[:lib_debug:[user_debug]]]\n"
+			  "\t[-n] [-p] [-q] [-u] [-v] [[--] calc_cmd ...]\n",
+			  program);
 			exit(1);
 		}
 	}
--- a/calc.h
+++ b/calc.h
@@ -14,6 +14,8 @@

 #include "value.h"

+#include "have_const.h"
+

 /*
 * Configuration definitions
@@ -140,6 +142,7 @@ extern void givehelp(char *type);
 extern void libcalc_call_me_first(void);
 extern void libcalc_call_me_last(void);
 extern void showerrors(void);
+extern char *calc_strdup(CONST char *);

 /*
 * Initialization
--- a/calc.man
+++ b/calc.man
@@ -570,7 +570,11 @@ The majority of
 was written by David I. Bell.
 .sp
 .B Calc
-archives and calc-tester mailing list maintained by Landon Curt Noll.
+The Calc primary mirror, calc mailing list and calc bug report
+processing is performed by Landon Curt Noll.
+.sp
+Landon Curt Noll maintains the master reference source, performs
+release control functions as well as other calc maintenance functions.
 .sp
 Thanks for suggestions and encouragement from Peter Miller,
 Neil Justusson, and Landon Noll.
@@ -610,9 +614,36 @@ scripts that you would like you see included
 in future distributions to:
 .sp
 .in +0.5i
-calc-tester@postofc.corp.sgi.com
+.nf
+calc-tester at postofc dot corp dot sgi dot com
+
+[[ Replace 'at' with @, 'dot' is with . and remove the spaces ]]
+.fi
 .in -0.5i
 .sp
+Bug reports are sent to:
+.in +0.5i
+.nf
+calc-bugs at postofc dot corp dot sgi dot com
+
+[[ Replace 'at' with @, 'dot' is with . and remove the spaces ]]
+.fi
+.in -0.5i
+.sp
+See the
+.I BUGS
+source file or use the 
+.I calc
+command:
+.sp
+.in +0.5i
+.nf
+help bugs
+.fi
+.in -0.5i
+.sp
+for more information about bug reporting.
+.sp
 Landon Noll maintains the the
 .B calc
 web site is located at:
@@ -626,7 +657,11 @@ One may join the
 testing group by sending a request to:
 .sp
 .in +0.5i
-calc-tester-request@postofc.corp.sgi.com
+.nf
+calc-tester-request at postofc dot corp dot sgi dot com
+
+[[ Replace 'at' with @, 'dot' is with . and remove the spaces ]]
+.fi
 .in -0.5i
 .sp
 Your message body (not the subject) should consist of:
--- a/calc_errno.c
+++ b/calc_errno.c
@@ -1,123 +0,0 @@
-/*
- * errno - Determine how to declare errno, sys_errlist and sys_nerr
- *
- * On most machines: errno sys_errlist and sys_nerr are declared
- * by either <stdio.h> and/or <errno.h>.  But some systems declare
- * them somewhere else or do not declare them at all!
- *
- * If the system were doing a proper job in headers, this should declare them:
- *
- *	#include <stdio.h>
- *	#include <errno.h>
- *
- * But one some systems one must explicitly declare them as:
- *
- *	extern int errno;
- *	extern const char *const sys_errlist[];
- *	extern int sys_nerr;
- *
- * and on some old systems they must be explicitly and incorrectly declared as:
- *
- *	extern int errno;
- *	extern char *sys_errlist[];
- *	extern int sys_nerr;
- *
- * The purpose of this utility is try and find the right way to declare
- * them and to output the middle of a header file called calc_errno.h.
- */
-
-/*
- * Copyright (c) 1999 by Landon Curt Noll.  All Rights Reserved.
- *
- * Permission to use, copy, modify, and distribute this software and
- * its documentation for any purpose and without fee is hereby granted,
- * provided that the above copyright, this permission notice and text
- * this comment, and the disclaimer below appear in all of the following:
- *
- *	supporting documentation
- *	source copies
- *	source works derived from this source
- *	binaries derived from this source or from derived source
- *
- * LANDON CURT NOLL DISCLAIMS ALL WARRANTIES WITH REGARD TO THIS SOFTWARE,
- * INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS. IN NO
- * EVENT SHALL LANDON CURT NOLL BE LIABLE FOR ANY SPECIAL, INDIRECT OR
- * CONSEQUENTIAL DAMAGES OR ANY DAMAGES WHATSOEVER RESULTING FROM LOSS OF
- * USE, DATA OR PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR
- * OTHER TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR
- * PERFORMANCE OF THIS SOFTWARE.
- *
- * chongo was here	/\../\
- */
-
-#include <stdio.h>
-#if defined(TRY_ERRNO_NO_DECL)
-#  include <errno.h>
-#endif /* TRY_ERRNO_NO_DECL */
-
-int
-main(void)
-{
-
-#if defined(TRY_ERRNO_NO_DECL)
-	printf("/*\n");
-	printf(" * The following comments were produced by calc_errno\n");
-	printf(" * in an effort to see if these values were correctly\n");
-	printf(" * declared when calc_errno.c was compiled.\n");
-	printf(" */\n\n");
-	printf("/* Initially errno is %d */\n", errno);
-	printf("/* There are %d entries in sys_errlist[] */\n", sys_nerr);
-	printf("/* The 2nd sys_errlist entry is \"%s\" */\n\n", sys_errlist[1]);
-	printf("/*\n");
-	printf(" * Based on the above, calc_errno now knows\n");
-	printf(" * how to declare errno and friends.\n");
-	printf(" */\n\n");
-	printf("#include <stdio.h>\n");
-	printf("#include <errno.h>\n");
-#elif defined(TRY_ERRNO_OLD_DECL)
-	extern int errno;		/* last system error */
-	extern char *sys_errlist[];	/* system error messages */
-	extern int sys_nerr;		/* number of system errors*/
-
-	printf("/*\n");
-	printf(" * The following comments were produced by calc_errno\n");
-	printf(" * in an effort to see if these values were correctly\n");
-	printf(" * declared when calc_errno.c was compiled.\n");
-	printf(" */\n\n");
-	printf("/* Initially errno is %d */\n", errno);
-	printf("/* There are %d entries in sys_errlist[] */\n", sys_nerr);
-	printf("/* The 2nd sys_errlist entry is \"%s\" */\n\n", sys_errlist[1]);
-	printf("/*\n");
-	printf(" * Based on the above, calc_errno now knows\n");
-	printf(" * how to declare errno and friends.\n");
-	printf(" */\n\n");
-	printf("extern int errno;\t\t/* last system error */\n");
-	printf("extern char *sys_errlist[];\t"
-	       "/* system error messages */\n");
-	printf("extern int sys_nerr;\t\t/* number of system errors*/\n");
-#else	/* assume defined(TRY_ERRNO_STD_DECL) */
-	extern int errno;			/* last system error */
-	extern const char *const sys_errlist[];	/* system error messages */
-	extern int sys_nerr;			/* number of system errors*/
-
-	printf("/*\n");
-	printf(" * The following comments were produced by calc_errno\n");
-	printf(" * in an effort to see if these values were correctly\n");
-	printf(" * declared when calc_errno.c was compiled.\n");
-	printf(" */\n\n");
-	printf("/* Initially errno is %d */\n", errno);
-	printf("/* There are %d entries in sys_errlist[] */\n", sys_nerr);
-	printf("/* The 2nd sys_errlist entry is \"%s\" */\n\n", sys_errlist[1]);
-	printf("/*\n");
-	printf(" * Based on the above, calc_errno now knows\n");
-	printf(" * how to declare errno and friends.\n");
-	printf(" */\n\n");
-	printf("extern int errno;\t\t\t/* last system error */\n");
-	printf("extern const char *const sys_errlist[];\t"
-	       "/* system error messages */\n");
-	printf("extern int sys_nerr;\t\t\t/* number of system errors*/\n");
-#endif
-
-	/* exit(0); */
-	return 0;
-}
--- a/codegen.c
+++ b/codegen.c
@@ -263,7 +263,7 @@ ungetfunction(void)
 				return;
 			default:
 				scanerror(T_SEMICOLON,
-					 "Non-name arg for undefine"); 
+					 "Non-name arg for undefine");
 				return;
 		}
 	}
@@ -521,11 +521,9 @@ getonevariable(int symtype)
 *	| BREAK ';'
 *	| RETURN assignment ';'
 *	| GOTO label ';'
- *	| MAT name '[' value [ ':' value ] [',' value [ ':' value ] ] ']' ';'
- *	| OBJ type '{' arg [ ',' arg ] ... '}' ] ';'
- *	| OBJ type name [ ',' name ] ';'
 *	| PRINT assignment [, assignment ] ... ';'
 *	| QUIT [ string ] ';'
+ *	| ABORT [ string ] ';'
 *	| SHOW item ';'
 *	| body
 *	| assignment ';'
@@ -707,14 +705,14 @@ getstatement(LABEL *contlabel, LABEL *breaklabel, LABEL *nextcaselabel, LABEL *d
 			}
 		}
 		if (gettoken() != T_RIGHTPAREN) {	/* have 'c' part */
-			if (label1.l_offset <= 0)
+			if (label1.l_offset < 0)
 				addoplabel(OP_JUMP, &label3);
 			setlabel(&label2);
 			contlabel = &label2;
 			rescantoken();
 			(void) getexprlist();
 			addop(OP_POP);
-			if (label1.l_offset > 0)
+			if (label1.l_offset >= 0)
 				addoplabel(OP_JUMP, &label1);
 			if (gettoken() != T_RIGHTPAREN) {
 				(void) tokenmode(oldmode);
@@ -734,15 +732,21 @@ getstatement(LABEL *contlabel, LABEL *breaklabel, LABEL *nextcaselabel, LABEL *d
 	case T_WHILE:
 		oldmode = tokenmode(TM_DEFAULT);
 		contlabel = &label1;
-		breaklabel = &label2;
 		clearlabel(contlabel);
-		clearlabel(breaklabel);
 		setlabel(contlabel);
 		getcondition();
-		addoplabel(OP_JUMPZ, breaklabel);
-		getstatement(contlabel, breaklabel, NULL_LABEL, NULL_LABEL);
-		addoplabel(OP_JUMP, contlabel);
-		setlabel(breaklabel);
+		if (gettoken() != T_SEMICOLON) {
+			breaklabel = &label2;
+			clearlabel(breaklabel);
+			addoplabel(OP_JUMPZ, breaklabel);
+			rescantoken();
+			getstatement(contlabel, breaklabel,
+				NULL_LABEL, NULL_LABEL);
+			addoplabel(OP_JUMP, contlabel);
+			setlabel(breaklabel);
+		} else {
+			addoplabel(OP_JUMPNZ, contlabel);
+		}
 		(void) tokenmode(oldmode);
 		return;

@@ -1053,18 +1057,14 @@ getoneobj(long index, int symtype)
 }

 /*
- * Routine to collect a set of variables for the specified object type
- * and initialize them as being that type of object.
- * Here
- *	objlist = name initlist [ ',' name initlist ] ... ';'.
- * If symtype is SYM_UNDEFINED, then this is an OBJ statement where the
- * values can be any variable expression, and no symbols are to be defined.
- * Otherwise this is part of a declaration, and the variables must be raw
- * symbol names which are defined with the specified symbol type.
+ * Routine to assign a specified object-type value to each of a set of
+ * variables in a "global", "local" or "static" declaration, or, if
+ * symtype is SYM_UNDEFINED, to create one object value of the specified
+ * type.
 *
 * given:
 *	name		object name
- *	symtype		type of symbol to collect for
+ *	symtype		declaration type
 */
 static void
 getobjvars(char *name, int symtype)
@@ -1078,6 +1078,8 @@ getobjvars(char *name, int symtype)
 	}
 	for (;;) {
 		getoneobj(index, symtype);
+		if (symtype == SYM_UNDEFINED)
+			return;
 		if (gettoken() != T_COMMA) {
 			rescantoken();
 			return;
@@ -1921,11 +1923,6 @@ getterm(void)

 		case T_MAT:
 			getonematrix(SYM_UNDEFINED);
-			while (gettoken() == T_COMMA) {
-				addop(OP_POP);
-				getonematrix(SYM_UNDEFINED);
-			}
-			rescantoken();
 			type = EXPR_ASSIGN;
 			break;

--- a/config.c
+++ b/config.c
@@ -1000,6 +1000,7 @@ config_value(CONFIG *cfg, int type, VALUE *vp)
 	 * convert element to value
 	 */
 	vp->v_type = V_NUM;
+	vp->v_subtype = V_NOSUBTYPE;
 	switch (type) {
 	case CONFIG_ALL:
 		vp->v_type = V_CONFIG;
--- a/custom/Makefile
+++ b/custom/Makefile
@@ -310,6 +310,15 @@ libcustcalc.a: ${CUSTCALC_OBJ} ${MAKE_FILE} ../Makefile
 	ar qc libcustcalc.a ${CUSTCALC_OBJ}
 	${RANLIB} libcustcalc.a

+##
+#
+# Special .o files
+#
+##
+
+c_sysinfo.o: c_sysinfo.c ${MAKE_FILE}
+	${CC} ${CFLAGS} c_sysinfo.c -c
+
 ##
 #
 # used by the upper level Makefile
--- a/custom/c_sysinfo.c
+++ b/custom/c_sysinfo.c
@@ -33,6 +33,7 @@
 #include "../config.h"
 #include "../calc.h"
 #include "../longbits.h"
+#define CHECK_L_FORMAT
 #include "../longlong.h"
 #include "../block.h"
 #include "../calcerr.h"
@@ -306,17 +307,33 @@ static void
 dump_name_value(void)
 {
 	struct infoname *p;	/* current infoname */
+	char *fmt;		/* printf value format */

 	/* dump the entire table */
 	for (p = sys_info; p->name != NULL; ++p) {
 		if (p->str == NULL) {
 #if LONG_BITS == FULL_BITS || FULL_BITS == 32 || !defined(HAVE_LONGLONG)
-			printf("%s%-23s\t%-8lu\t(0x%lx)\n",
+			fmt = "%s%-23s\t%-8lu\t(0x%lx)\n";
+			printf(fmt,
 			    (conf->tab_ok ? "\t" : ""), p->name,
 			    (unsigned long)p->nmbr,
 			    (unsigned long)p->nmbr);
 #else
-			printf("%s%-23s\t%-8llu\t(0x%llx)\n",
+			/*
+			 * Determine of %ld can print a 64 bit long long.
+			 *
+			 * Some systems that can make use of %ld to print a
+			 * a 64 bit value do not support the %lld type.
+			 * So we will only try %lld if %ld does not work.
+			 */
+			if (l_format < 0) {
+				/* %ld prints lower 32 bits only, use %lld */
+				fmt = "%s%-23s\t%-8llu\t(0x%llx)\n";
+			} else {
+				/* %ld prints all 64 bits, use %ld */
+				fmt = "%s%-23s\t%-8lu\t(0x%lx)\n";
+			}
+			printf(fmt,
 			    (conf->tab_ok ? "\t" : ""), p->name,
 			    (unsigned long long)p->nmbr,
 			    (unsigned long long)p->nmbr);
@@ -337,17 +354,33 @@ static void
 dump_mening_value(void)
 {
 	struct infoname *p;	/* current infoname */
+	char *fmt;		/* printf value format */

 	/* dump the entire table */
 	for (p = sys_info; p->name != NULL; ++p) {
 		if (p->str == NULL) {
 #if LONG_BITS == FULL_BITS || FULL_BITS == 32 || !defined(HAVE_LONGLONG)
-			printf("%s%-36.36s\t%-8lu\t(0x%lx)\n",
+			fmt = "%s%-36.36s\t%-8lu\t(0x%lx)\n";
+			printf(fmt,
 			    (conf->tab_ok ? "\t" : ""), p->meaning,
 			    (unsigned long)p->nmbr,
 			    (unsigned long)p->nmbr);
 #else
-			printf("%s%-36.36s\t%-8llu\t(0x%llx)\n",
+			/*
+			 * Determine of %ld can print a 64 bit long long.
+			 *
+			 * Some systems that can make use of %ld to print a
+			 * a 64 bit value do not support the %lld type.
+			 * So we will only try %lld if %ld does not work.
+			 */
+			if (l_format < 0) {
+				/* %ld prints lower 32 bits only, use %lld */
+				fmt = "%s%-36.36s\t%-8llu\t(0x%llx)\n";
+			} else {
+				/* %ld prints all 64 bits, use %ld */
+				fmt = "%s%-36.36s\t%-8lu\t(0x%lx)\n";
+			}
+			printf(fmt,
 			    (conf->tab_ok ? "\t" : ""), p->meaning,
 			    (unsigned long long)p->nmbr,
 			    (unsigned long long)p->nmbr);
--- a/file.c
+++ b/file.c
@@ -2228,15 +2228,16 @@ freadnum(FILE *fp, VALUE *valptr)
 			ch = fgetc(fp);
 		}
 	}
-	if (ch == 'i' || ch == 'I')
+	if (ch == 'i' || ch == 'I') {
 		imag = TRUE;
-	else {
+	} else {
 		ungetc(ch, fp);
 	}

 	if (ziszero(num)) {
 		zfree(num);
 		val.v_type = V_NUM;
+		val.v_subtype = V_NOSUBTYPE;
 		val.v_num = qlink(&_qzero_);
 		*valptr = val;
 		return;
@@ -2281,11 +2282,11 @@ freadnum(FILE *fp, VALUE *valptr)
 		c->imag = q;
 		val.v_type = V_COM;
 		val.v_com = c;
-	}
-	else {
+	} else {
 		val.v_type = V_NUM;
 		val.v_num = q;
 	}
+	val.v_subtype = V_NOSUBTYPE;
 	*valptr = val;
 }

--- a/func.c
+++ b/func.c
--- a/have_strdup.c
+++ b/have_strdup.c
@@ -0,0 +1,62 @@
+/*
+ * have_strdup - Determine if we strdup()
+ *
+ * usage:
+ *	have_strdup
+ *
+ * Not all systems have the strdup() function, so this may not
+ * compile on your system.
+ *
+ * This prog outputs several defines:
+ *
+ *	HAVE_STRDUP
+ *		defined ==> use strdup()
+ *		undefined ==> do not call or cannot call strdup()
+ */
+/*
+ * Copyright (c) 1999 by Landon Curt Noll.  All Rights Reserved.
+ *
+ * Permission to use, copy, modify, and distribute this software and
+ * its documentation for any purpose and without fee is hereby granted,
+ * provided that the above copyright, this permission notice and text
+ * this comment, and the disclaimer below appear in all of the following:
+ *
+ *	supporting documentation
+ *	source copies
+ *	source works derived from this source
+ *	binaries derived from this source or from derived source
+ *
+ * LANDON CURT NOLL DISCLAIMS ALL WARRANTIES WITH REGARD TO THIS SOFTWARE,
+ * INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS. IN NO
+ * EVENT SHALL LANDON CURT NOLL BE LIABLE FOR ANY SPECIAL, INDIRECT OR
+ * CONSEQUENTIAL DAMAGES OR ANY DAMAGES WHATSOEVER RESULTING FROM LOSS OF
+ * USE, DATA OR PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR
+ * OTHER TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR
+ * PERFORMANCE OF THIS SOFTWARE.
+ *
+ * chongo was here	/\../\
+ */
+
+#include <string.h>
+
+int
+main(void)
+{
+#if defined(HAVE_NO_STRDUP)
+
+	printf("#undef HAVE_STRDUP /* no */\n");
+
+#else /* HAVE_NO_STRDUP */
+
+	char *p;
+
+	p = strdup("#define HAVE_STRDUP /* yes */");
+	if (p != NULL) {
+		printf("%s\n", p);
+	}
+
+#endif /* HAVE_NO_STRDUP */
+
+	/* exit(0); */
+	return 0;
+}
--- a/help/contrib
+++ b/help/contrib
@@ -36,17 +36,33 @@ gziped (or compressed) tar file).

 You should send submissions to:

-	calc-tester@postofc.corp.sgi.com
+    calc-tester at postofc dot corp dot sgi dot com
+
+    [[ Replace 'at' with @, 'dot' is with . and remove the spaces ]]

 Thanks for considering submitting code to calc.  Calc is a collective
 work by a number of people.  It would not be what it is today without
 your efforts and submissions!

+Calc bug reports, however, should be sent to:
+
+    calc-bugs at postofc dot corp dot sgi dot com
+
+    [[ Replace 'at' with @, 'dot' is with . and remove the spaces ]]
+
+See the BUGS file or try the help command:
+
+    help bugs
+
+for details on bug reporting.
+
 =-=

 One may join the calc testing group by sending a request to:

-	calc-tester-request@postofc.corp.sgi.com
+	calc-tester-request at postofc dot corp dot sgi dot com
+
+	[[ Replace 'at' with @, 'dot' is with . and remove the spaces ]]

    Your message body (not the subject) should consist of:

--- a/help/credit
+++ b/help/credit
@@ -2,7 +2,11 @@ Credits

    The majority of calc was written by David I. Bell.

-    Calc archives and calc-tester mailing list maintained by Landon Curt Noll.
+    The Calc primary mirror, calc mailing list and calc bug report
+    processing is performed by Landon Curt Noll.
+
+    Landon Curt Noll maintains the master reference source, performs
+    release control functions as well as other calc maintenance functions.

    Thanks for suggestions and encouragement from Peter Miller,
    Neil Justusson, and Landon Noll.
--- a/help/funclist.sed
+++ b/help/funclist.sed
@@ -3,6 +3,5 @@ s/NUMBER[	 ]*\*/int /
 s/NUMBER/int/
 s/STRINGHEAD/int/
 s/\(".*",.*,.*\),.*,.*,.*,.*,/\1, 0, 0, 0, 0,/
-s/showbuiltins/main/
 s/[	 ][	 ]*$//
 p
--- a/help/todo
+++ b/help/todo
@@ -5,10 +5,22 @@ Calc Todo Items:

    Code contributions are welcome.  Send patches to:

-	    calc-tester@postofc.corp.sgi.com
+	    calc-tester at postofc dot corp dot sgi dot com
+
+	    [[ Replace 'at' with @, 'dot' is with . and remove the spaces ]]

    See also the 'wishlist' help files for the calc enhancement wish list.

+    Calc bug reports, however, should send to:
+
+	    calc-tester at postofc dot corp dot sgi dot com
+
+	    [[ Replace 'at' with @, 'dot' is with . and remove the spaces ]]
+
+    see the BUGS file or try the calc command:
+
+	    help bugs
+
 =-=

 Very High priority items:
@@ -50,6 +62,8 @@ Very High priority items:
    * Fix any 'Known bugs' as noted in the BUGS file or as
      displayed by 'calc help bugs'.

+    * Copyleft calc and place it under a LGPL.
+
 =-=

 High priority items:
@@ -79,10 +93,9 @@ High priority items:
      then a $PAGER should be forked and execed after doing things such
      as closing open file descriptors that the $PAGER does not need.

-    * Inspect changes made since the last major calc release (the
-      initial x.y.z version ... prior to any t sub-version) to help
-      ensure that they have not introduced new or re-introduced old bugs
-      into calc.
+    * The arg passdown to the lower level Makfiles should be reviewed
+      to determine which items really need to be passed down.  The lower
+      level Makefiles should default to the shipped values.

    * Consider using configure to build the calc Makefile.

--- a/help/wishlist
+++ b/help/wishlist
@@ -4,11 +4,19 @@ Calc Enhancement Wish List:
    interesting calc scripts that you would like you see included in
    future distributions to:

-		calc-tester@postofc.corp.sgi.com
+	    calc-tester at postofc dot corp dot sgi dot com
+
+	    [[ Replace 'at' with @, 'dot' is with . and remove the spaces ]]

    The following items are in the calc wish list.  Programs like this
    can be extended and improved forever.

+    Calc bug repoers, however, should be sent to:
+
+	    calc-bugs at postofc dot corp dot sgi dot com
+
+	    [[ Replace 'at' with @, 'dot' is with . and remove the spaces ]]
+
    See the 'todo' help file for higher priority todo items.

 =-=
--- a/hist.c
+++ b/hist.c
@@ -31,6 +31,11 @@
 #include "terminal.h"
 #include "have_string.h"

+#include "have_strdup.h"
+#if !defined(HAVE_STRDUP)
+# define strdup(x) calc_strdup((CONST char *)(x))
+#endif /* HAVE_STRDUP */
+

 #if defined(USE_TERMIOS)
 # include <termios.h>
--- a/lib/README
+++ b/lib/README
@@ -33,7 +33,9 @@ the calc language, and/or because the authors thought them to be useful!

 If you write something that you think is useful, please send it to:

-    calc-tester@postofc.corp.sgi.com
+    calc-tester at postofc dot corp dot sgi dot com
+
+    [[ Replace 'at' with @, 'dot' is with . and remove the spaces ]]

 By convention, a lib file only defines and/or initializes functions,
 objects and variables.  (The regress.cal and testxxx.cal regression test
--- a/lib/regress.cal
+++ b/lib/regress.cal
@@ -1694,6 +1694,7 @@ define test_obj()
 {
 	static obj surd a;
 	static obj surd b;
+	local PP;

 	print '1800: Beginning object test';

@@ -1731,7 +1732,10 @@ define test_obj()
 	vrfy(a <= a,			'1828: a < a');
 	vrfy(isobj(a) == 1,		'1829: isobj(a) == 1');

-	print '1830: Ending object test';
+	obj pt {x,y}, PP = obj pair {A,B} = {obj pt, obj pt};
+	print '1830: obj pt {x,y}, PP = obj pair {A,B} = {obj pt, obj pt}';
+
+	print '1831: Ending object test';
 }
 print '028: parsed test_obj()';

@@ -2867,7 +2871,7 @@ print '053: parsed test_frem()';
 */
 define test_error()
 {
-	local strx, e99, list1, e999;
+	local strx, e99, list1, e9999;
 	local a, b, c, n, x;		/* used by newerror() */

 	print '3600: Beginning test_error';
@@ -3040,11 +3044,12 @@ define test_error()
 					'3710: newerror() == newerror("???")');
 	vrfy(newerror("") == newerror(),
 					'3711: newerror("") == newerror()');
-	e999 = error(999);
-	print				'3712: e999 = error(999)';
-	vrfy(errno() == 999,		'3713: errno() == 999');
-	vrfy(error() == e999,		'3714: error() == e999');
-	vrfy(strerror() == "Error 999",	'3715: strerror() == "Error 999"');
+	e9999 = error(9999);
+	print				'3712: e9999 = error(9999)';
+	vrfy(errno() == 9999,		'3713: errno() == 9999');
+	vrfy(error() == e9999,		'3714: error() == e9999');
+	vrfy(strerror() == "Unknown error 9999",
+				'3715: strerror() == "Unknown error 9999"');
 	x = newerror("Alpha");
 	print				'3716: x = newerror("Alpha")';
 	n = iserror(x);
@@ -3052,10 +3057,11 @@ define test_error()
 	vrfy(errno() == n,		'3718: errno() == n');
 	vrfy(error() == x,		'3719: error() == x');
 	vrfy(strerror() == "Alpha",	'3720: strerror() == "Alpha"');
-	vrfy(errno(999) == n,		'3721: errno() == n');
-	vrfy(errno() == 999,		'3722: errno() == 999');
-	vrfy(error() == e999,		'3723: error() == e999');
-	vrfy(strerror() == "Error 999",	'3724: strerror() == "Error 999"');
+	vrfy(errno(9999) == n,		'3721: errno() == n');
+	vrfy(errno() == 9999,		'3722: errno() == 9999');
+	vrfy(error() == e9999,		'3723: error() == e9999');
+	vrfy(strerror() == "Unknown error 9999",
+				'3724: strerror() == "Unknown error 9999"');
 	a = 1/0;
 	print				'3725: a = 1/0';
 	vrfy(strerror() == "Division by zero",
@@ -4687,8 +4693,8 @@ define test_commaeq()
 	/*
 	 * matrix assignment
 	 */
-	mat A[3] = {1,2,3}, B[2] = {1,2};
-	print		'5625: mat A[3] = {1,2,3}, B[2] = {1,2}';
+	mat A[3] = {1,2,3}, mat B[2] = {1,2};
+	print		'5625: mat A[3] = {1,2,3}, mat B[2] = {1,2}';
 	vrfy(A[0] == 1,			  '5626: A[0] == 1');
 	vrfy(B[1] == 2,			  '5627: B[1] == 2');

@@ -4916,8 +4922,8 @@ define test_assign(base, work)
 	print				  base+3: ': X5800 = base+3';
 	Y5800 = base+4;
 	print				  base+4: ': Y5800 = base+4';
-	obj xy5800 A={1,2}, B={3,4};
-	print base+5: ': obj xy5800 A={1,2}, B={3,4}';
+	obj xy5800 A={1,2}, obj xy5800 B={3,4};
+	print base+5: ': obj xy5800 A={1,2}, obj xy5000 B={3,4}';

 	/*
 	 * test assignment
@@ -7372,7 +7378,7 @@ vrfy(X5800 == (obj xy5800 = {3,9}),
 				  '5859: X5800 == (obj xy5800 = {3,9})');
 vrfy(errno() > 0,		  '5860: errno() > 0');
 vrfy(strerror() == "Incompatible types for =",
-			'5861: strerror() == "Incompatible types for ="');
+		'5861: strerror() == "Incompatible types for ="');
 X5800 = 2;
 print				  '5862: X5800 = 2';
 vrfy(X5800 == (obj xy5800 = {2,0}),
@@ -7381,7 +7387,10 @@ X5800 = obj xy5800 = {1,2};
 print				  '5864: X5800 = obj xy5800 = {1,2}';
 vrfy(X5800 == (obj xy5800 = {1,2}),
 				  '5865: X5800 == (obj xy5800 = {1,2})');
-print '5866: End of 5800 sequence';
+define f5800(a8500 = mat[2] = {3,4}) = 5 * a8500;
+print		'5866: define f5800(a8500 = mat[2] = {3,4}) = 5 * a8500;'
+vrfy(f5800() == (mat[] = {15,20}),'5867: f5800() == (mat[] = {15,20})');
+print '5868: End of 5800 sequence';

 print;
 return test_is();
--- a/lib_calc.c
+++ b/lib_calc.c
@@ -34,6 +34,11 @@
 #include "symbol.h"
 #include "func.h"

+#include "have_strdup.h"
+#if !defined(HAVE_STRDUP)
+# define strdup(x) calc_strdup((CONST char *)(x))
+#endif /* HAVE_STRDUP */
+
 #include "have_unistd.h"
 #if defined(HAVE_UNISTD_H)
 #include <unistd.h>
@@ -313,17 +318,16 @@ initenv(void)
 	char *c;

 	/* determine the $CALCPATH value */
-	c = getenv(CALCPATH);
-	calcpath = ((no_env || c == NULL) ? NULL : strdup(c));
+	c = (no_env ? NULL : getenv(CALCPATH));
+	calcpath = (c ? strdup(c) : NULL);
 	if (calcpath == NULL)
 		calcpath = DEFAULTCALCPATH;

 	/* determine the $CALCRC value */
-	c = getenv(CALCRC);
-	calcrc = ((no_env || c == NULL) ? NULL : strdup(c));
-	if (calcrc == NULL) {
+	c = (no_env ? NULL : getenv(CALCRC));
+	calcrc = (c ? strdup(c) : NULL);
+	if (calcrc == NULL)
 		calcrc = DEFAULTCALCRC;
-	}
 	if (strlen(calcrc) > MAX_CALCRC) {
 		math_error("The $CALCRC variable is longer than %d chars",
 			   MAX_CALCRC);
@@ -331,15 +335,14 @@ initenv(void)
 	}

 	/* determine the $CALCBINDINGS value */
-	c = getenv(CALCBINDINGS);
-	calcbindings = ((no_env || c == NULL) ? NULL : strdup(c));
-	if (calcbindings == NULL) {
+	c = (no_env ? NULL : getenv(CALCBINDINGS));
+	calcbindings = (c ? strdup(c) : NULL);
+	if (calcbindings == NULL)
 		calcbindings = DEFAULTCALCBINDINGS;
-	}

 	/* determine the $HOME value */
-	c = getenv(HOME);
-	home = ((no_env || c == NULL) ? NULL : strdup(c)); 
+	c = (no_env ? NULL : getenv(HOME));
+	home = (c ? strdup(c) : NULL);
 	if (home == NULL || home[0] == '\0') {
 		ent = (struct passwd *)getpwuid(geteuid());
 		if (ent == NULL) {
@@ -351,16 +354,15 @@ initenv(void)
 	}

 	/* determine the $PAGER value */
-	c = getenv(PAGER);
-	pager = ((no_env || c == NULL) ? NULL : strdup(c));
-	if (pager == NULL || *pager == '\0') {
+	c = (no_env ? NULL : getenv(PAGER));
+	pager = (c ? strdup(c) : NULL);
+	if (pager == NULL || *pager == '\0')
 		pager = DEFAULTCALCPAGER;
-	}

 	/* determine the $SHELL value */
-	c = getenv(SHELL);
-	shell = ((no_env || c == NULL) ? NULL : strdup(c));
-	if (shell == NULL)
+	c = (no_env ? NULL : getenv(SHELL));
+	shell = (c ? strdup(c) : NULL);
+	if (shell == NULL || *shell == '\0')
 		shell = DEFAULTSHELL;
 }

@@ -431,3 +433,45 @@ run_state_name(run state)
 	}
 	return "RUN_invalid";
 }
+
+
+/*
+ * calc_strdup - calc interface to provide or simulate strdup()
+ */
+char *
+calc_strdup(CONST char *s1)
+{
+#if defined(HAVE_STRDUP)
+
+	return strdup(s1);
+
+#else /* HAVE_STRDUP */
+
+	char *ret;	/* return string */
+
+	/*
+	 * firewall
+	 */
+	if (s1 == NULL) {
+		return NULL;
+	}
+
+	/*
+	 * allocate duplicate storage
+	 */
+	ret = (char *)malloc(sizeof(char) * (strlen(s1)+1));
+
+	/*
+	 * if we have storage, duplicate the string
+	 */
+	if (ret != NULL) {
+		strcpy(ret, s1);
+	}
+
+	/*
+	 * return the new string, or NULL if malloc failed
+	 */
+	return ret;
+
+#endif /* HAVE_STRDUP */
+}
--- a/listfunc.c
+++ b/listfunc.c
@@ -130,10 +130,12 @@ removelistfirst(LIST *lp, VALUE *vp)
 {
 	if (lp->l_count == 0) {
 		vp->v_type = V_NULL;
+		vp->v_subtype = V_NOSUBTYPE;
 		return;
 	}
 	*vp = lp->l_first->e_value;
 	lp->l_first->e_value.v_type = V_NULL;
+	lp->l_first->e_value.v_type = V_NOSUBTYPE;
 	removelistelement(lp, lp->l_first);
 }

@@ -151,10 +153,12 @@ removelistlast(LIST *lp, VALUE *vp)
 {
 	if (lp->l_count == 0) {
 		vp->v_type = V_NULL;
+		vp->v_subtype = V_NOSUBTYPE;
 		return;
 	}
 	*vp = lp->l_last->e_value;
 	lp->l_last->e_value.v_type = V_NULL;
+	lp->l_last->e_value.v_subtype = V_NOSUBTYPE;
 	removelistelement(lp, lp->l_last);
 }

@@ -181,6 +185,7 @@ removelistmiddle(LIST *lp, long index, VALUE *vp)
 	}
 	*vp = ep->e_value;
 	ep->e_value.v_type = V_NULL;
+	ep->e_value.v_subtype = V_NOSUBTYPE;
 	removelistelement(lp, ep);
 }

--- a/longlong.c
+++ b/longlong.c
@@ -62,8 +62,8 @@
 /*
 * have the compiler try its hand with unsigned and signed long longs
 */
-unsigned long long val = 4294967297ULL;
-long long val2 = -4294967297LL;
+unsigned long long val = 0x1234567890123456ULL;
+long long val2 = -1311768467284833366LL;	/* -0x1234567890123456 */


 int
@@ -90,14 +90,20 @@ main(int argc, char **argv)
 	if (longlong_bits > 0) {

 		/*
-		 * if size is longer than an unsigned long, use it
+		 * if size is longer than an unsigned long,
+		 * and the negative 'long long' works, then use long long's
 		 */
-		if (longlong_bits > sizeof(unsigned long)*8) {
+		if (longlong_bits > sizeof(unsigned long)*8 && val2 < 0) {

 			/* use long long length */
 			printf("#define HAVE_LONGLONG\n");
 			printf("#define LONGLONG_BITS %d  /* yes */\n",
 			    longlong_bits);
+
+			printf("\n/* does %%lld work or does %%ld? */\n");
+			printf("#if defined(CHECK_L_FORMAT)\n");
+			printf("long long l_format = %ld;\n", val);
+			printf("#endif /* CHECK_L_FORMAT */\n");
 		}
 	}
 	/* exit(0); */
--- a/matfunc.c
+++ b/matfunc.c
@@ -154,6 +154,7 @@ matmul(MATRIX *m1, MATRIX *m2)
 	for (i1 = 0; i1 < max1; i1++) {
 		for (i2 = 0; i2 < max2; i2++) {
 			sum.v_type = V_NULL;
+			sum.v_subtype = V_NOSUBTYPE;
 			v1 = &m1->m_table[i1 * maxindex];
 			v2 = &m2->m_table[i2];
 			for (index = 0; index < maxindex; index++) {
@@ -202,6 +203,7 @@ matsquare(MATRIX *m)
 	for (i1 = 0; i1 < max; i1++) {
 		for (i2 = 0; i2 < max; i2++) {
 			sum.v_type = V_NULL;
+			sum.v_subtype = V_NOSUBTYPE;
 			v1 = &m->m_table[i1 * max];
 			v2 = &m->m_table[i2];
 			for (index = 0; index < max; index++) {
@@ -405,6 +407,7 @@ matscale(MATRIX *m, long n)
 	if (n == 0)
 		return matcopy(m);
 	temp.v_type = V_NUM;
+	temp.v_subtype = V_NOSUBTYPE;
 	temp.v_num = itoq(n);
 	res = matalloc(m->m_size);
 	*res = *m;
@@ -436,6 +439,7 @@ matshift(MATRIX *m, long n)
 	if (n == 0)
 		return matcopy(m);
 	temp.v_type = V_NUM;
+	temp.v_subtype = V_NOSUBTYPE;
 	temp.v_num = itoq(n);
 	res = matalloc(m->m_size);
 	*res = *m;
@@ -1062,6 +1066,7 @@ matdet(MATRIX *m)
 		while (!testvalue(val)) {
 			if (--i <= 0) {
 				tmp1.v_type = V_NUM;
+				tmp1.v_subtype = V_NOSUBTYPE;
 				tmp1.v_num = qlink(&_qzero_);
 				matfree(m);
 				return tmp1;
--- a/obj.c
+++ b/obj.c
@@ -141,6 +141,10 @@ objcall(int action, VALUE *v1, VALUE *v2, VALUE *v3)
 	VALUE tmp;		/* temp value */
 	char name[SYMBOLSIZE+1];	/* full name of user routine to call */

+	/* initialize VALUEs */
+	val.v_subtype = V_NOSUBTYPE;
+	tmp.v_subtype = V_NOSUBTYPE;
+
 	if ((unsigned)action > OBJ_MAXFUNC) {
 		math_error("Illegal action for object call");
 		/*NOTREACHED*/
@@ -404,6 +408,7 @@ objpowi(VALUE *vp, NUMBER *q)
 			case 1:
 				res.v_obj = objcopy(vp->v_obj);
 				res.v_type = V_OBJ;
+				res.v_subtype = V_NOSUBTYPE;
 				return res;
 			case -1:
 				return objcall(OBJ_INV, vp, NULL_VALUE, NULL_VALUE);
@@ -579,6 +584,16 @@ findelement(char *name)
 }


+/*
+ * Returns the name of object type with specified index
+ */
+char *
+objtypename(unsigned long index)
+{
+	return namestr(&objectnames, (long)index);
+}
+
+
 /*
 * Return the value table offset to be used for an object element name.
 * This converts the element index from the element table into an offset
--- a/opcodes.c
+++ b/opcodes.c
@@ -230,6 +230,7 @@ o_argvalue(FUNC *fp, int argcount, VALUE *args)
 			qfree(stack->v_num);
 		stack->v_num = itoq((long) argcount);
 		stack->v_type = V_NUM;
+		stack->v_subtype = V_NOSUBTYPE;
 		return;
 	}
 	index = qtoi(vp->v_num) - 1;
@@ -292,6 +293,7 @@ o_string(FUNC *fp, long arg)
 	stack++;
 	stack->v_str = slink(findstring(arg));
 	stack->v_type = V_STR;
+	stack->v_subtype = V_NOSUBTYPE;
 }


@@ -300,6 +302,7 @@ o_undef(void)
 {
 	stack++;
 	stack->v_type = V_NULL;
+	stack->v_subtype = V_NOSUBTYPE;
 }


@@ -940,6 +943,7 @@ o_swap(void)
 		/*NOTREACHED*/
 	}
 	stack->v_type = V_NULL;
+	stack->v_subtype = V_NOSUBTYPE;
 }


@@ -958,11 +962,13 @@ o_add(void)
 		v2 = v2->v_addr;
 	if (v1->v_type == V_OCTET) {
 		w1.v_type = V_NUM;
+		w1.v_subtype = V_NOSUBTYPE;
 		w1.v_num = itoq(*v1->v_octet);
 		v1 = &w1;
 	}
 	if (v2->v_type == V_OCTET) {
 		w2.v_type = V_NUM;
+		w2.v_subtype = V_NOSUBTYPE;
 		w2.v_num = itoq(*v2->v_octet);
 		v2 = &w2;
 	}
@@ -993,11 +999,13 @@ o_sub(void)
 		v2 = v2->v_addr;
 	if (v1->v_type == V_OCTET) {
 		w1.v_type = V_NUM;
+		w1.v_subtype = V_NOSUBTYPE;
 		w1.v_num = itoq((unsigned char) *v1->v_octet);
 		v1 = &w1;
 	}
 	if (v2->v_type == V_OCTET) {
 		w2.v_type = V_NUM;
+		w2.v_subtype = V_NOSUBTYPE;
 		w2.v_num = itoq((unsigned char) *v2->v_octet);
 		v2 = &w2;
 	}
@@ -1028,11 +1036,13 @@ o_mul(void)
 		v2 = v2->v_addr;
 	if (v1->v_type == V_OCTET) {
 		w1.v_type = V_NUM;
+		w1.v_subtype = V_NOSUBTYPE;
 		w1.v_num = itoq(*v1->v_octet);
 		v1 = &w1;
 	}
 	if (v2->v_type == V_OCTET) {
 		w2.v_type = V_NUM;
+		w2.v_subtype = V_NOSUBTYPE;
 		w2.v_num = itoq(*v2->v_octet);
 		v2 = &w2;
 	}
@@ -1081,11 +1091,13 @@ o_div(void)
 		v2 = v2->v_addr;
 	if (v1->v_type == V_OCTET) {
 		w1.v_type = V_NUM;
+		w1.v_subtype = V_NOSUBTYPE;
 		w1.v_num = itoq(*v1->v_octet);
 		v1 = &w1;
 	}
 	if (v2->v_type == V_OCTET) {
 		w2.v_type = V_NUM;
+		w2.v_subtype = V_NOSUBTYPE;
 		w2.v_num = itoq(*v2->v_octet);
 		v2 = &w2;
 	}
@@ -1113,6 +1125,7 @@ o_quo(void)
 	if (v2->v_type == V_ADDR)
 		v2 = v2->v_addr;
 	null.v_type = V_NULL;
+	null.v_subtype = V_NOSUBTYPE;
 	quovalue(v1, v2, &null, &tmp);
 	freevalue(stack--);
 	freevalue(stack);
@@ -1133,6 +1146,7 @@ o_mod(void)
 	if (v2->v_type == V_ADDR)
 		v2 = v2->v_addr;
 	null.v_type = V_NULL;
+	null.v_subtype = V_NOSUBTYPE;
 	modvalue(v1, v2, &null, &tmp);
 	freevalue(stack--);
 	freevalue(stack);
@@ -1175,7 +1189,9 @@ o_quomod(void)
 	}

 	valquo.v_type = V_NUM;
+	valquo.v_subtype = V_NOSUBTYPE;
 	valmod.v_type = V_NUM;
+	valmod.v_subtype = V_NOSUBTYPE;
 	res = qquomod(v1->v_num, v2->v_num, &valquo.v_num, &valmod.v_num);
 	stack -= 2;
 	if (stack->v_type == V_NUM)
@@ -1185,6 +1201,7 @@ o_quomod(void)
 		qfree(stack->v_num);
 	stack->v_num = (res ? qlink(&_qone_) : qlink(&_qzero_));
 	stack->v_type = V_NUM;
+	stack->v_subtype = V_NOSUBTYPE;

 	freevalue(v3);
 	freevalue(v4);
@@ -1289,6 +1306,7 @@ o_not(void)
 	freevalue(stack);
 	stack->v_num = (r ? qlink(&_qzero_) : qlink(&_qone_));
 	stack->v_type = V_NUM;
+	stack->v_subtype = V_NOSUBTYPE;
 }


@@ -1335,6 +1353,7 @@ o_negate(void)
 			qfree(stack->v_num);
 		stack->v_num = q;
 		stack->v_type = V_NUM;
+		stack->v_subtype = V_NOSUBTYPE;
 		return;
 	}
 	negvalue(vp, &tmp);
@@ -1440,6 +1459,7 @@ o_abs(void)
 		qfree(stack->v_num);
 	stack->v_num = q;
 	stack->v_type = V_NUM;
+	stack->v_subtype = V_NOSUBTYPE;
 }


@@ -1459,6 +1479,7 @@ o_norm(void)
 			qfree(stack->v_num);
 		stack->v_num = q;
 		stack->v_type = V_NUM;
+		stack->v_subtype = V_NOSUBTYPE;
 		return;
 	}
 	normvalue(vp, &tmp);
@@ -1483,6 +1504,7 @@ o_square(void)
 			qfree(stack->v_num);
 		stack->v_num = q;
 		stack->v_type = V_NUM;
+		stack->v_subtype = V_NOSUBTYPE;
 		return;
 	}
 	squarevalue(vp, &tmp);
@@ -1503,6 +1525,7 @@ o_test(void)
 	i = testvalue(vp);
 	freevalue(stack);
 	stack->v_type = V_NUM;
+	stack->v_subtype = V_NOSUBTYPE;
 	stack->v_num = i ? qlink(&_qone_) : qlink(&_qzero_);
 }

@@ -1534,6 +1557,7 @@ o_links(void)
 	if (!haveaddress)
 		links--;
 	stack->v_type = V_NUM;
+	stack->v_subtype = V_NOSUBTYPE;
 	stack->v_num = itoq(links);
 }

@@ -1584,6 +1608,7 @@ o_bit (void)
 		stack->v_type = V_NUM;
 		stack->v_num = itoq(r);
 	}
+	stack->v_subtype = V_NOSUBTYPE;
 }

 static void
@@ -1628,6 +1653,7 @@ o_highbit (void)
 			return;
 		default:
 			stack->v_type = V_NUM;
+			stack->v_subtype = V_NOSUBTYPE;
 			stack->v_num = itoq(index);
 	}
 }
@@ -1679,6 +1705,7 @@ o_lowbit (void)
 			return;
 		default:
 			stack->v_type = V_NUM;
+			stack->v_subtype = V_NOSUBTYPE;
 			stack->v_num = itoq(index);
 	}
 }
@@ -1772,6 +1799,7 @@ o_istype(void)
 	freevalue(stack);
 	stack->v_num = itoq((long) r);
 	stack->v_type = V_NUM;
+	stack->v_subtype = V_NOSUBTYPE;
 }


@@ -1788,6 +1816,7 @@ o_isint(void)
 		freevalue(stack);
 		stack->v_num = qlink(&_qzero_);
 		stack->v_type = V_NUM;
+		stack->v_subtype = V_NOSUBTYPE;
 		return;
 	}
 	if (qisint(vp->v_num))
@@ -1798,6 +1827,7 @@ o_isint(void)
 		qfree(stack->v_num);
 	stack->v_num = q;
 	stack->v_type = V_NUM;
+	stack->v_subtype = V_NOSUBTYPE;
 }


@@ -1822,10 +1852,12 @@ o_isnum(void)
 			freevalue(stack);
 			stack->v_num = qlink(&_qzero_);
 			stack->v_type = V_NUM;
+			stack->v_subtype = V_NOSUBTYPE;
 			return;
 	}
 	stack->v_num = qlink(&_qone_);
 	stack->v_type = V_NUM;
+	stack->v_subtype = V_NOSUBTYPE;
 }


@@ -1841,10 +1873,12 @@ o_ismat(void)
 		freevalue(stack);
 		stack->v_num = qlink(&_qzero_);
 		stack->v_type = V_NUM;
+		stack->v_subtype = V_NOSUBTYPE;
 		return;
 	}
 	freevalue(stack);
 	stack->v_type = V_NUM;
+	stack->v_subtype = V_NOSUBTYPE;
 	stack->v_num = qlink(&_qone_);
 }

@@ -1862,6 +1896,7 @@ o_islist(void)
 	freevalue(stack);
 	stack->v_num = (r ? qlink(&_qone_) : qlink(&_qzero_));
 	stack->v_type = V_NUM;
+	stack->v_subtype = V_NOSUBTYPE;
 }


@@ -1878,6 +1913,7 @@ o_isobj(void)
 	freevalue(stack);
 	stack->v_num = (r ? qlink(&_qone_) : qlink(&_qzero_));
 	stack->v_type = V_NUM;
+	stack->v_subtype = V_NOSUBTYPE;
 }


@@ -1894,6 +1930,7 @@ o_isstr(void)
 	freevalue(stack);
 	stack->v_num = (r ? qlink(&_qone_) : qlink(&_qzero_));
 	stack->v_type = V_NUM;
+	stack->v_subtype = V_NOSUBTYPE;
 }


@@ -1910,6 +1947,7 @@ o_isfile(void)
 	freevalue(stack);
 	stack->v_num = (r ? qlink(&_qone_) : qlink(&_qzero_));
 	stack->v_type = V_NUM;
+	stack->v_subtype = V_NOSUBTYPE;
 }


@@ -1926,6 +1964,7 @@ o_isrand(void)
 	freevalue(stack);
 	stack->v_num = (r ? qlink(&_qone_) : qlink(&_qzero_));
 	stack->v_type = V_NUM;
+	stack->v_subtype = V_NOSUBTYPE;
 }


@@ -1942,6 +1981,7 @@ o_israndom(void)
 	freevalue(stack);
 	stack->v_num = (r ? qlink(&_qone_) : qlink(&_qzero_));
 	stack->v_type = V_NUM;
+	stack->v_subtype = V_NOSUBTYPE;
 }


@@ -1958,6 +1998,7 @@ o_isconfig(void)
 	freevalue(stack);
 	stack->v_num = (r ? qlink(&_qone_) : qlink(&_qzero_));
 	stack->v_type = V_NUM;
+	stack->v_subtype = V_NOSUBTYPE;
 }


@@ -1976,6 +2017,7 @@ o_ishash(void)
 	freevalue(stack);
 	stack->v_num = itoq((long) r);
 	stack->v_type = V_NUM;
+	stack->v_subtype = V_NOSUBTYPE;
 }


@@ -1992,6 +2034,7 @@ o_isassoc(void)
 	freevalue(stack);
 	stack->v_num = (r ? qlink(&_qone_) : qlink(&_qzero_));
 	stack->v_type = V_NUM;
+	stack->v_subtype = V_NOSUBTYPE;
 }


@@ -2012,6 +2055,7 @@ o_isblock(void)
 	freevalue(stack);
 	stack->v_num = itoq(r);
 	stack->v_type = V_NUM;
+	stack->v_subtype = V_NOSUBTYPE;
 }


@@ -2028,6 +2072,7 @@ o_isoctet(void)
 	freevalue(stack);
 	stack->v_num = itoq(r);
 	stack->v_type = V_NUM;
+	stack->v_subtype = V_NOSUBTYPE;
 }


@@ -2050,6 +2095,7 @@ o_isptr(void)
 	freevalue(stack);
 	stack->v_num = itoq(r);
 	stack->v_type = V_NUM;
+	stack->v_subtype = V_NOSUBTYPE;
 }


@@ -2079,6 +2125,7 @@ o_isdefined(void)
 	freevalue(stack);
 	stack->v_num = itoq(r);
 	stack->v_type = V_NUM;
+	stack->v_subtype = V_NOSUBTYPE;
 }


@@ -2099,6 +2146,7 @@ o_isobjtype(void)
 	freevalue(stack);
 	stack->v_num = itoq(index >= 0);
 	stack->v_type = V_NUM;
+	stack->v_subtype = V_NOSUBTYPE;
 }


@@ -2122,6 +2170,7 @@ o_issimple(void)
 	freevalue(stack);
 	stack->v_num = (r ? qlink(&_qone_) : qlink(&_qzero_));
 	stack->v_type = V_NUM;
+	stack->v_subtype = V_NOSUBTYPE;
 }


@@ -2138,11 +2187,13 @@ o_isodd(void)
 			qfree(stack->v_num);
 		stack->v_num = qlink(&_qone_);
 		stack->v_type = V_NUM;
+		stack->v_subtype = V_NOSUBTYPE;
 		return;
 	}
 	freevalue(stack);
 	stack->v_num = qlink(&_qzero_);
 	stack->v_type = V_NUM;
+	stack->v_subtype = V_NOSUBTYPE;
 }


@@ -2159,11 +2210,13 @@ o_iseven(void)
 			qfree(stack->v_num);
 		stack->v_num = qlink(&_qone_);
 		stack->v_type = V_NUM;
+		stack->v_subtype = V_NOSUBTYPE;
 		return;
 	}
 	freevalue(stack);
 	stack->v_num = qlink(&_qzero_);
 	stack->v_type = V_NUM;
+	stack->v_subtype = V_NOSUBTYPE;
 }


@@ -2180,11 +2233,13 @@ o_isreal(void)
 			qfree(stack->v_num);
 		stack->v_num = qlink(&_qone_);
 		stack->v_type = V_NUM;
+		stack->v_subtype = V_NOSUBTYPE;
 		return;
 	}
 	freevalue(stack);
 	stack->v_num = qlink(&_qzero_);
 	stack->v_type = V_NUM;
+	stack->v_subtype = V_NOSUBTYPE;
 }


@@ -2200,11 +2255,13 @@ o_isnull(void)
 		freevalue(stack);
 		stack->v_num = qlink(&_qzero_);
 		stack->v_type = V_NUM;
+		stack->v_subtype = V_NOSUBTYPE;
 		return;
 	}
 	freevalue(stack);
 	stack->v_num = qlink(&_qone_);
 	stack->v_type = V_NUM;
+	stack->v_subtype = V_NOSUBTYPE;
 }


@@ -2221,6 +2278,7 @@ o_re(void)
 		if (stack->v_type == V_ADDR) {
 			stack->v_num = qlink(vp->v_num);
 			stack->v_type = V_NUM;
+			stack->v_subtype = V_NOSUBTYPE;
 		}
 		return;
 	}
@@ -2233,6 +2291,7 @@ o_re(void)
 		comfree(stack->v_com);
 	stack->v_num = q;
 	stack->v_type = V_NUM;
+	stack->v_subtype = V_NOSUBTYPE;
 }


@@ -2250,6 +2309,7 @@ o_im(void)
 			qfree(stack->v_num);
 		stack->v_num = qlink(&_qzero_);
 		stack->v_type = V_NUM;
+		stack->v_subtype = V_NOSUBTYPE;
 		return;
 	}
 	if (vp->v_type != V_COM) {
@@ -2261,6 +2321,7 @@ o_im(void)
 		comfree(stack->v_com);
 	stack->v_num = q;
 	stack->v_type = V_NUM;
+	stack->v_subtype = V_NOSUBTYPE;
 }


@@ -2277,6 +2338,7 @@ o_conjugate(void)
 		if (stack->v_type == V_ADDR) {
 			stack->v_num = qlink(vp->v_num);
 			stack->v_type = V_NUM;
+			stack->v_subtype = V_NOSUBTYPE;
 		}
 		return;
 	}
@@ -2382,6 +2444,7 @@ o_sgn(void)
 			qfree(vp->v_num);
 		stack->v_num = q;
 		stack->v_type = V_NUM;
+		stack->v_subtype = V_NOSUBTYPE;
 		return;
 	}
 	sgnvalue(vp, &tmp);
@@ -2410,6 +2473,7 @@ o_numerator(void)
 		qfree(stack->v_num);
 	stack->v_num = q;
 	stack->v_type = V_NUM;
+	stack->v_subtype = V_NOSUBTYPE;
 }


@@ -2431,6 +2495,7 @@ o_denominator(void)
 		qfree(stack->v_num);
 	stack->v_num = q;
 	stack->v_type = V_NUM;
+	stack->v_subtype = V_NOSUBTYPE;
 }


@@ -2687,6 +2752,7 @@ o_eq(void)
 	freevalue(stack);
 	stack->v_num = itoq((long) (r == 0));
 	stack->v_type = V_NUM;
+	stack->v_subtype = V_NOSUBTYPE;
 }


@@ -2707,6 +2773,7 @@ o_ne(void)
 	freevalue(stack);
 	stack->v_num = itoq((long) (r != 0));
 	stack->v_type = V_NUM;
+	stack->v_subtype = V_NOSUBTYPE;
 }


@@ -2727,15 +2794,15 @@ o_le(void)
 	freevalue(stack);

 	stack->v_type = V_NUM;
+	stack->v_subtype = V_NOSUBTYPE;
 	if (tmp.v_type == V_NUM) {
 		stack->v_num =  !qispos(tmp.v_num) ? qlink(&_qone_):
 				qlink(&_qzero_);
-	}
-	else if (tmp.v_type == V_COM) {
+	} else if (tmp.v_type == V_COM) {
 		stack->v_num = qlink(&_qzero_);
-	}
-	else
+	} else {
 		stack->v_type = V_NULL;
+	}
 	freevalue(&tmp);
 }

@@ -2756,14 +2823,13 @@ o_ge(void)
 	freevalue(stack--);
 	freevalue(stack);
 	stack->v_type = V_NUM;
+	stack->v_subtype = V_NOSUBTYPE;
 	if (tmp.v_type == V_NUM) {
 		stack->v_num =  !qisneg(tmp.v_num) ? qlink(&_qone_):
 				qlink(&_qzero_);
-	}
-	else if (tmp.v_type == V_COM) {
+	} else if (tmp.v_type == V_COM) {
 		stack->v_num = qlink(&_qzero_);
-	}
-	else {
+	} else {
 		stack->v_type = V_NULL;
 	}
 	freevalue(&tmp);
@@ -2786,15 +2852,15 @@ o_lt(void)
 	freevalue(stack--);
 	freevalue(stack);
 	stack->v_type = V_NUM;
+	stack->v_subtype = V_NOSUBTYPE;
 	if (tmp.v_type == V_NUM) {
 		stack->v_num =  qisneg(tmp.v_num) ? qlink(&_qone_):
 				qlink(&_qzero_);
-	}
-	else if (tmp.v_type == V_COM) {
+	} else if (tmp.v_type == V_COM) {
 		stack->v_num = qlink(&_qzero_);
-	}
-	else
+	} else {
 		stack->v_type = V_NULL;
+	}
 	freevalue(&tmp);
 }

@@ -2815,15 +2881,15 @@ o_gt(void)
 	freevalue(stack--);
 	freevalue(stack);
 	stack->v_type = V_NUM;
+	stack->v_subtype = V_NOSUBTYPE;
 	if (tmp.v_type == V_NUM) {
 		stack->v_num = qispos(tmp.v_num) ? qlink(&_qone_):
 				qlink(&_qzero_);
-	}
-	else if (tmp.v_type == V_COM) {
+	} else if (tmp.v_type == V_COM) {
 		stack->v_num = qlink(&_qzero_);
-	}
-	else
+	} else {
 		stack->v_type = V_NULL;
+	}
 	freevalue(&tmp);
 }

@@ -2886,6 +2952,7 @@ o_postinc(void)
 	if (stack->v_type == V_OCTET) {
 		stack[1] = stack[0];
 		stack->v_type = V_NUM;
+		stack->v_subtype = V_NOSUBTYPE;
 		stack->v_num = itoq((long) stack->v_octet[0]);
 		stack++;
 		stack->v_octet[0]++;
@@ -2938,8 +3005,8 @@ o_postdec(void)
 	freevalue(vp);
 	*vp = tmp;
 	stack->v_type = V_ADDR;
-	stack->v_addr = vp;
 	stack->v_subtype = V_NOSUBTYPE;
+	stack->v_addr = vp;
 }


@@ -3067,6 +3134,7 @@ o_zero(void)
 {
 	stack++;
 	stack->v_type = V_NUM;
+	stack->v_subtype = V_NOSUBTYPE;
 	stack->v_num = qlink(&_qzero_);
 }

@@ -3076,6 +3144,7 @@ o_one(void)
 {
 	stack++;
 	stack->v_type = V_NUM;
+	stack->v_subtype = V_NOSUBTYPE;
 	stack->v_num = qlink(&_qone_);
 }

@@ -3162,6 +3231,7 @@ o_getepsilon(void)
 {
 	stack++;
 	stack->v_type = V_NUM;
+	stack->v_subtype = V_NOSUBTYPE;
 	stack->v_num = qlink(conf->epsilon);
 }

@@ -3185,6 +3255,7 @@ o_setepsilon(void)
 	if (stack->v_type == V_NUM)
 		qfree(newep);
 	stack->v_type = V_NUM;
+	stack->v_subtype = V_NOSUBTYPE;
 }


@@ -3254,6 +3325,7 @@ updateoldvalue(FUNC *fp)
 	freevalue(&oldvalue);
 	oldvalue = fp->f_savedvalue;
 	fp->f_savedvalue.v_type = V_NULL;
+	fp->f_savedvalue.v_subtype = V_NOSUBTYPE;
 }


@@ -3275,6 +3347,7 @@ error_value(int e)
 		/*NOTREACHED*/
 	}
 	res.v_type = (short) -e;
+	res.v_subtype = V_NOSUBTYPE;
 	return res;
 }

@@ -3773,11 +3846,14 @@ dumpop(unsigned long *pc)
 			return 3;
 		case OP_PRINT: case OP_JUMPZ: case OP_JUMPNZ: case OP_JUMP:
 		case OP_CONDORJUMP: case OP_CONDANDJUMP: case OP_CASEJUMP:
-		case OP_INITSTATIC: case OP_MATCREATE: case OP_OBJCREATE:
+		case OP_INITSTATIC: case OP_MATCREATE:
 		case OP_SHOW: case OP_ELEMINIT: case OP_ELEMADDR:
 		case OP_ELEMVALUE: case OP_JUMPNN:
 			printf(" %ld\n", *pc);
 			return 2;
+		case OP_OBJCREATE:
+			printf(" %s\n", objtypename(*pc));
+			return 2;
 		case OP_NUMBER: case OP_IMAGINARY:
 			qprintf(" %r", constvalue(*pc));
 			printf("\n");
--- a/qmath.c
+++ b/qmath.c
@@ -1382,10 +1382,23 @@ qalloc(void)
 		}
 		freeNum[NNALLOC - 1].next = NULL;
 		freeNum[NNALLOC - 1].links = 0;
-		for (temp = freeNum + NNALLOC - 2; temp >= freeNum; --temp) {
+
+		/*
+		 * We prevent the temp pointer from walking behind freeNum
+		 * by stopping one short of the end and running the loop one
+		 * more time.
+		 *
+		 * We would stop the loop with just temp >= freeNum, but
+		 * doing this helps make code checkers such as insure happy.
+		 */
+		for (temp = freeNum + NNALLOC - 2; temp > freeNum; --temp) {
 			temp->next = temp + 1;
 			temp->links = 0;
 		}
+		/* run the loop manually one last time */
+		temp->next = temp + 1;
+		temp->links = 0;
+
 		blockcount++;
 		if (firstNums == NULL) {
 		    newfn = (NUMBER **) malloc(blockcount * sizeof(NUMBER *));
--- a/seed.c
+++ b/seed.c
@@ -53,9 +53,10 @@
 */
 #include <sys/types.h>
 #include <sys/stat.h>
-#include <sys/resource.h>
 #include <time.h>
+#include <sys/time.h>
 #include <sys/times.h>
+#include <sys/resource.h>
 #include <setjmp.h>
 #include "qmath.h"
 #include "longbits.h"
--- a/string.c
+++ b/string.c
@@ -979,10 +979,23 @@ stralloc(void)
 		}
 		freeStr[STRALLOC - 1].s_next = NULL;
 		freeStr[STRALLOC - 1].s_links = 0;
-		for (temp = freeStr + STRALLOC - 2; temp >= freeStr; --temp) {
+
+		/*
+		 * We prevent the temp pointer from walking behind freeStr
+		 * by stopping one short of the end and running the loop one
+		 * more time.
+		 *
+		 * We would stop the loop with just temp >= freeStr, but
+		 * doing this helps make code checkers such as insure happy.
+		 */
+		for (temp = freeStr + STRALLOC - 2; temp > freeStr; --temp) {
 			temp->s_next = temp + 1;
 			temp->s_links = 0;
 		}
+		/* run the loop manually one last time */
+		temp->s_next = temp + 1;
+		temp->s_links = 0;
+
 		blockcount++;
 		if (firstStrs == NULL) {
 		    newfn = (STRING **) malloc( blockcount * sizeof(STRING *));
--- a/symbol.c
+++ b/symbol.c
@@ -175,7 +175,10 @@ showglobals(void)
 			printf("\n");
 		}
 	}
-	printf(count ? "\n" : "No real-valued global variables\n");
+	if (count == 0) {
+		printf("No real-valued global variables\n");
+	}
+	putchar('\n');
 }


@@ -342,8 +345,17 @@ freeglobals(void)
 	GLOBAL *sp;		/* current global symbol pointer */
 	long count;		/* number of global variables freed */

+	/*
+	 * We prevent the hp pointer from walking behind globalhash
+	 * by stopping one short of the end and running the loop one
+	 * more time.
+	 *
+	 * We could stop the loop with just hp >= globalhash, but stopping
+	 * short and running the loop one last time manually helps make 
+	 * code checkers such as insure happy.
+	 */
 	count = 0;
-	for (hp = &globalhash[HASHSIZE-1]; hp >= globalhash; hp--) {
+	for (hp = &globalhash[HASHSIZE-1]; hp > globalhash; hp--) {
 		for (sp = *hp; sp; sp = sp->g_next) {
 			if (sp->g_value.v_type != V_NULL) {
 				freevalue(&sp->g_value);
@@ -351,6 +363,13 @@ freeglobals(void)
 			}
 		}
 	}
+	/* run the loop manually one last time */
+	for (sp = *hp; sp; sp = sp->g_next) {
+		if (sp->g_value.v_type != V_NULL) {
+			freevalue(&sp->g_value);
+			count++;
+		}
+	}
 }

 /*
@@ -539,14 +558,22 @@ unscope(void)
 	register GLOBAL *sp;		/* current global symbol pointer */
 	GLOBAL *prevsp;			/* previous kept symbol pointer */

-	for (hp = &globalhash[HASHSIZE-1]; hp >= globalhash; hp--) {
+	/*
+	 * We prevent the hp pointer from walking behind globalhash
+	 * by stopping one short of the end and running the loop one
+	 * more time.
+	 *
+	 * We could stop the loop with just hp >= globalhash, but stopping
+	 * short and running the loop one last time manually helps make 
+	 * code checkers such as insure happy.
+	 */
+	for (hp = &globalhash[HASHSIZE-1]; hp > globalhash; hp--) {
 		prevsp = NULL;
 		for (sp = *hp; sp; sp = sp->g_next) {
 			if ((sp->g_filescope == SCOPE_GLOBAL) ||
 				(sp->g_filescope < filescope) ||
 				((sp->g_filescope == filescope) &&
-					(sp->g_funcscope <= funcscope)))
-			{
+					(sp->g_funcscope <= funcscope))) {
 				prevsp = sp;
 				continue;
 			}
@@ -561,6 +588,26 @@ unscope(void)
 				*hp = sp->g_next;
 		}
 	}
+	/* run the loop manually one last time */
+	prevsp = NULL;
+	for (sp = *hp; sp; sp = sp->g_next) {
+		if ((sp->g_filescope == SCOPE_GLOBAL) ||
+			(sp->g_filescope < filescope) ||
+			((sp->g_filescope == filescope) &&
+				(sp->g_funcscope <= funcscope))) {
+			prevsp = sp;
+			continue;
+		}
+
+		/*
+		 * This symbol needs removing.
+		 */
+		addstatic(sp);
+		if (prevsp)
+			prevsp->g_next = sp->g_next;
+		else
+			*hp = sp->g_next;
+	}
 }


--- a/value.c
+++ b/value.c
@@ -22,8 +22,6 @@

 #define LINELEN 80		/* length of a typical tty line */

-extern int sys_nerr;
-
 /*
 * Free a value and set its type to undefined.
 *
@@ -194,6 +192,7 @@ copyvalue(VALUE *oldvp, VALUE *newvp)
 			break;
 		case V_OCTET:
 			newvp->v_type = V_NUM;
+			newvp->v_subtype = V_NOSUBTYPE;
 			newvp->v_num = itoq((long) *oldvp->v_octet);
 			break;
 		case V_NBLOCK:
@@ -312,6 +311,7 @@ negvalue(VALUE *vp, VALUE *vres)
 			return;
 		case V_OCTET:
 			vres->v_type = V_NUM;
+			vres->v_subtype = V_NOSUBTYPE;
 			vres->v_num = itoq(- (long) *vp->v_octet);
 			return;

@@ -343,6 +343,7 @@ addnumeric(VALUE *v1, VALUE *v2, VALUE *vres)
 	/*
 	 * add numeric values
 	 */
+	vres->v_subtype = V_NOSUBTYPE;
 	switch (TWOVAL(v1->v_type, v2->v_type)) {
 	case TWOVAL(V_NUM, V_NUM):
 		vres->v_num = qqadd(v1->v_num, v2->v_num);
@@ -2420,9 +2421,11 @@ acceptvalue(VALUE *v1, VALUE *v2)
 	if (fp) {
 		++stack;
 		stack->v_type = V_ADDR;
+		stack->v_subtype = V_NOSUBTYPE;
 		stack->v_addr = v1;
 		++stack;
 		stack->v_type = V_ADDR;
+		stack->v_subtype = V_NOSUBTYPE;
 		stack->v_addr = v2;
 		calculate(fp, 2);
 		ret = testvalue(stack);
@@ -2447,9 +2450,11 @@ precvalue(VALUE *v1, VALUE *v2)
 	if (fp) {
 		++stack;
 		stack->v_type = V_ADDR;
+		stack->v_subtype = V_NOSUBTYPE;
 		stack->v_addr = v1;
 		++stack;
 		stack->v_type = V_ADDR;
+		stack->v_subtype = V_NOSUBTYPE;
 		stack->v_addr = v2;
 		calculate(fp, 2);
 		ret = testvalue(stack);
@@ -2632,6 +2637,7 @@ sgnvalue(VALUE *vp, VALUE *vres)
 	switch (vp->v_type) {
 		case V_NUM:
 			vres->v_num = qsign(vp->v_num);
+			vres->v_subtype = vp->v_subtype;
 			return;
 		case V_COM:
 			c = comalloc();
@@ -2641,9 +2647,11 @@ sgnvalue(VALUE *vp, VALUE *vres)
 			c->imag = qsign(vp->v_com->imag);
 			vres->v_com = c;
 			vres->v_type = V_COM;
+			vres->v_subtype = V_NOSUBTYPE;
 			return;
 		case V_OCTET:
 			vres->v_type = V_NUM;
+			vres->v_subtype = V_NOSUBTYPE;
 			vres->v_num = itoq((long) (*vp->v_octet != 0));
 			return;
 		case V_OBJ:
@@ -2694,7 +2702,7 @@ printvalue(VALUE *vp, int flags)
 	if (type < 0) {
 		if (userfunc("error_print", vp))
 			return;
-		if (-type >= sys_nerr)
+		if (-type >= E__BASE)
 			printf("Error %d", -type);
 		else
 			printf("System error %d", -type);
--- a/value.h
+++ b/value.h
@@ -451,6 +451,7 @@ extern int checkobject(char *name);
 extern void showobjfuncs(void);
 extern void showobjtypes(void);
 extern int findelement(char *name);
+extern char *objtypename(unsigned long index);
 extern int objoffset(OBJECT *op, long index);


--- a/version.c
+++ b/version.c
@@ -18,7 +18,7 @@ static char *program;
 #define MAJOR_VER	2	/* major version */
 #define MINOR_VER	11	/* minor version */
 #define MAJOR_PATCH	0	/* patch level or 0 if no patch */
-#define MINOR_PATCH	"9.4.1"	/* test number or empty string if no patch */
+#define MINOR_PATCH	"9.4.5"	/* test number or empty string if no patch */

 /*
 * calc version constants
--- a/zfunc.c
+++ b/zfunc.c
@@ -1109,7 +1109,17 @@ zlog(ZVALUE z1, ZVALUE z2)
 	 */
 	val = _one_;
 	power = 0;
-	for (; zp >= squares; zp--, worth /= 2) {
+
+	/*
+	 * We prevent the zp pointer from walking behind squares
+	 * by stopping one short of the end and running the loop one
+	 * more time.
+	 *
+	 * We could stop the loop with just zp >= squares, but stopping
+	 * short and running the loop one last time manually helps make 
+	 * code checkers such as insure happy.
+	 */
+	for (; zp > squares; zp--, worth /= 2) {
 		if ((val.len + zp->len - 1) <= z1.len) {
 			zmul(val, *zp, &temp);
 			if (zrel(z1, temp) >= 0) {
@@ -1123,6 +1133,22 @@ zlog(ZVALUE z1, ZVALUE z2)
 		if (zp != squares)
 			zfree(*zp);
 	}
+	/* run the loop manually one last time */
+	if (zp == squares) {
+		if ((val.len + zp->len - 1) <= z1.len) {
+			zmul(val, *zp, &temp);
+			if (zrel(z1, temp) >= 0) {
+				zfree(val);
+				val = temp;
+				power += worth;
+			} else {
+				zfree(temp);
+			}
+		}
+		if (zp != squares)
+			zfree(*zp);
+	}
+
 	zfree(val);
 	return power;
 }
@@ -1163,7 +1189,17 @@ zlog10(ZVALUE z)
 	 */
 	val = _one_;
 	power = 0;
-	for (; zp >= _tenpowers_; zp--, worth /= 2) {
+
+	/*
+	 * We prevent the zp pointer from walking behind _tenpowers_
+	 * by stopping one short of the end and running the loop one
+	 * more time.
+	 *
+	 * We could stop the loop with just zp >= _tenpowers_, but stopping
+	 * short and running the loop one last time manually helps make 
+	 * code checkers such as insure happy.
+	 */
+	for (; zp > _tenpowers_; zp--, worth /= 2) {
 		if ((val.len + zp->len - 1) <= z.len) {
 			zmul(val, *zp, &temp);
 			if (zrel(z, temp) >= 0) {
@@ -1175,6 +1211,20 @@ zlog10(ZVALUE z)
 			}
 		}
 	}
+	/* run the loop manually one last time */
+	if (zp == _tenpowers_) {
+		if ((val.len + zp->len - 1) <= z.len) {
+			zmul(val, *zp, &temp);
+			if (zrel(z, temp) >= 0) {
+				zfree(val);
+				val = temp;
+				power += worth;
+			} else {
+				zfree(temp);
+			}
+		}
+	}
+
 	zfree(val);
 	return power;
 }
@@ -1281,11 +1331,21 @@ zfacrem(ZVALUE z1, ZVALUE z2, ZVALUE *rem)
 		worth *= 2;
 		count += worth;
 	}
+
 	/*
 	 * Now back down the list of squares, and see if the lower powers
 	 * will divide any more times.
 	 */
-	for (; zp >= squares; zp--, worth /= 2) {
+	/*
+	 * We prevent the zp pointer from walking behind squares
+	 * by stopping one short of the end and running the loop one
+	 * more time.
+	 *
+	 * We could stop the loop with just zp >= squares, but stopping
+	 * short and running the loop one last time manually helps make 
+	 * code checkers such as insure happy.
+	 */
+	for (; zp > squares; zp--, worth /= 2) {
 		if (zp->len <= z1.len) {
 			zdiv(z1, *zp, &temp1, &temp2, 0);
 			if (ziszero(temp2)) {
@@ -1300,6 +1360,23 @@ zfacrem(ZVALUE z1, ZVALUE z2, ZVALUE *rem)
 		if (zp != squares)
 			zfree(*zp);
 	}
+	/* run the loop manually one last time */
+	if (zp == squares) {
+		if (zp->len <= z1.len) {
+			zdiv(z1, *zp, &temp1, &temp2, 0);
+			if (ziszero(temp2)) {
+				temp3 = z1;
+				z1 = temp1;
+				temp1 = temp3;
+				count += worth;
+			}
+			zfree(temp1);
+			zfree(temp2);
+		}
+		if (zp != squares)
+			zfree(*zp);
+	}
+
 	*rem = z1;
 	return count;
 }
--- a/zmath.c
+++ b/zmath.c
@@ -733,18 +733,18 @@ done:	while (m > 0 && A[m - 1] == 0)
 		f = g = 0;
 		t = -1;
 		if (s) {
-			while (--i > 0 ) {
+			while (--i > 0) {
 				g = (FULL) *--a + (*--b >> 1 | f);
 				if (g != BASE1)
 					break;
 				f = *b & 1 ? TOPHALF : 0;
 			}
 			if (g == BASE && f == 0) {
-				while (i-- && (*a-- | *b--) == 0);
+				while ((--i > 0) && ((*--a | *--b) == 0));
 				t = (i > 0);
-			}
-			else if (g >= BASE)
+			} else if (g >= BASE) {
 				t = 1;
+			}
 		} else {
 			while (--i > 0) {
 				g = (FULL) *--a - (*--b >> 1 | f);
--- a/zmul.c
+++ b/zmul.c
@@ -280,6 +280,8 @@ domul(HALF *v1, LEN size1, HALF *v2, LEN size2, HALF *ans)
 	baseA = v1 + shift;
 	baseB = v1;
 	/*
+	 * XXX - is this still an issue?
+	 *
 	 * Saber-C Version 3.1 says:
 	 *
 	 *    W#26, Storing a bad pointer into auto variable dmul`baseC.
@@ -296,11 +298,13 @@ domul(HALF *v1, LEN size1, HALF *v2, LEN size2, HALF *ans)
 	 *	config("mul2", 2);
 	 *	pmod(3,a-1,a);
 	 *
+	 *	[[ NOTE: The above code no longer invokes this code. ]]
+	 *
 	 * When this code is executed, shift == 6 and v2 is 3 shorts
 	 * long (size2 == 2).  This baseC points 3 shorts beyond the
 	 * allocated end of v2.
 	 *
-	 * The stack was as follows:
+	 * The stack was as follows:   [[NOTE: line numbers may have changed]]
 	 *
 	 *	domul(v1=0x2d93d8, size1=12,
 	 *	      v2=0x2ded30, size2=2, ans=0x2ee8a8) at "zmul.c":313
@@ -314,6 +318,64 @@ domul(HALF *v1, LEN size1, HALF *v2, LEN size2, HALF *ans)
 	 *	evaluate(...) at "codegen.c":170
 	 *	getcommands(...) at "codegen.c":109
 	 *	main(...) at "calc.c":167
+	 *
+	 * The final domul() call point is the next executable line below.
+	 *
+	 ****
+	 *
+	 * The insure tool also reports a problem at this position:
+	 *
+	 * [zmul.c:319] **COPY_BAD_RANGE**
+	 * >>         baseC = v2 + shift;
+	 *
+	 *   Copying pointer which is out-of-range: v2 + shift
+	 *
+	 *		[[NOTE: line numbers may have changed]]
+	 *
+	 *   Pointer      : 0x1400919cc
+	 *   Actual block : 0x140090c80 thru 0x140090def (368 bytes,92 elements)
+	 *                  hp, allocated at:
+	 *                           malloc()
+	 *                            alloc()  zmath.c, 221
+	 *                             zmul()  zmul.c, 73
+	 *                          ztenpow()  zfunc.c, 441
+	 *                            str2q()  qio.c, 537
+	 *                        addnumber()  const.c, 52
+	 *                        eatnumber()  token.c, 594
+	 *                         gettoken()  token.c, 319
+	 *                      getcallargs()  codegen.c, 2358
+	 *
+	 *   Stack trace where the error occurred:
+	 *                            domul()  zmul.c, 319
+	 *                             zmul()  zmul.c, 74
+	 *                          ztenpow()  zfunc.c, 441
+	 *                            str2q()  qio.c, 537
+	 *                        addnumber()  const.c, 52
+	 *                        eatnumber()  token.c, 594
+	 *                         gettoken()  token.c, 319
+	 *                      getcallargs()  codegen.c, 2358
+	 *                        getidexpr()  codegen.c, 1998
+	 *                          getterm()  codegen.c, 1936
+	 *                    getincdecexpr()  codegen.c, 1820
+	 *                     getreference()  codegen.c, 1804
+	 *                     getshiftexpr()  codegen.c, 1758
+	 *                       getandexpr()  codegen.c, 1704
+	 *                        getorexpr()  codegen.c, 1682
+	 *                       getproduct()  codegen.c, 1654
+	 *                           getsum()  codegen.c, 1626
+	 *                      getrelation()  codegen.c, 1585
+	 *                       getandcond()  codegen.c, 1556
+	 *                        getorcond()  codegen.c, 1532
+	 *                       getaltcond()  codegen.c, 1499
+	 *                    getassignment()  codegen.c, 1442
+	 *                  getopassignment()  codegen.c, 1352
+	 *                      getexprlist()  codegen.c, 1318
+	 *                     getstatement()  codegen.c, 921
+	 *                         evaluate()  codegen.c, 219
+	 *                      getcommands()  codegen.c, 165
+	 *                             main()  calc.c, 321
+	 *
+	 * The final domul() call point is the next executable line below.
 	 */
 	/* 	ok to ignore on name domul`baseC */
 	baseC = v2 + shift;
--- a/zrandom.c
+++ b/zrandom.c
@@ -2527,7 +2527,8 @@ random_libcalc_cleanup(void)
 static void
 zfree_random(ZVALUE z)
 {
-	if (z.v != h_ndefvec && z.v != h_rdefvec && z.v != h_rdefvec_2 &&
+	if (z.v != NULL &&
+	    z.v != h_ndefvec && z.v != h_rdefvec && z.v != h_rdefvec_2 &&
 	    z.v != h_nvec01 && z.v != h_rvec01 &&
 	    z.v != h_nvec02 && z.v != h_rvec02 &&
 	    z.v != h_nvec03 && z.v != h_rvec03 &&
Author	SHA1	Message	Date
Landon Curt Noll	86c8e6dcf1	Release calc version 2.11.0t9.4.5	2017-05-21 15:38:33 -07:00
Landon Curt Noll	58d32c68f9	Release calc version 2.11.0t9.4.4	2017-05-21 15:38:33 -07:00
Landon Curt Noll	7d0b761de3	Release calc version 2.11.0t9.4.3	2017-05-21 15:38:33 -07:00
Landon Curt Noll	82ff31f246	Release calc version 2.11.0t9.4.2	2017-05-21 15:38:32 -07:00