Release calc version 2.12.5.6

BUGS

@@ -32,16 +32,25 @@ 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
 and bug fixes reports to:
 
-    calc-bugs at asthe dot com
+	calc-bug-report at asthe dot com
 
-    [[ NOTE: Replace 'at' with @, 'dot' is with . and remove the spaces ]]
-    [[ NOTE: The EMail address uses 'asthe', the web site URL uses 'isthe' ]]
+	NOTE: Remove spaces and replace 'at' with @, 'dot' with .
 
-Your subject must contain the words:
+    This replaces the old calc-bugs at asthe dot com address.
 
-       calc bug report
+    To be sure we see your EMail reporting a calc bug, please use the
+    following phase in your EMail Subject line:
 
-You may have additional words in your subject line.
+	calc bug report
+
+    That phrase in your subject line will help ensure your request
+    will get past our anti-spam filters.  You may have additional
+    words in your subject line.
+
+    However, you may find it more helpful to simply subscribe
+    to the calc-tester mailing list (see below) and then to
+    send your report to that mailing list as a wider set calc
+    testers may be able to help you.
 
 When you send your report, please include the following information:
 
@@ -68,23 +77,18 @@ of a context diff patch).
 
 Known bugs:
 
+    The output of the alg_config.cal resource file is bogus.
+    We would welcome a replacement for this code.
+
     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.
 
 =-=
 
-Problems that have known work-a-rounds:
-
-    * There is a bug in gcc v4.1.0 that causes calc to fail the regression
-      test.  The work-a-round is to compile with gcc v4.1.1 or later.  This
-      problems was observed on Fedora 5.
-
-=-=
-
 mis-features in calc:
 
-    Some problems are not bugs but rarther mis-features / things that could
+    Some problems are not bugs but rather mis-features / things that could
     work better.  The following is a list of mis-features that should be
     addressed and improved someday.
 
@@ -129,7 +133,34 @@ mis-features in calc:
 
       will not.
 
-## Copyright (C) 1999-2007  Landon Curt Noll
+=-=
+
+To subscribe to the calc-tester mailing list, visit the following URL:
+
+	http://www.isthe.com/chongo/tech/comp/calc/calc-tester.html
+
+    This is a low volume moderated mailing list.
+
+    This mailing list replaces calc-tester at asthe dot com list.
+
+    If you need a human to help you with your mailing list subscription,
+    please send EMail to our special:
+
+	calc-tester-maillist-help at asthe dot com
+
+	NOTE: Remove spaces and replace 'at' with @, 'dot' with .
+
+    address.  To be sure we see your EMail asking for help with your
+    mailing list subscription, please use the following phase in your
+    EMail Subject line:
+
+	calc tester mailing list help
+
+    That phrase in your subject line will help ensure your
+    request will get past our anti-spam filters.  You may have
+    additional words in your subject line.
+
+## Copyright (C) 1999-2014  Landon Curt Noll
 ##
 ## Calc is open software; you can redistribute it and/or modify it under
 ## the terms of the version 2.1 of the GNU Lesser General Public License
@@ -145,8 +176,8 @@ mis-features in calc:
 ## received a copy with calc; if not, write to Free Software Foundation, Inc.
 ## 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
 ##
-## @(#) $Revision: 30.1 $
-## @(#) $Id: BUGS,v 30.1 2007/03/16 11:09:46 chongo Exp $
+## @(#) $Revision: 30.5 $
+## @(#) $Id: BUGS,v 30.5 2014/10/12 12:23:43 chongo Exp $
 ## @(#) $Source: /usr/local/src/bin/calc/RCS/BUGS,v $
 ##
 ## Under source code control:	1994/03/18 14:06:13

CHANGES

@@ -1,4 +1,344 @@
-The following are the changes from calc version 2.12.4.6 to date:
+The following are the changes from calc version 2.12.5.6 to date:
+
+    Fixed a crash that showed up on Mac OS that was reported
+    by Richard Outerbridge <outer at interlog dot com> and
+    fixed by Stuart Henderson <stu at spacehopper dot org>.
+    Thanks goes to both!
+
+
+The following are the changes from calc version 2.12.5.4 to 2.12.5.5:
+
+    Recompile to match current RHEL7.2 libc and friends.
+
+    Added fix by Alexandre Fedotov <fedotov at mail dot desy dot de>
+    to prepend ${T} in front of the CALCPATH path components
+    ${CALC_SHAREDIR} and ${CUSTOMDIR}.  Add ${T} in front of ${HELPDIR}
+    and ${CUSTONHELPDIR} when making conf.h.
+
+    Improved the jacobi help page.
+
+    Rewrote gen_v1() in the lucas.cal resource file using the method
+    based on a paper:
+
+ 	"A note on primality tests for N = h*2^n-1", by Oystein J. Rodseth,
+	Department of Mathematics, University of Bergen, BIT Numerical
+	Mathematics. 34 (3): pp 451-454.
+
+	http://folk.uib.no/nmaoy/papers/luc.pdf
+
+    The improved gen_v1() function is capable of returning a value
+    for all valid values of h and n.  As a result, the trial tables
+    used by gen_v1() have been changed to a short list of values
+    to try, in order ot likelyhood of success, before doing an
+    enhaustive search for a v1 value to return.
+
+    Removed lucas_tbl.cal calc resource file.  This file was made
+    obsolete by the above rewrite of the lucas.cal resource file.
+    This file will be removed from the local cal directory and
+    from CALC_SHAREDIR during a 'make install', 'make cobber',
+    and 'make uninstall'.
+
+    Renamed gen_u0() to gen_u2() in lucas.cal.  Provided a gen_u0()
+    stub function that calls gen_u2() for backward compatibility.
+
+    The old gen_v1() method used by the Amdahl 6 group has been
+    renamed legacy_gen_v1() in lucas.cal.  This function is no
+    longer usde by the lucas(h, n) function to test the primality of
+    h*2^n-1.  It is preserved in lucas.cal for historical purposes.
+
+    The 'make clobber' rule will attempt to remove all files that
+    start with libcalc and start with libcustcalc.
+
+    The 'man' command is now an alias for the 'help' command.
+
+    Fixed extra /'s that were put into CALCPATH because of ${T}.
+    Fixed extra /'s that were compiled into HELPDIR and CUSTOMHELPDIR.
+
+    The fix in 2.12.5.4 to to prepend ${T} in front of the CALCPATH
+    path components ${CALC_SHAREDIR} and ${CUSTOMDIR} broke the
+    calc rpm build process.  The check-buildroot tool discovered
+    that the BUILDROOT directory had been improperly put into various
+    paths and binaries.  This has been fixed in 2.12.5.5.
+
+
+The following are the changes from calc version 2.12.5.3 to 2.12.5.3:
+
+    Calc version 2.12.5.2 for Darwin (Mac OS X) users, code to installed
+    calc under /opt/calc.  Moreover the CHANGES file did not mention
+    /opt/calc.  Sorry about that!.
+
+    A much better tree for Darwin (Mac OS X) users would have been
+    to install cal under /opt/calc.  This release ONLY changes the Darwin
+    (Mac OS X) install tree to /usr/local.
+
+    Darwin (Mac OS X) users who installed calc version 2.12.5.2
+    should, after installing version 2.12.5.3:
+
+    	rm -rf /opt/calc
+
+
+The following are the changes from calc version 2.12.5.1 to 2.12.5.2:
+
+    NOTE: calc version 2.12.5.2, for Darwin (Mac OS X) users,
+    	  installed under /opt/calc.  We neglected to mention this
+	  AND /usr/local would have been a better choice.  Sorry!
+	  Fixed in calc version 2.12.5.3.
+
+    Removed rules and makefile variables assocated with shortened
+    calc version numbers of less than 4 levels.
+
+    Under OS X (darwin), if /usr/include is missing, warnings
+    are issued to help the user use xcode-select --install
+    so that one may properly compile C code.
+
+    Lowered REDC levels:
+
+	#define MAXREDC 256	/* number of entries in REDC cache */
+
+	#define SQ_ALG2 28	/* size for alternative squaring */
+	    config("sq2") == 28		/* was 3388 */
+	#define MUL_ALG2 28	/* size for alternative multiply */
+	    config("mul2") == 28	/* was 1780 */
+	#define POW_ALG2 20	/* size for using REDC for powers */
+	    config("pow2") == 20	/* was 176 */
+	#define REDC_ALG2 25	/* size for using alternative REDC */
+	    config("redc2") == 25	/* was 220 */
+
+    The alg_config.cal script appears to be not correctly finding the
+    best REDC values.  While it has been improved, alg_config.cal still
+    seems to be suspect on how it attempts to find the best values.
+
+    Fixed an intro help file mistake found by Roger Hardiman
+    <roger at rjh dot org dot uk>.
+
+
+The following are the changes from calc version 2.12.5.0 to 2.12.5.1:
+
+    Calc has a new calc-tester mailing list.  This list is for those
+    who are using/testing calc.  We also use this list to announce
+    new versions of calc.  To subscribe to the calc-tester mailing
+    list, visit the following URL:
+
+	    http://www.isthe.com/chongo/tech/comp/calc/calc-tester.html
+
+	This is a low volume moderated mailing list.
+
+	This mailing list replaces calc-tester at asthe dot com list.
+
+	If you need a human to help you with your mailing list subscription,
+	please send EMail to our special:
+
+	    calc-tester-maillist-help at asthe dot com
+
+	address.  To be sure we see your EMail asking for help with your
+	mailing list subscription, please use the following phase in your
+	EMail Subject line:
+
+	    calc tester mailing list help
+
+	That phrase in your subject line will help ensure your
+	request will get past our anti-spam filters.  You may have
+	additional words in your subject line.
+
+    There is a new calc bug report EMail address:
+
+	    calc-bug-report at asthe dot com
+
+	This replaces the old calc-bugs at asthe dot com address.
+
+	To be sure we see your EMail reporting a calc bug, please use the
+	following phase in your EMail Subject line:
+
+	    calc bug report
+
+	That phrase in your subject line will help ensure your
+	request will get past our anti-spam filters.  You may have
+	additional words in your subject line.
+
+	However, you may find it more helpful to simply subscribe
+	to the calc-tester mailing list (see above) and then to
+	send your report to that mailing list as a wider set calc
+	testers may be able to help you.
+
+    The following makefile rules that were related to printing the
+    upper values of the calc version, rules that were made obsolete
+    in calc version 2.12.4.14, have been removed:
+
+	calc_vers calc_ver calc_ve
+	vers ver ve
+
+    Noted that the hash() builtin function, internally known as
+    quickhash (used for internal objects such as the associative
+    arrays as well as other internal processes) uses the deprcated
+    32-bit FNV-0 hash.  The use of this deprecated hash is sufficient
+    for calc internal purposes.  Use of FNV-1a is recommended for
+    a general non-cryptographic quick hash.
+
+
+The following are the changes from calc version 2.12.4.14 to 2.12.5.0:
+
+    For Apple OS X / Darwin target:
+
+    	MACOSX_DEPLOYMENT_TARGET is no longer defined
+	using clang compiler
+
+	By default, -install-name is used when forming shared libs.
+	To force -install-name to not be used, set SET_INSTALL_NAME=no.
+
+    The have_stdvs.c test uses <stdlib.h> and fixed va_start() test call
+    that didn't use last arg.
+
+    Fixed math_fmt (printf) in value.c where a LEN (SB32) be printed as %d.
+
+    Fixed a significant bug where that resulted in an incorrect
+    complex number comparison.  Thanks goes to David Binderman
+    <dcb314 at hotmail dot com> for identifying the subtle typo!
+
+    Make minor fixes to the make depend rule.
+
+    Fixed places were calc defined a reserved identifier that
+    begin with either __ or _[A-Z].  For example, __FILE_H__ has
+    been replaced with INCLUDE_FILE_H.
+
+    Fixed the addall3 example in the script help file.  Thanks for this
+    fix goes to Igor Furlan <igor dot furlan at gmail dot com>.
+
+    We made important fixes to the calc command line history:
+
+	Fixed a bug in the command line history where calc would somtimes
+	crash.  There was code that used memcpy() instead of memmove()
+	that could corrupt the command line history when entering a
+	into into history that was similar to a previous entry.  Thanks
+	goes to Einar Lielmanis <einars at spicausis dot lv> for first
+	identifying this mistake.
+
+	The calc command line history code, in general was not robust.
+	We made use a patch from Mathias Buhr <napcode at users dot sf
+	dot net>, that while it uses a bit more memory: is much more
+	flexible, readable and robust.  This patch replaced the improer
+	use of memcpy() (see above) with better code.  Thanks!
+
+    The alg_config.cal calc resource file has been reworked to produce
+    better diagnostics while attempting to determine the ideal values
+    for mul2, sq2, and pow2.  However, it has been shown that this
+    code is not correct.  Suggestions for a replacement are welcome!
+
+   	calc -u 'read alg_config; config("user_debug", 2),; best_mul2();'
+   	calc -u 'read alg_config; config("user_debug", 2),; best_sq2();'
+   	calc -u 'read alg_config; config("user_debug", 2),; best_pow2();'
+
+    Fixed a number of pedantic compiler warnings.
+
+    Removed -W and -Wno-comment from the the CCWARN makefile variable.
+
+    Removed no_implicit.arg makefile rule.  Removed HAVE_NO_IMPLICIT
+    makefile variable.  Removed no_implicit.c source file.
+
+    Added WNO_IMPLICT makefile variable to hold the compiler flag
+    -Wno-implicit for use on selective compile lines.
+
+    Added WNO_ERROR_LONG_LONG makefile variable to hold the compiler flag
+    -Wno-error=long-long for use on selective compile lines.
+
+    Added WNO_LONG_LONG makefile variable to hold the compiler flag
+    -Wno-long-long for use on selective compile lines.
+
+    The makefile variable ${MKDIR_ARG} has been replaced with just -p.
+
+    Minor fixes were made to the calc.spec.in file.
+
+    The target rpm architecture changed from i686 to x86_64.  For those
+    who do not run machine with x86_64, we continue to release a src
+    rpm. For those without the ability to process an rpm, we will always
+    to release src tarball.
+
+    When building the libcalc and libcustcalc shared  libraries,
+    ONLY the .so and .so.${VERSION} files are created.  The .so is
+    a symlink to the .so.${VERSION} file.  Here ${VERSION} is the
+    full "w.x.y.z" calc version.
+
+
+The following are the changes from calc version 2.12.4.11 to 2.12.4.13:
+
+    Fixed many typos in comments of the Makefile thanks to the review
+    work of Michael Somos.
+
+    Fixed typo in "help sysinfo".
+
+    The Makefile rule, debug, is now more verbose and prints more information
+    about the calc compiled constants.
+
+    Added a more of calc resource files by
+    Christoph Zurnieden <czurnieden at gmx dot de> including:
+
+    infinities.cal - handle infinities symbolically, a little helper file
+    intnum.cal - implementation of tanh sinh and Gauss-Legendre quadrature
+    smallfactors.cal - find the factors of a number < 2^32
+    strings.cal - implementation of isascii() and isblank()
+
+    Reformatted some calc resource files.  Cleanup in comment the headers
+    of some calc resource files.
+
+    Minor formatting changes to a few help files.
+
+    No need to be special picky about the test8900.cal calc resource file.
+
+    Added a number of ctype-like builtins:
+
+    isalnum - whether character is alpha-numeric
+    isalpha - whether character is alphabetic
+    iscntrl - whether character is a control character
+    isdigit - whether character is a digit character
+    isgraph - whether character is a graphical character
+    islower - whether character is lower case
+    isprint - whether character is a printable
+    ispunct - whether character is a punctuation
+    isspace - whether character is a space character
+    isupper - whether character is upper case
+    isxdigit - whether character a hexadecimal digit
+    strcasecmp - compare two strings, case independent
+    strncasecmp - compare two strings up to n characters, case independent
+    strtolower - transform an ASCII string to lower case
+    strtoupper - transform an ASCII string to upper case
+
+    For details on these new builtins, see their help messages.
+    Thanks goes to Inge Zurnieden <inge dot zurnieden at gmx dot de> for
+    these new builtins.
+
+    Calc source code is now picky v2.3 clean using:
+
+	picky -s -v file file2 ..
+
+    With the exception of:
+
+	help/errorcodes.sed
+	cal/set8700.line
+
+    Due to the long lines in those files, we use:
+
+	picky -w -s -v help/errorcodes.sed cal/set8700.line
+
+    For more information about the picky tool, see:
+
+	http://cis.csuohio.edu/~somos/picky.html
+
+    Removed functions from strings.cal that have been replaced by
+    the new ctype-like builtin functions.
+
+    Fixed cal/Makefile to include missing intnum.cal file.
+
+    Added detail_help_list make target to cal/Makefile.
+
+    The detaillist make target in help/Makefile is now
+    called detail_help_list.
+
+    Removed requirement of gen_u0(h, n, v1) in lucas.cal that h
+    be odd.  While still lucas(h, n) converts even h into an odd h
+    internally by incrementing n, gen_u0(h, n, v1) will output even
+    when h is even.
+
+
+The following are the changes from calc version 2.12.4.6 to version 2.12.4.10:
 
     Updated RPM build process to remove use of deprecated flags.
 
@@ -11,6 +351,56 @@ The following are the changes from calc version 2.12.4.6 to date:
     Libraries are chmodded as 0644 to allow for building rpms
     without root.
 
+    Silenced annoying warning about unused variable 'intp'
+    while compiline endian.c under some circumstances.
+
+    Fixed typo in redeclaration warnings.  Thanks to
+    Christoph Zurnieden <czurnieden at gmx dot de> for this report.
+
+    Added a number of calc resource files by
+    Christoph Zurnieden <czurnieden at gmx dot de> including:
+
+    bernpoly.cal - Computes the nth Bernoulli polynomial at z for any n,z
+    brentsolve.cal - root-finder implementwed with the Brent-Dekker trick
+    factorial.cal - product of the positive integers
+    factorial2.cal - variety of integer functions quasi-related to factoral
+    lambertw.cal - Computes Lambert's W-function at "z" at branch "branch"
+    lnseries.cal - Calculates a series of natural logarithms at 1,2,3,4...n
+    specialfunctions.cal - Calculates the value of the beta function
+    statistics.cal - a wide vareity of stastical functions
+    toomcook.cal - Multiply by way of the Toom-Cook algorithm
+    zeta2.cal - Calculate the value of the Hurwitz Zeta function
+
+    Fixed a makefile bug that prevented the those new calc resource
+    files from being installed.
+
+    Improved the formatting of the output from:
+
+	help resource
+
+    We replaced COPYING-LGPL with the version that is found at
+    http://www.gnu.org/licenses/lgpl-2.1.txt because that version
+    contans some whitespace formatting cleanup.  Otherwise the
+    license is the same.
+
+    We fixed a number of places where "the the" was used
+    when just "the" should be used.
+
+	NOTE: Fixes to grammar, spelling and minor formatting
+	      problems are welcome.  Please send us your patches!
+
+    With the exception of 3 source files, we became "picky" about
+    line lengths and other issues reported by the picky tool:
+
+    	cal/test8900.cal
+	cal/set8700.line
+	help/errorcodes.sed
+
+    The above 3 files now pass picky -w (OK except for line length).
+    For more information about the picky tool, see:
+
+	http://cis.csuohio.edu/~somos/picky.html
+
 
 The following are the changes from calc version 2.12.4.3 to 2.12.4.5:
 
@@ -1166,8 +1556,7 @@ The following are the changes from calc version 2.11.10 to 2.11.10:
 The following are the changes from calc version 2.11.9 to 2.11.9.3:
 
     Fixed calc man page examples to move -f to the end of the line.
-    Thanks goes to Michael Somos <somos at grail dot cba dot csuohio
-    dot edu> for pointing this out.
+    Thanks goes to Michael Somos for pointing this out.
 
     Linux and gcc now compiled with -Wall -W -Wno-comment.
 
@@ -2122,7 +2511,7 @@ The following are the changes from calc version 2.11.1 to 2.11.1t2.2:
     Cleaned up help display system.  Help file lines that begin with
     '##' are not displayed.
 
-    Calc source and documentation now uses the the these terms:
+    Calc source and documentation now uses these terms:
 
 	*.cal files	calc resource file
 	*.a files	calc binary link library
@@ -2160,7 +2549,7 @@ The following are the changes from calc version 2.11.0t10 to 2.11.0t11:
     Fixed whitespace to make the best use of 8 character tabs.
 
     Fixed some bugs relating to '// and %' in combination with some
-    of the the rounding modes based on a patch from Ernest Bowen
+    of the rounding modes based on a patch from Ernest Bowen
     <ernie at turing dot une dot edu dot au>.
 
     A patch from Klaus Alexander Seistrup <klaus at seistrup dot dk>, when
@@ -2354,7 +2743,7 @@ The following are the changes from calc version 2.11.0t10 to 2.11.0t11:
 
 	Where ``/usr/local/bin/calc'' is the path to the calc binary.
 	Additional -options may be added to the line, but it MUST
-	start with -S.	For example, the the executable file ``plus''
+	start with -S.	For example, the executable file ``plus''
 	contain the following:
 
 	    #!/usr/local/bin/calc -S -e
@@ -4817,7 +5206,7 @@ The following are the changes from calc version 2.10.3t0 to 2.10.3t2:
 
 	global variable: to the end of the session with calc
 
-	static within a function definition: to the the first of:
+	static within a function definition: to the first of:
 
 	    an end of a global, static or local declaration (including
 	    initialization code) with the same identifier
@@ -4915,7 +5304,7 @@ The following are the changes from calc version 2.10.3t0 to 2.10.3t2:
 
 	global a, b, mat A, B[2] = {3,4}, C[2] = {4,5}, obj point P = {5,6}, Q
 
-    Fixed some bugs related to global and static scoping.  See the the
+    Fixed some bugs related to global and static scoping.  See the
     5200 regress test and lib/test5200.cal for details.
 
     Optimized opcode generator so that functions defined using '=' do not
@@ -5412,7 +5801,7 @@ The following are the changes from calc version 2.10.2t4 to 2.10.2t24:
 	 ILDFLAGS are flags given to ${CC} for linking .o files
 		  for intermediate progs
 
-	 CC is how the the C compiler is invoked
+	 CC is how the C compiler is invoked
 
     Added more tests to regress.cal.
 
@@ -5637,7 +6026,7 @@ The following are the changes from calc version 2.10.2t1 to 2.10.2t3:
     Some systems has a libc symbolic qadd() that conflicted with calc's
     qadd function.  To avoid this, qadd() has been renamed to qqadd().
 
-    The calc error codes are produced from the the calcerr.tbl file.
+    The calc error codes are produced from the calcerr.tbl file.
     Instead of changing #defines in value.h, one can not edit calcerr.tbl.
     The Makefile builds calcerr.h from this file.
 
@@ -6180,7 +6569,7 @@ The following are the changes from calc version 2.9.3t11 to 2.10.0t12:
 	LDFLAGS are flags given to ${CC} for linking .o files
 	ILDFLAGS are given to ${CC} for linking .o's for intermediate progs
 
-	CC is how the the C compiler is invoked
+	CC is how the C compiler is invoked
 
     The syntax error:
 
@@ -6815,7 +7204,7 @@ Following is a list of visible changes to calc from version 1.24.7 to 1.26.1:
 
     A small bug in the library file regress.cal was fixed.
 
-## Copyright (C) 2001  Landon Curt Noll
+## Copyright (C) 2001-2017  Landon Curt Noll
 ##
 ## Calc is open software; you can redistribute it and/or modify it under
 ## the terms of the version 2.1 of the GNU Lesser General Public License
@@ -6831,9 +7220,9 @@ Following is a list of visible changes to calc from version 1.24.7 to 1.26.1:
 ## received a copy with calc; if not, write to Free Software Foundation, Inc.
 ## 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
 ##
-## @(#) $Revision: 30.25 $
-## @(#) $Id: CHANGES,v 30.25 2013/05/05 10:15:30 chongo Exp $
-## @(#) $Source: /usr/local/src/bin/calc/RCS/CHANGES,v $
+## @(#) $Revision: 30.55 $
+## @(#) $Id: CHANGES,v 30.55 2017/05/21 02:22:22 chongo Exp $
+## @(#) $Source: /usr/local/src/bin/calc-RHEL7/RCS/CHANGES,v $
 ##
 ## Under source code control:	1993/06/02 18:12:57
 ## File existed as early as:	1989

COPYING

@@ -6,17 +6,17 @@ This file is Copyrighted
 
     This file is covered under the following Copyright:
 
-	Copyright (C) 1999-2008  Landon Curt Noll
+	Copyright (C) 1999-2014  Landon Curt Noll
 	All rights reserved.
 
 	Everyone is permitted to copy and distribute verbatim copies
 	of this license document, but changing it is not allowed.
 
-    # @(#) $Revision: 30.2 $
-    # @(#) $Id: COPYING,v 30.2 2008/10/24 10:46:52 chongo Exp $
+    # @(#) $Revision: 30.6 $
+    # @(#) $Id: COPYING,v 30.6 2014/10/12 12:23:43 chongo Exp $
     # @(#) $Source: /usr/local/src/bin/calc/RCS/COPYING,v $
 
-=-=
+-=-
 
 Calc is covered by the GNU Lesser General Public License
 --------------------------------------------------------
@@ -53,47 +53,64 @@ Calc is covered by the GNU Lesser General Public License
 	    Boston, MA  02110-1301
 	    USA
 
-    The contact addresses for calc is as follows:
+    To subscribe to the calc-tester mailing list, visit the following URL:
 
-	Web:	http://www.isthe.com/chongo/tech/comp/calc/email.html
+	    http://www.isthe.com/chongo/tech/comp/calc/calc-tester.html
 
-    To join the low volume calc mailing list.  Send a EMail message to:
+	This is a low volume moderated mailing list.
 
-	calc-tester-request at asthe dot com
+	This mailing list replaces calc-tester at asthe dot com list.
 
-    Your subject must contain the words:
+	If you need a human to help you with your mailing list subscription,
+	please send EMail to our special:
 
-	calc mailing list subscription
+	    calc-tester-maillist-help at asthe dot com
 
-    You may have additional words in your subject line.
+	    NOTE: Remove spaces and replace 'at' with @, 'dot' with .
 
-    Your message body (not the subject) should consist of:
+	address.  To be sure we see your EMail asking for help with your
+	mailing list subscription, please use the following phase in your
+	EMail Subject line:
 
-	 subscribe calc-tester address
-	 end
-	 name your_full_name
+	    calc tester mailing list help
 
-    where ``address'' is your EMail address and ``your_full_name'' is
-    your full name.
+	That phrase in your subject line will help ensure your
+	request will get past our anti-spam filters.  You may have
+	additional words in your subject line.
 
-    Feel free to follow the name line with additional EMail text as desired.
-
-=-=
+-=-
 
     Calc bug reports and calc bug fixes should be sent to:
 
-	calc-bugs at asthe dot com
+	    calc-bug-report at asthe dot com
 
-    [[ NOTE: Replace 'at' with @, 'dot' is with . and remove the spaces ]]
-    [[ NOTE: The EMail address uses 'asthe' and the web site URL uses 'isthe' ]]
+	    NOTE: Remove spaces and replace 'at' with @, 'dot' with .
 
-    Your subject must contain the words:
+	This replaces the old calc-bugs at asthe dot com address.
 
-	   calc bug report
+	To be sure we see your EMail reporting a calc bug, please use the
+	following phase in your EMail Subject line:
 
-    You may have additional words in your subject line.
+	    calc bug report
 
-=-=
+	That phrase in your subject line will help ensure your
+	request will get past our anti-spam filters.  You may have
+	additional words in your subject line.
+
+	However, you may find it more helpful to simply subscribe
+	to the calc-tester mailing list (see above) and then to
+	send your report to that mailing list as a wider set calc
+	testers may be able to help you.
+
+-=-
+
+    The calc web site is located at:
+
+	http://www.isthe.com/chongo/tech/comp/calc/
+
+    NOTE: The EMail address uses 'asthe', while the web site uses 'isthe'.
+
+-=-
 
 Calc's relationship to the GNU Lesser General Public License
 ------------------------------------------------------------
@@ -149,7 +166,7 @@ Calc's relationship to the GNU Lesser General Public License
      except for the exception files explicitly listed in the ``Calc
      copyrights and exception files'' section below.
 
-=-=
+-=-
 
 Calc copyrights and exception files
 -----------------------------------
@@ -165,6 +182,7 @@ Calc copyrights and exception files
 	Copyright (C) year  Ernest Bowen and Landon Curt Noll
 	Copyright (C) year  Ernest Bowen
 	Copyright (C) year  Petteri Kettunen and Landon Curt Noll
+	Copyright (C) year  Christoph Zurnieden
 
     These files are not covered under one of the Copyrights listed above:
 
@@ -188,7 +206,7 @@ Calc copyrights and exception files
     And because one may freely distribute the LGPL covered files, the
     entire calc source may be freely used and distributed.
 
-=-=
+-=-
 
 General Copyleft and License info
 ---------------------------------
@@ -202,7 +220,7 @@ General Copyleft and License info
 	http://www.gnu.org/copyleft/lesser.html
 	http://www.gnu.org/copyleft/lesser.txt
 
-=-=
+-=-
 
 Why calc did not use the GNU General Public License
 ---------------------------------------------------

COPYING-LGPL

@@ -1,8 +1,8 @@
-		  GNU LESSER GENERAL PUBLIC LICENSE
-		       Version 2.1, February 1999
+                  GNU LESSER GENERAL PUBLIC LICENSE
+                       Version 2.1, February 1999
 
  Copyright (C) 1991, 1999 Free Software Foundation, Inc.
-     51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA
+ 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301  USA
  Everyone is permitted to copy and distribute verbatim copies
  of this license document, but changing it is not allowed.
 
@@ -10,7 +10,7 @@
  as the successor of the GNU Library Public License, version 2, hence
  the version number 2.1.]
 
-			    Preamble
+                            Preamble
 
   The licenses for most software are designed to take away your
 freedom to share and change it.  By contrast, the GNU General Public
@@ -112,7 +112,7 @@ modification follow.  Pay close attention to the difference between a
 former contains code derived from the library, whereas the latter must
 be combined with the library in order to run.
 
-		  GNU LESSER GENERAL PUBLIC LICENSE
+                  GNU LESSER GENERAL PUBLIC LICENSE
    TERMS AND CONDITIONS FOR COPYING, DISTRIBUTION AND MODIFICATION
 
   0. This License Agreement applies to any software library or other
@@ -146,7 +146,7 @@ such a program is covered only if its contents constitute a work based
 on the Library (independent of the use of the Library in a tool for
 writing it).  Whether that is true depends on what the Library does
 and what the program that uses the Library does.
-  
+
   1. You may copy and distribute verbatim copies of the Library's
 complete source code as you receive it, in any medium, provided that
 you conspicuously and appropriately publish on each copy an
@@ -432,7 +432,7 @@ decision will be guided by the two goals of preserving the free status
 of all derivatives of our free software and of promoting the sharing
 and reuse of software generally.
 
-			    NO WARRANTY
+                            NO WARRANTY
 
   15. BECAUSE THE LIBRARY IS LICENSED FREE OF CHARGE, THERE IS NO
 WARRANTY FOR THE LIBRARY, TO THE EXTENT PERMITTED BY APPLICABLE LAW.
@@ -455,7 +455,7 @@ FAILURE OF THE LIBRARY TO OPERATE WITH ANY OTHER SOFTWARE), EVEN IF
 SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH
 DAMAGES.
 
-		     END OF TERMS AND CONDITIONS
+                     END OF TERMS AND CONDITIONS
 
            How to Apply These Terms to Your New Libraries
 
@@ -485,7 +485,8 @@ convey the exclusion of warranty; and each file should have at least the
 
     You should have received a copy of the GNU Lesser General Public
     License along with this library; if not, write to the Free Software
-    Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA
+    Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301
+    USA
 
 Also add information on how to contact you by electronic and paper mail.
 
@@ -500,5 +501,3 @@ necessary.  Here is a sample; alter the names:
   Ty Coon, President of Vice
 
 That's all there is to it!
-
-

README

@@ -73,52 +73,64 @@ for a wish/todo list.  Code contributions are welcome.
 
 =-=
 
-To join the calc-tester mailing list.  Send an EMail message to:
+To subscribe to the calc-tester mailing list, visit the following URL:
 
-    calc-tester-request at asthe dot com
+	http://www.isthe.com/chongo/tech/comp/calc/calc-tester.html
 
-    [[ NOTE: Replace 'at' with @, 'dot' is with . and remove the spaces ]]
-    [[ NOTE: The EMail address uses 'asthe' and the web site URL uses 'isthe' ]]
+    This is a low volume moderated mailing list.
 
-Your subject must contain the words:
+    This mailing list replaces calc-tester at asthe dot com list.
 
-    calc mailing list subscription
+    If you need a human to help you with your mailing list subscription,
+    please send EMail to our special:
 
-You may have additional words in your subject line.
+	calc-tester-maillist-help at asthe dot com
 
-Your message body (not the subject) should consist of:
+	NOTE: Remove spaces and replace 'at' with @, 'dot' with .
 
-     subscribe calc-tester address
-     end
-     name your_full_name
+    address.  To be sure we see your EMail asking for help with your
+    mailing list subscription, please use the following phase in your
+    EMail Subject line:
 
-where ``address'' is your EMail address and ``your_full_name'' is
-your full name.
+	calc tester mailing list help
 
-Feel free to follow the name line with additional EMail text as desired.
+    That phrase in your subject line will help ensure your
+    request will get past our anti-spam filters.  You may have
+    additional words in your subject line.
 
-=-=
+-=-
 
-Send Calc bug and bug fixes to:
+Calc bug reports and calc bug fixes should be sent to:
 
-    calc-bugs at asthe dot com
+	calc-bug-report at asthe dot com
 
-    [[ NOTE: Replace 'at' with @, 'dot' is with . and remove the spaces ]]
-    [[ NOTE: The EMail address uses 'asthe' and the web site URL uses 'isthe' ]]
+	NOTE: Remove spaces and replace 'at' with @, 'dot' with .
 
-but see the BUGS file first.
+    This replaces the old calc-bugs at asthe dot com address.
 
-Your subject must contain the words:
+    To be sure we see your EMail reporting a calc bug, please use the
+    following phase in your EMail Subject line:
 
-       calc bug report
+	calc bug report
 
-You may have additional words in your subject line.
+    That phrase in your subject line will help ensure your
+    request will get past our anti-spam filters.  You may have
+    additional words in your subject line.
+
+    However, you may find it more helpful to simply subscribe
+    to the calc-tester mailing list (see above) and then to
+    send your report to that mailing list as a wider set calc
+    testers may be able to help you.
+
+-=-
 
 The calc web site is located at:
 
-	http://www.isthe.com/chongo/tech/comp/calc/
+    http://www.isthe.com/chongo/tech/comp/calc/
 
-## Copyright (C) 1999  Landon Curt Noll
+NOTE: The EMail address uses 'asthe', while the web site uses 'isthe'.
+
+## Copyright (C) 1999,2014  Landon Curt Noll
 ##
 ## Calc is open software; you can redistribute it and/or modify it under
 ## the terms of the version 2.1 of the GNU Lesser General Public License
@@ -134,8 +146,8 @@ The calc web site is located at:
 ## received a copy with calc; if not, write to Free Software Foundation, Inc.
 ## 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
 ##
-## @(#) $Revision: 30.1 $
-## @(#) $Id: README,v 30.1 2007/03/16 11:09:46 chongo Exp $
+## @(#) $Revision: 30.3 $
+## @(#) $Id: README,v 30.3 2014/10/12 12:23:43 chongo Exp $
 ## @(#) $Source: /usr/local/src/bin/calc/RCS/README,v $
 ##
 ## Under source code control:	1995/10/25 05:27:59

alloc.h

@@ -1,7 +1,7 @@
 /*
  * alloc - storage allocation and storage debug macros
  *
- * Copyright (C) 1999-2007  David I. Bell
+ * Copyright (C) 1999-2007,2014  David I. Bell
  *
  * Calc is open software; you can redistribute it and/or modify it under
  * the terms of the version 2.1 of the GNU Lesser General Public License
@@ -17,8 +17,8 @@
  * received a copy with calc; if not, write to Free Software Foundation, Inc.
  * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
  *
- * @(#) $Revision: 30.2 $
- * @(#) $Id: alloc.h,v 30.2 2008/04/15 21:17:57 chongo Exp $
+ * @(#) $Revision: 30.5 $
+ * @(#) $Id: alloc.h,v 30.5 2014/08/24 21:56:51 chongo Exp $
  * @(#) $Source: /usr/local/src/bin/calc/RCS/alloc.h,v $
  *
  * Under source code control:	1990/02/15 01:48:29
@@ -28,8 +28,8 @@
  */
 
 
-#if !defined(__ALLOC_H__)
-#define __ALLOC_H__
+#if !defined(INCLUDE_ALLOC_H)
+#define INCLUDE_ALLOC_H
 
 
 #if defined(CALC_SRC)	/* if we are building from the calc source tree */
@@ -53,7 +53,8 @@
 # if defined(HAVE_NEWSTR)
 E_FUNC void *memcpy();
 E_FUNC void *memset();
-#if defined(FORCE_STDC) || (defined(__STDC__) && __STDC__ != 0) || defined(__cplusplus)
+#if defined(FORCE_STDC) || \
+   (defined(__STDC__) && __STDC__ != 0) || defined(__cplusplus)
 E_FUNC size_t strlen();
 #  else
 E_FUNC long strlen();
@@ -82,7 +83,8 @@ E_FUNC int strcmp();
 
 #if !defined(HAVE_MEMMOVE)
 # undef MEMMOVE_SIZE_T
-#if defined(FORCE_STDC) || (defined(__STDC__) && __STDC__ != 0) || defined(__cplusplus)
+#if defined(FORCE_STDC) || \
+   (defined(__STDC__) && __STDC__ != 0) || defined(__cplusplus)
 #  define MEMMOVE_SIZE_T size_t
 # else
 #  define MEMMOVE_SIZE_T long
@@ -90,4 +92,4 @@ E_FUNC int strcmp();
 E_FUNC void *memmove(void *s1, CONST void *s2, MEMMOVE_SIZE_T n);
 #endif
 
-#endif /* !__ALLOC_H__ */
+#endif /* !INCLUDE_ALLOC_H */

assocfunc.c

@@ -17,8 +17,8 @@
  * received a copy with calc; if not, write to Free Software Foundation, Inc.
  * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
  *
- * @(#) $Revision: 30.1 $
- * @(#) $Id: assocfunc.c,v 30.1 2007/03/16 11:09:46 chongo Exp $
+ * @(#) $Revision: 30.3 $
+ * @(#) $Id: assocfunc.c,v 30.3 2014/09/30 00:55:11 chongo Exp $
  * @(#) $Source: /usr/local/src/bin/calc/RCS/assocfunc.c,v $
  *
  * Under source code control:	1993/07/20 23:04:27
@@ -83,7 +83,7 @@ associndex(ASSOC *ap, BOOL create, long dim, VALUE *indices)
 	 * so that we can first select the correct hash chain, and
 	 * also so we can quickly compare each element for a match.
 	 */
-	hash = FNV1_32_BASIS;
+	hash = QUICKHASH_BASIS;
 	for (i = 0; i < dim; i++)
 		hash = hashvalue(&indices[i], hash);
 
@@ -332,7 +332,8 @@ assoccopy(ASSOC *oldap)
 			oldep = oldep->e_next) {
 			ep = (ASSOCELEM *) malloc(ELEMSIZE(oldep->e_dim));
 			if (ep == NULL) {
-				math_error("Cannot allocate association element");
+				math_error("Cannot allocate "
+					   "association element");
 				/*NOTREACHED*/
 			}
 			ep->e_dim = oldep->e_dim;
@@ -340,7 +341,8 @@ assoccopy(ASSOC *oldap)
 			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_indices[i],
+					  &ep->e_indices[i]);
 			copyvalue(&oldep->e_value, &ep->e_value);
 			listhead = &ap->a_table[ep->e_hash % ap->a_size];
 			ep->e_next = *listhead;

blkcpy.c

@@ -19,8 +19,8 @@
  * received a copy with calc; if not, write to Free Software Foundation, Inc.
  * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
  *
- * @(#) $Revision: 30.1 $
- * @(#) $Id: blkcpy.c,v 30.1 2007/03/16 11:09:46 chongo Exp $
+ * @(#) $Revision: 30.2 $
+ * @(#) $Id: blkcpy.c,v 30.2 2013/08/11 08:41:38 chongo Exp $
  * @(#) $Source: /usr/local/src/bin/calc/RCS/blkcpy.c,v $
  *
  * Under source code control:	1997/04/18 20:41:26
@@ -374,7 +374,8 @@ copyblk2mat(BLOCK *blk, long ssi, long num, MATRIX *dmat, long dsi)
  * copymat2blk - copy matrix to block
  */
 int
-copymat2blk(MATRIX *smat, long ssi, long num, BLOCK *dblk, long dsi, BOOL noreloc)
+copymat2blk(MATRIX *smat, long ssi, long num, BLOCK *dblk, long dsi,
+	    BOOL noreloc)
 {
 	long i;
 	long newlen;
@@ -720,7 +721,8 @@ copystr2file(STRING *str, long ssi, long num, FILEID id, long dsi)
  * copyblk2blk - copy block to block
  */
 int
-copyblk2blk(BLOCK *sblk, long ssi, long num, BLOCK *dblk, long dsi, BOOL noreloc)
+copyblk2blk(BLOCK *sblk, long ssi, long num, BLOCK *dblk, long dsi,
+	    BOOL noreloc)
 {
 	long	newlen;
 	long	newsize;
@@ -762,7 +764,8 @@ copyblk2blk(BLOCK *sblk, long ssi, long num, BLOCK *dblk, long dsi, BOOL noreloc
  * copystr2blk - copy string to block
  */
 int
-copystr2blk(STRING *str, long ssi, long num, BLOCK *dblk, long dsi, BOOL noreloc)
+copystr2blk(STRING *str, long ssi, long num, BLOCK *dblk, long dsi,
+	    BOOL noreloc)
 {
 	long	len;
 	long	newlen;
@@ -982,7 +985,8 @@ memmove(void *s1, CONST void *s2, MEMMOVE_SIZE_T n)
  * copynum2blk - copy number numerator to block
  */
 int
-copynum2blk(NUMBER *snum, long ssi, long num, BLOCK *dblk, long dsi, BOOL noreloc)
+copynum2blk(NUMBER *snum, long ssi, long num, BLOCK *dblk, long dsi,
+	    BOOL noreloc)
 {
 	size_t	newlen;
 	size_t	newsize;
@@ -1033,7 +1037,8 @@ copynum2blk(NUMBER *snum, long ssi, long num, BLOCK *dblk, long dsi, BOOL norelo
  * copyblk2num - copy block to number
  */
 int
-copyblk2num(BLOCK *sblk, long ssi, long num, NUMBER *dnum, long dsi, NUMBER **res)
+copyblk2num(BLOCK *sblk, long ssi, long num, NUMBER *dnum, long dsi,
+	    NUMBER **res)
 {
 	size_t	newlen;
 	NUMBER *ret;		/* cloned and modified numerator */

blkcpy.h

@@ -1,7 +1,7 @@
 /*
  * blkcpy - general values and related routines used by the calculator
  *
- * Copyright (C) 1999-2007  Landon Curt Noll and Ernest Bowen
+ * Copyright (C) 1999-2007,2014  Landon Curt Noll and Ernest Bowen
  *
  * Primary author:  Landon Curt Noll
  *
@@ -19,8 +19,8 @@
  * received a copy with calc; if not, write to Free Software Foundation, Inc.
  * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
  *
- * @(#) $Revision: 30.1 $
- * @(#) $Id: blkcpy.h,v 30.1 2007/03/16 11:09:46 chongo Exp $
+ * @(#) $Revision: 30.3 $
+ * @(#) $Id: blkcpy.h,v 30.3 2014/08/24 21:56:51 chongo Exp $
  * @(#) $Source: /usr/local/src/bin/calc/RCS/blkcpy.h,v $
  *
  * Under source code control:	1997/04/18 20:41:25
@@ -30,8 +30,8 @@
  */
 
 
-#if !defined(__BLKCPY_H__)
-#define __BLKCPY_H__
+#if !defined(INCLUDE_BLKCPY_H)
+#define INCLUDE_BLKCPY_H
 
 /*
  * the main copy gateway function
@@ -59,4 +59,4 @@ E_FUNC int copystr2blk(STRING *, long, long, BLOCK *, long, BOOL);
 E_FUNC int copystr2file(STRING *, long, long, FILEID, long);
 E_FUNC int copystr2str(STRING *, long, long, STRING *, long);
 
-#endif /* !__BLKCPY_H__ */
+#endif /* !INCLUDE_BLKCPY_H */

block.h

@@ -1,7 +1,7 @@
 /*
  * block - fixed, dynamic, fifo and circular memory blocks
  *
- * Copyright (C) 1999-2007  Landon Curt Noll and Ernest Bowen
+ * Copyright (C) 1999-2007,2014  Landon Curt Noll and Ernest Bowen
  *
  * Primary author:  Landon Curt Noll
  *
@@ -19,8 +19,8 @@
  * received a copy with calc; if not, write to Free Software Foundation, Inc.
  * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
  *
- * @(#) $Revision: 30.1 $
- * @(#) $Id: block.h,v 30.1 2007/03/16 11:09:46 chongo Exp $
+ * @(#) $Revision: 30.3 $
+ * @(#) $Id: block.h,v 30.3 2014/08/24 21:56:51 chongo Exp $
  * @(#) $Source: /usr/local/src/bin/calc/RCS/block.h,v $
  *
  * Under source code control:	1997/02/21 05:03:39
@@ -31,8 +31,8 @@
  */
 
 
-#if !defined(__BLOCK_H__)
-#define __BLOCK_H__
+#if !defined(INCLUDE_BLOCK_H)
+#define INCLUDE_BLOCK_H
 
 
 /*
@@ -222,4 +222,4 @@ E_FUNC int countnblocks(void);
 E_FUNC void shownblocks(void);
 
 
-#endif /* !__BLOCK_H__ */
+#endif /* !INCLUDE_BLOCK_H */

byteswap.c

@@ -17,8 +17,8 @@
  * received a copy with calc; if not, write to Free Software Foundation, Inc.
  * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
  *
- * @(#) $Revision: 30.1 $
- * @(#) $Id: byteswap.c,v 30.1 2007/03/16 11:09:46 chongo Exp $
+ * @(#) $Revision: 30.2 $
+ * @(#) $Id: byteswap.c,v 30.2 2013/08/11 08:41:38 chongo Exp $
  * @(#) $Source: /usr/local/src/bin/calc/RCS/byteswap.c,v $
  *
  * Under source code control:	1995/10/11 04:44:01
@@ -102,7 +102,8 @@ swap_b8_in_ZVALUE(ZVALUE *dest, ZVALUE *src, BOOL all)
 		 */
 		dest = malloc(sizeof(ZVALUE));
 		if (dest == NULL) {
-			math_error("swap_b8_in_ZVALUE: swap_b8_in_ZVALUE: Not enough memory");
+			math_error("swap_b8_in_ZVALUE: swap_b8_in_ZVALUE: "
+				    "Not enough memory");
 			/*NOTREACHED*/
 		}
 

byteswap.h

@@ -1,7 +1,7 @@
 /*
  * byteswap - byte swapping macros
  *
- * Copyright (C) 1999  Landon Curt Noll
+ * Copyright (C) 1999,2014  Landon Curt Noll
  *
  * Calc is open software; you can redistribute it and/or modify it under
  * the terms of the version 2.1 of the GNU Lesser General Public License
@@ -17,8 +17,8 @@
  * received a copy with calc; if not, write to Free Software Foundation, Inc.
  * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
  *
- * @(#) $Revision: 30.1 $
- * @(#) $Id: byteswap.h,v 30.1 2007/03/16 11:09:46 chongo Exp $
+ * @(#) $Revision: 30.3 $
+ * @(#) $Id: byteswap.h,v 30.3 2014/08/24 21:56:51 chongo Exp $
  * @(#) $Source: /usr/local/src/bin/calc/RCS/byteswap.h,v $
  *
  * Under source code control:	1995/10/11 04:44:01
@@ -29,8 +29,8 @@
  */
 
 
-#if !defined(__BYTESWAP_H__)
-#define __BYTESWAP_H__
+#if !defined(INCLUDE_BYTESWAP_H)
+#define INCLUDE_BYTESWAP_H
 
 
 #if defined(CALC_SRC)	/* if we are building from the calc source tree */
@@ -178,4 +178,4 @@
 #endif /* LONG_BITS == 64 */
 
 
-#endif /* !__BYTESWAP_H__ */
+#endif /* !INCLUDE_BYTESWAP_H */

cal/Makefile

@@ -2,7 +2,7 @@
 #
 # cal - makefile for calc standard resource files
 #
-# Copyright (C) 1999-2006  Landon Curt Noll
+# Copyright (C) 1999-2006,2017  Landon Curt Noll
 #
 # Calc is open software; you can redistribute it and/or modify it under
 # the terms of the version 2.1 of the GNU Lesser General Public License
@@ -18,9 +18,9 @@
 # received a copy with calc; if not, write to Free Software Foundation, Inc.
 # 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
 #
-# @(#) $Revision: 30.5 $
-# @(#) $Id: Makefile,v 30.5 2011/05/23 22:50:18 chongo Exp $
-# @(#) $Source: /usr/local/src/bin/calc/cal/RCS/Makefile,v $
+# @(#) $Revision: 30.12 $
+# @(#) $Id: Makefile,v 30.12 2017/05/19 16:09:14 chongo Exp $
+# @(#) $Source: /usr/local/src/bin/calc-RHEL7/cal/RCS/Makefile,v $
 #
 # Under source code control:	1991/07/21 05:00:54
 # File existed as early as:	1991
@@ -179,22 +179,41 @@ MV= mv
 CO= co
 TRUE= true
 TOUCH= touch
+SED= sed
+SORT= sort
+FMT= fmt
 
 # The calc files to install
 #
-CALC_FILES= README bigprime.cal deg.cal ellip.cal lucas.cal lucas_chk.cal \
-	lucas_tbl.cal mersenne.cal mod.cal pell.cal pi.cal pix.cal \
-	pollard.cal poly.cal psqrt.cal quat.cal regress.cal solve.cal \
-	sumsq.cal surd.cal unitfrac.cal varargs.cal chrem.cal mfactor.cal \
-	bindings randmprime.cal test1700.cal randrun.cal linear.cal \
-	randbitrun.cal bernoulli.cal test2300.cal test2600.cal \
-	test2700.cal test3100.cal test3300.cal test3400.cal prompt.cal \
-	test3500.cal seedrandom.cal test4000.cal test4100.cal test4600.cal \
-	beer.cal hello.cal test5100.cal test5200.cal randombitrun.cal \
-	randomrun.cal repeat.cal xx_print.cal natnumset.cal qtime.cal \
-	test8400.cal test8500.cal test8600.cal chi.cal intfile.cal screen.cal \
-	dotest.cal set8700.cal set8700.line alg_config.cal sumtimes.cal \
-	dms.cal hms.cal gvec.cal
+# This list is prodiced by the detaillist rule when no WARNINGS are detected.
+#
+# Please use:
+#
+#	make calc_files_list
+#
+# to keep this list in nice sorted order and to check that these
+# deailed help files are under RCS control.
+#
+CALC_FILES= README alg_config.cal beer.cal bernoulli.cal \
+	bernpoly.cal bigprime.cal bindings brentsolve.cal chi.cal chrem.cal \
+	constants.cal deg.cal dms.cal dotest.cal ellip.cal factorial.cal \
+	factorial2.cal gvec.cal hello.cal hms.cal infinities.cal \
+	intfile.cal intnum.cal lambertw.cal linear.cal lnseries.cal \
+	lucas.cal lucas_chk.cal mersenne.cal mfactor.cal \
+	mod.cal natnumset.cal pell.cal pi.cal pix.cal pollard.cal poly.cal \
+	prompt.cal psqrt.cal qtime.cal quat.cal randbitrun.cal randmprime.cal \
+	randombitrun.cal randomrun.cal randrun.cal regress.cal repeat.cal \
+	screen.cal seedrandom.cal set8700.cal set8700.line smallfactors.cal \
+	solve.cal specialfunctions.cal statistics.cal strings.cal sumsq.cal \
+	sumtimes.cal surd.cal test1700.cal test2300.cal test2600.cal \
+	test2700.cal test3100.cal test3300.cal test3400.cal test3500.cal \
+	test4000.cal test4100.cal test4600.cal test5100.cal test5200.cal \
+	test8400.cal test8500.cal test8600.cal test8900.cal toomcook.cal \
+	unitfrac.cal varargs.cal xx_print.cal zeta2.cal
+
+# These calc files are now obsolete and are removed by the install rule.
+#
+DEAD_CALC_FILES= lucas_tbl.cal
 
 # These files are found (but not built) in the distribution
 #
@@ -242,6 +261,27 @@ calcliblist:
 	    fi; \
 	done
 
+# These next rule help form the ${CALC_FILES} makefile variables above.
+#
+calc_files_list:
+	${Q} -(find .  -mindepth 1 -maxdepth 1 -type f -name '*.cal' -print | \
+	   while read i; do \
+		if [ X"$$i" != X"/dev/null" ]; then \
+		    if [ ! -f RCS/$$i,v ]; then \
+			echo "WARNING: $$i not under RCS control" 1>&2; \
+		    else \
+			echo $$i; \
+		    fi; \
+		fi; \
+	done; \
+	echo '--first_line--'; \
+	echo README; \
+	echo set8700.line; \
+	echo bindings) | \
+	    ${SED} -e 's:^\./::' | LANG=C ${SORT} | ${FMT} -70 | \
+	    ${SED} -e '1s/--first_line--/CALC_FILES=/' -e '2,$$s/^/	/' \
+		   -e 's/$$/ \\/' -e '$$s/ \\$$//'
+
 ##
 #
 # rpm rules
@@ -265,6 +305,23 @@ clean:
 
 clobber: clean
 	${RM} -f .all
+	-${Q} if [ -e .DS_Store ]; then \
+	    echo ${RM} -rf .DS_Store; \
+	    ${RM} -rf .DS_Store; \
+	fi
+	-${Q} for i in ${DEAD_CALC_FILES} /dev/null; do \
+	    if [ "$$i" = "/dev/null" ]; then \
+		continue; \
+	    fi; \
+	    if [ -e "${T}${CALC_SHAREDIR}/$$i" ]; then \
+		echo "${RM} -f ${T}${CALC_SHAREDIR}/$$i"; \
+		${RM} -f ${T}${CALC_SHAREDIR}/$$i; \
+	    fi; \
+	    if [ -e "./$$i" ]; then \
+		echo "${RM} -f ./$$i"; \
+		${RM} -f ./$$i; \
+	    fi; \
+	done
 
 # install everything
 #
@@ -272,8 +329,8 @@ clobber: clean
 #
 install: all
 	-${Q} if [ ! -d ${T}${CALC_SHAREDIR} ]; then \
-	    echo ${MKDIR} ${T}${CALC_SHAREDIR}; \
-	    ${MKDIR} ${T}${CALC_SHAREDIR}; \
+	    echo ${MKDIR} -p ${T}${CALC_SHAREDIR}; \
+	    ${MKDIR} -p ${T}${CALC_SHAREDIR}; \
 	    if [ ! -d "${T}${CALC_SHAREDIR}" ]; then \
 		echo ${MKDIR} -p "${T}${CALC_SHAREDIR}"; \
 		${MKDIR} -p "${T}${CALC_SHAREDIR}"; \
@@ -293,16 +350,42 @@ install: all
 		${RM} -f ${T}${CALC_SHAREDIR}/$$i.new; \
 		${CP} -f $$i ${T}${CALC_SHAREDIR}/$$i.new; \
 		${CHMOD} 0444 ${T}${CALC_SHAREDIR}/$$i.new; \
-		${MV} -f ${T}${CALC_SHAREDIR}/$$i.new ${T}${CALC_SHAREDIR}/$$i; \
+		${MV} -f ${T}${CALC_SHAREDIR}/$$i.new ${T}${CALC_SHAREDIR}/$$i;\
 		echo "installed ${T}${CALC_SHAREDIR}/$$i"; \
 	    fi; \
 	done
+	${Q} for i in ${DEAD_CALC_FILES} /dev/null; do \
+	    if [ "$$i" = "/dev/null" ]; then \
+		continue; \
+	    fi; \
+	    if [ -e "${T}${CALC_SHAREDIR}/$$i" ]; then \
+		echo "${RM} -f ${T}${CALC_SHAREDIR}/$$i"; \
+		${RM} -f ${T}${CALC_SHAREDIR}/$$i; \
+	    fi; \
+	    if [ -e "./$$i" ]; then \
+		echo "${RM} -f ./$$i"; \
+		${RM} -f ./$$i; \
+	    fi; \
+	done
 
 # Try to remove everything that was installed
 #
 # NOTE: Keep the uninstall rule in reverse order to the install rule
 #
 uninstall:
+	- ${Q} for i in ${DEAD_CALC_FILES} /dev/null; do \
+	    if [ "$$i" = "/dev/null" ]; then \
+		continue; \
+	    fi; \
+	    if [ -e "${T}${CALC_SHAREDIR}/$$i" ]; then \
+		echo "${RM} -f ${T}${CALC_SHAREDIR}/$$i"; \
+		${RM} -f ${T}${CALC_SHAREDIR}/$$i; \
+	    fi; \
+	    if [ -e "./$$i" ]; then \
+		echo "${RM} -f ./$$i"; \
+		${RM} -f ./$$i; \
+	    fi; \
+	done
 	-${Q} for i in ${CALC_FILES} /dev/null; do \
 	    if [ "$$i" = "/dev/null" ]; then \
 		continue; \

cal/alg_config.cal

@@ -1,7 +1,7 @@
 /*
  * alg_config - help determine optimal values for algorithm levels
  *
- * Copyright (C) 2006  Landon Curt Noll
+ * Copyright (C) 2006,2014  Landon Curt Noll
  *
  * Calc is open software; you can redistribute it and/or modify it under
  * the terms of the version 2.1 of the GNU Lesser General Public License
@@ -17,8 +17,8 @@
  * received a copy with calc; if not, write to Free Software Foundation, Inc.
  * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
  *
- * @(#) $Revision: 30.1 $
- * @(#) $Id: alg_config.cal,v 30.1 2007/03/16 11:09:54 chongo Exp $
+ * @(#) $Revision: 30.12 $
+ * @(#) $Id: alg_config.cal,v 30.12 2016/02/06 08:38:56 chongo Exp $
  * @(#) $Source: /usr/local/src/bin/calc/cal/RCS/alg_config.cal,v $
  *
  * Under source code control:	2006/06/07 14:10:11
@@ -28,8 +28,37 @@
  * Share and enjoy!  :-)	http://www.isthe.com/chongo/tech/comp/calc/
  */
 
+static test_time;	/* try for this many seconds in loop test */
 
-global test_time;	/* try for this many seconds in loop test */
+
+/*
+ * close_to_one - set to 1 if the ratio is close enough to 1
+ *
+ * given:
+ *	ratio	the ratio of time between two algorithms
+ *
+ * retuns:
+ *	1	When ratio is near 1.0
+ *	0	otherwise
+ *
+ * We consider the range [0.995, 1.005] to be close enough to 1 to be call unity
+ * because of the precision of the CPU timing.
+ */
+define close_to_one(ratio)
+{
+    /* firewall */
+    if (!isreal(ratio)) {
+	quit "close: 1st arg: must be a real number";
+    }
+
+    /* check if the ratio is far from unity */
+    if ((ratio < 0.995) || (ratio > 1.005)) {
+	return 0;
+    }
+
+    /* we are close to unity */
+    return 1;
+}
 
 
 /*
@@ -80,7 +109,8 @@ define mul_loop(repeat, x)
     len = sizeof((*x)[0]) / baseb_bytes;
     for (i=1; i < 4; ++i) {
 	if ((sizeof((*x)[i]) / baseb_bytes) != len) {
-	    quit "mul_loop: 2nd arg matrix elements are not of equal BASEB-bit word length";
+	    quit "mul_loop: 2nd arg matrix elements are not of "
+	         "equal BASEB-bit word length";
 	}
     }
 
@@ -142,7 +172,7 @@ define mul_loop(repeat, x)
  *	len	length in BASEB-bit words to multiply
  *
  * return:
- *	ratio of (1st / 2nd) algorithm rate
+ *	ratio of (1st / 2nd) algorithm rate.
  *
  * When want to determine a rate to a precision of 1 part in 1000.
  * Most systems today return CPU time to at least 10 msec precision.
@@ -165,6 +195,7 @@ define mul_ratio(len)
     local tover;		/* est of time for loop overhead */
     local alg1_rate;		/* loop rate of 1st algorithm */
     local alg2_rate;		/* loop rate of 2nd algorithm */
+    local ret;			/* return ratio, or 1.0 */
     local i;
 
     /*
@@ -217,12 +248,12 @@ define mul_ratio(len)
      * determine the 1st algorithm rate
      */
     loops = max(1, ceil(loops * test_time / tlen));
-    if (loops < 8) {
+    if (loops < 16) {
 	if (config("user_debug") > 1) {
-	    printf("    we must expand loop test time to more than %d secs\n",
-		ceil(test_time * (8 / loops)));
+	    printf("    we must expand alg1 loop test time to about %d secs\n",
+		ceil(test_time * (16 / loops)));
 	}
-	loops = 8;
+	loops = 16;
     }
     if (config("user_debug") > 3) {
 	printf("\t    will try alg1 %d loops\n", loops);
@@ -263,12 +294,12 @@ define mul_ratio(len)
      * determine the 2nd algorithm rate
      */
     loops = max(1, ceil(loops * test_time / tlen));
-    if (loops < 8) {
+    if (loops < 16) {
 	if (config("user_debug") > 1) {
-	    printf("    we must expand loop test time to more than %d secs\n",
-		ceil(test_time * (8 / loops)));
+	    printf("    we must expand alg2 loop test time to about %d secs\n",
+		ceil(test_time * (16 / loops)));
 	}
-	loops = 8;
+	loops = 16;
     }
     tlen = mul_loop(loops, &x);
     if (config("user_debug") > 3) {
@@ -297,7 +328,12 @@ define mul_ratio(len)
     /*
      * return alg1 / alg2 rate ratio
      */
-    return (alg1_rate / alg2_rate);
+    ret = alg1_rate / alg2_rate;
+    if (config("user_debug") > 2) {
+	printf("\tprecise ratio is: %.f mul_ratio will return: %.3f\n",
+		alg1_rate / alg2_rate, ret);
+    }
+    return ret;
 }
 
 
@@ -319,30 +355,40 @@ define best_mul2()
 {
     local ratio;	/* previously calculated alg1/alg2 ratio */
     local low;		/* low loop value tested */
-    local high;		/* low loop value tested */
+    local high;		/* high loop value tested */
+    local mid;		/* between low and high */
+    local best_val;	/* value found with ratio closest to unity */
+    local best_ratio;	/* cloest ratio found to unity */
     local expand;	/* how fast to expand the length */
 
     /*
      * setup
      */
-    test_time = 15.0;
-    if (config("user_debug") > 0) {
-	printf("will start with loop test time of %d secs\n", test_time);
+    printf("WARNING: This tool may not be computing the correct best value\n");
+    test_time = 5.0;
+    printf("The best_mul2() function will take a LONG time to run!\n");
+    printf("It is important that best_mul2() run on an othwewise idle host!\n");
+    if (config("user_debug") <= 0) {
+	printf("To monitor progress, set user_debug to 2: "
+	       "config(\"user_debug\", 2)\n");
     }
+    printf("Starting with loop test time of %d secs\n", test_time);
 
     /*
      * firewall - must have a >1 ratio for the initial length
      */
-    high = 16;
+    high = 8;
+    best_val = high;
     if (config("user_debug") > 0) {
 	printf("testing multiply alg1/alg2 ratio for len = %d\n", high);
     }
     ratio = mul_ratio(high);
+    best_ratio = ratio;
     if (config("user_debug") > 1) {
-	printf("    multiply alg1/alg2 ratio = %.3f\n", ratio);
+	printf("    multiply alg1/alg2 ratio = %.6f\n", ratio);
     }
-    if (ratio <= 1.0) {
-	quit "best_mul2: tests imply config(\"mul2\") should be < 4";
+    if (ratio < 1.0) {
+	quit "best_mul2: tests imply mul2 should be < 16, which seems bogus";
     }
 
     /*
@@ -355,7 +401,7 @@ define best_mul2()
 	 * We will multiplicatively expand our test level until
 	 * the ratio drops below 1.0.
 	 */
-	expand = ((ratio >= 3.5) ? 16 : 2^round(ratio));
+	expand = 2;
 	low = high;
 	high *= expand;
 	if (config("user_debug") > 1) {
@@ -367,19 +413,63 @@ define best_mul2()
 	 * determine the alg1/alg2 test ratio for this new length
 	 */
 	if (high >= 2^31) {
-	    quit "best_mul2: tests imply config(\"mul2\") should be >= 2^31";
+	    quit "best_mul2: test implies mul2 >= 2^31, which seems bogus";
 	}
 	if (config("user_debug") > 0) {
 	    printf("testing multiply alg1/alg2 ratio for len = %d\n", high);
 	}
 	ratio = mul_ratio(high);
-	if (config("user_debug") > 1) {
-	    printf("    multiply alg1/alg2 ratio = %.3f\n", ratio);
+	if (abs(ratio - 1.0) < abs(best_ratio - 1.0)) {
+	    best_val = high;
+	    best_ratio = ratio;
+	    if (config("user_debug") > 1) {
+		printf("    len %d has a new cloest ratio to unity: %.6f\n",
+		       best_val, best_ratio);
+	    }
 	}
-    } while (ratio >= 1.0);
+	if (config("user_debug") > 1) {
+	    printf("    multiply alg1/alg2 ratio = %.6f\n", ratio);
+	}
+    } while (ratio > 1.0);
+
+    /*
+     * If we previously expanded more than by a factor of 2, then
+     * we may have jumped over the crossover point.  So now
+     * drop down powers of two until the ratio is again >= 1.0
+     */
+    if (expand > 2) {
+	do {
+
+	    /*
+	     * contract by 2
+	     */
+	    high /= 2;
+	    low = high / 2;
+	    if (config("user_debug") > 0) {
+		printf("retesting multiply alg1/alg2 ratio for len = %d\n",
+		       high);
+	    }
+	    ratio = mul_ratio(high);
+	    if (abs(ratio - 1.0) < abs(best_ratio - 1.0)) {
+		best_val = high;
+		best_ratio = ratio;
+		if (config("user_debug") > 1) {
+		    printf("    len %d has a new cloest ratio to unity: %.6f\n",
+			   best_val, best_ratio);
+		}
+	    }
+	    if (config("user_debug") > 1) {
+		printf("    multiply alg1/alg2 ratio = %.6f\n", ratio);
+	    }
+
+	} while (ratio <= 1.0);
+
+	/* now that the ratio flipped again, use the previous range */
+	low = high;
+	high = high * 2;
+    }
     if (config("user_debug") > 0) {
-	printf("alg1/alg2 ratio now < 1.0, starting binary search between %d and %d\n",
-	    low, high);
+	printf("Starting binary search between %d and %d\n", low, high);
     }
 
     /*
@@ -388,30 +478,54 @@ define best_mul2()
     while (low+1 < high) {
 
     	/* try the mid-point */
+	mid = int((low+high)/2);
 	if (config("user_debug") > 0) {
-	    printf("testing multiply alg1/alg2 ratio for len = %d\n",
-	    	int((low+high)/2));
+	    printf("testing multiply alg1/alg2 ratio for len = %d\n", mid);
+	}
+	ratio = mul_ratio(mid);
+	if (abs(ratio - 1.0) < abs(best_ratio - 1.0)) {
+	    best_val = mid;
+	    best_ratio = ratio;
+	    if (config("user_debug") > 1) {
+		printf("    len %d has a new cloest ratio to unity: %.6f\n",
+		       best_val, best_ratio);
+	    }
 	}
-	ratio = mul_ratio(int((low+high)/2));
 	if (config("user_debug") > 1) {
-	    printf("    multiply alg1/alg2 ratio = %.3f\n", ratio);
+	    printf("    len %d multiply alg1/alg2 ratio = %.6f\n", mid, ratio);
+	}
+
+	/* stop search if near unity */
+	if (close_to_one(ratio)) {
+	    low = mid;
+	    high = mid;
+	    if (config("user_debug") > 0) {
+	    	printf("\twe are close enough to unity ratio at: %d\n", mid);
+	    }
+	    break;
 	}
 
 	/* bump lower range up if we went over */
-	if (ratio >= 1.0) {
+	if (ratio > 1.0) {
 	    if (config("user_debug") > 2) {
 	    	printf("\tmove low from %d up to %d\n",
-		    low, int((low+high)/2));
+		    low, mid);
 	    }
-	    low = int((low+high)/2);
+	    low = mid;
 
 	/* drop higher range down if we went under */
 	} else {
 	    if (config("user_debug") > 2) {
 	    	printf("\tmove high from %d down to %d\n",
-		     high, int((low+high)/2));
+		     high, mid);
 	    }
-	    high = int((low+high)/2);
+	    high = mid;
+	}
+
+	/* report on test loop progress */
+	if (config("user_debug") > 1) {
+	    printf("\tsetting low: %d high: %d diff: %d\n",
+	    	   low, high, high-low);
 	}
     }
 
@@ -419,10 +533,15 @@ define best_mul2()
      * return on the suggested config("mul2") value
      */
     if (config("user_debug") > 0) {
-	printf("best value of config(\"mul2\") is %d\n",
-	    (ratio >= 1.0) ? high : low);
+	printf("Best value for multiply is near %d\n", best_val);
+	printf("Best multiply alg1/alg2 ratio is: %.6f\n", best_ratio);
+	printf("We suggest placing this line in your .calcrc:\n");
+	printf("config(\"mul2\", %d),;\n", best_val);
+	printf("WARNING: It is believed that the output "
+	       "of this resource file is bogus!\n");
+	printf("WARNING: You may NOT wish to follow the above suggeston.\n");
     }
-    return ((ratio >= 1.0) ? high : low);
+    return mid;
 }
 
 
@@ -472,7 +591,8 @@ define sq_loop(repeat, x)
     len = sizeof((*x)[0]) / baseb_bytes;
     for (i=1; i < 4; ++i) {
 	if ((sizeof((*x)[i]) / baseb_bytes) != len) {
-	    quit "sq_loop: 2nd arg matrix elements are not of equal BASEB-bit word length";
+	    quit "sq_loop: 2nd arg matrix elements are not of equal "
+	         "BASEB-bit word length";
 	}
     }
 
@@ -559,6 +679,7 @@ define sq_ratio(len)
     local tover;		/* est of time for loop overhead */
     local alg1_rate;		/* loop rate of 1st algorithm */
     local alg2_rate;		/* loop rate of 2nd algorithm */
+    local ret;			/* return ratio, or 1.0 */
     local i;
 
     /*
@@ -611,12 +732,12 @@ define sq_ratio(len)
      * determine the 1st algorithm rate
      */
     loops = max(1, ceil(loops * test_time / tlen));
-    if (loops < 8) {
+    if (loops < 16) {
 	if (config("user_debug") > 1) {
-	    printf("    we must expand loop test time to more than %d secs\n",
-		ceil(test_time * (8 / loops)));
+	    printf("    we must expand alg1 loop test time to about %d secs\n",
+		ceil(test_time * (16 / loops)));
 	}
-	loops = 8;
+	loops = 16;
     }
     tlen = sq_loop(loops, &x);
     if (config("user_debug") > 3) {
@@ -654,12 +775,12 @@ define sq_ratio(len)
      * determine the 2nd algorithm rate
      */
     loops = max(1, ceil(loops * test_time / tlen));
-    if (loops < 8) {
+    if (loops < 16) {
 	if (config("user_debug") > 1) {
-	    printf("    we must expand loop test time to more than %d secs\n",
-		ceil(test_time * (8 / loops)));
+	    printf("    we must expand alg2 loop test time to about %d secs\n",
+		ceil(test_time * (16 / loops)));
 	}
-	loops = 8;
+	loops = 16;
     }
     tlen = sq_loop(loops, &x);
     if (config("user_debug") > 3) {
@@ -688,7 +809,12 @@ define sq_ratio(len)
     /*
      * return alg1 / alg2 rate ratio
      */
-    return (alg1_rate / alg2_rate);
+    ret = alg1_rate / alg2_rate;
+    if (config("user_debug") > 2) {
+	printf("\tprecise ratio is: %.f sq_ratio will return: %.3f\n",
+		alg1_rate / alg2_rate, ret);
+    }
+    return ret;
 }
 
 
@@ -710,30 +836,40 @@ define best_sq2()
 {
     local ratio;	/* previously calculated alg1/alg2 ratio */
     local low;		/* low loop value tested */
-    local high;		/* low loop value tested */
+    local high;		/* high loop value tested */
+    local mid;		/* between low and high */
+    local best_val;	/* value found with ratio closest to unity */
+    local best_ratio;	/* cloest ratio found to unity */
     local expand;	/* how fast to expand the length */
 
     /*
      * setup
      */
-    test_time = 15.0;
-    if (config("user_debug") > 0) {
-	printf("will start with loop test time of %d secs\n", test_time);
+    printf("WARNING: This tool may not be computing the correct best value\n");
+    test_time = 5.0;
+    printf("The best_sq2() function will take a LONG time to run!\n");
+    printf("It is important that best_sq2() run on an othwewise idle host!\n");
+    if (config("user_debug") <= 0) {
+	printf("To monitor progress, set user_debug to 2: "
+	       "config(\"user_debug\", 2)\n");
     }
+    printf("Starting with loop test time of %d secs\n", test_time);
 
     /*
      * firewall - must have a >1 ratio for the initial length
      */
-    high = 16;
+    high = 8;
+    best_val = high;
     if (config("user_debug") > 0) {
 	printf("testing square alg1/alg2 ratio for len = %d\n", high);
     }
     ratio = sq_ratio(high);
+    best_ratio = ratio;
     if (config("user_debug") > 1) {
 	printf("    square alg1/alg2 ratio = %.3f\n", ratio);
     }
-    if (ratio <= 1.0) {
-	quit "best_sq2: tests imply config(\"sq2\") should be < 4";
+    if (ratio < 1.0) {
+	quit "best_sq2: test implies sq2 < 16, which seems bogus";
     }
 
     /*
@@ -746,31 +882,75 @@ define best_sq2()
 	 * We will multiplicatively expand our test level until
 	 * the ratio drops below 1.0.
 	 */
-	expand = ((ratio >= 3.5) ? 16 : 2^round(ratio));
+	expand = 2;
 	low = high;
 	high *= expand;
 	if (config("user_debug") > 1) {
 	    printf("    expand the next test range by a factor of %d\n",
-	           expand);
+	    	   expand);
 	}
 
 	/*
 	 * determine the alg1/alg2 test ratio for this new length
 	 */
 	if (high >= 2^31) {
-	    quit "best_sq2: tests imply config(\"sq2\") should be >= 2^31";
+	    quit "best_sq2: tests imply sq2 >= 2^31, which seems bogus";
 	}
 	if (config("user_debug") > 0) {
 	    printf("testing square alg1/alg2 ratio for len = %d\n", high);
 	}
 	ratio = sq_ratio(high);
+	if (abs(ratio - 1.0) < abs(best_ratio - 1.0)) {
+	    best_val = high;
+	    best_ratio = ratio;
+	    if (config("user_debug") > 1) {
+		printf("    len %d has a new cloest ratio to unity: %.6f\n",
+		       best_val, best_ratio);
+	    }
+	}
 	if (config("user_debug") > 1) {
 	    printf("    square alg1/alg2 ratio = %.3f\n", ratio);
 	}
-    } while (ratio >= 1.0);
+    } while (ratio > 1.0);
+
+    /*
+     * If we previously expanded more than by a factor of 2, then
+     * we may have jumped over the crossover point.  So now
+     * drop down powers of two until the ratio is again >= 1.0
+     */
+    if (expand > 2) {
+	do {
+
+	    /*
+	     * contract by 2
+	     */
+	    high /= 2;
+	    low = high / 2;
+	    if (config("user_debug") > 0) {
+		printf("retesting multiply alg1/alg2 ratio for len = %d\n",
+		       high);
+	    }
+	    ratio = mul_ratio(high);
+	    if (abs(ratio - 1.0) < abs(best_ratio - 1.0)) {
+		best_val = high;
+		best_ratio = ratio;
+		if (config("user_debug") > 1) {
+		    printf("    len %d has a new cloest ratio to unity: %.6f\n",
+			   best_val, best_ratio);
+		}
+	    }
+	    if (config("user_debug") > 1) {
+		printf("    multiply alg1/alg2 ratio = %.6f\n", ratio);
+	    }
+
+	} while (ratio <= 1.0);
+
+	/* now that the ratio flipped again, use the previous range */
+	low = high;
+	high = high * 2;
+    }
     if (config("user_debug") > 0) {
-	printf("alg1/alg2 ratio now < 1.0, starting binary search between %d and %d\n",
-	    low, high);
+	printf("Starting binary search between %d and %d\n", low, high);
     }
 
     /*
@@ -779,41 +959,71 @@ define best_sq2()
     while (low+1 < high) {
 
     	/* try the mid-point */
+	mid = int((low+high)/2);
 	if (config("user_debug") > 0) {
-	    printf("testing square alg1/alg2 ratio for len = %d\n",
-	    	int((low+high)/2));
+	    printf("testing square alg1/alg2 ratio for len = %d\n", mid);
+	}
+	ratio = sq_ratio(mid);
+	if (abs(ratio - 1.0) < abs(best_ratio - 1.0)) {
+	    best_val = mid;
+	    best_ratio = ratio;
+	    if (config("user_debug") > 1) {
+		printf("    len %d has a new cloest ratio to unity: %.6f\n",
+		       best_val, best_ratio);
+	    }
 	}
-	ratio = sq_ratio(int((low+high)/2));
 	if (config("user_debug") > 1) {
-	    printf("    square alg1/alg2 ratio = %.3f\n", ratio);
+	    printf("    len %d square alg1/alg2 ratio = %.6f\n", mid, ratio);
+	}
+
+	/* stop search if near unity */
+	if (close_to_one(ratio)) {
+	    low = mid;
+	    high = mid;
+	    if (config("user_debug") > 0) {
+	    	printf("\twe are close enough to unity ratio at: %d\n", mid);
+	    }
+	    break;
 	}
 
 	/* bump lower range up if we went over */
-	if (ratio >= 1.0) {
+	if (ratio > 1.0) {
 	    if (config("user_debug") > 2) {
 	    	printf("\tmove low from %d up to %d\n",
-		    low, int((low+high)/2));
+		    low, mid);
 	    }
-	    low = int((low+high)/2);
+	    low = mid;
 
 	/* drop higher range down if we went under */
 	} else {
 	    if (config("user_debug") > 2) {
 	    	printf("\tmove high from %d down to %d\n",
-		     high, int((low+high)/2));
+		     high, mid);
 	    }
-	    high = int((low+high)/2);
+	    high = mid;
+	}
+
+	/* report on test loop progress */
+	if (config("user_debug") > 1) {
+	    printf("\tsetting low: %d high: %d diff: %d\n",
+	    	   low, high, high-low);
 	}
     }
 
     /*
      * return on the suggested config("sq2") value
      */
+    mid = int((low+high)/2);
     if (config("user_debug") > 0) {
-	printf("best value of config(\"sq2\") is %d\n",
-	    (ratio >= 1.0) ? high : low);
+	printf("Best value for square is near %d\n", best_val);
+	printf("Best square alg1/alg2 ratio is: %.6f\n", best_ratio);
+	printf("We suggest placing this line in your .calcrc:\n");
+	printf("config(\"sq2\", %d),;\n", best_val);
+	printf("WARNING: It is believed that the output "
+	       "of this resource file is bogus!\n");
+	printf("WARNING: You may NOT wish to follow the above suggeston.\n");
     }
-    return ((ratio >= 1.0) ? high : low);
+    return mid;
 }
 
 
@@ -866,7 +1076,8 @@ define pow_loop(repeat, x, ex)
     len = sizeof((*x)[0]) / baseb_bytes;
     for (i=1; i < 4; ++i) {
 	if ((sizeof((*x)[i]) / baseb_bytes) != len) {
-	    quit "pow_loop: 2nd arg matrix elements are not of equal BASEB-bit word length";
+	    quit "pow_loop: 2nd arg matrix elements are not of "
+	         "equal BASEB-bit word length";
 	}
     }
     if (!isint(ex) || ex < 3) {
@@ -963,6 +1174,7 @@ define pow_ratio(len)
     local alg1_rate;		/* loop rate of 1st algorithm */
     local alg2_rate;		/* loop rate of 2nd algorithm */
     local ex;			/* exponent to use in pow_loop() */
+    local ret;			/* return ratio, or 1.0 */
     local i;
 
     /*
@@ -985,7 +1197,7 @@ define pow_ratio(len)
     /*
      * setup
      */
-    ex = 5;
+    ex = 7;
 
     /*
      * initialize x, the values we will pmod
@@ -1021,12 +1233,12 @@ define pow_ratio(len)
      * determine the 1st algorithm rate
      */
     loops = max(1, ceil(loops * test_time / tlen));
-    if (loops < 8) {
+    if (loops < 16) {
 	if (config("user_debug") > 1) {
-	    printf("    we must expand loop test time to more than %d secs\n",
-		ceil(test_time * (8 / loops)));
+	    printf("    we must expand alg1 loop test time to about %d secs\n",
+		ceil(test_time * (16 / loops)));
 	}
-	loops = 8;
+	loops = 16;
     }
     tlen = pow_loop(loops, &x, ex);
     if (config("user_debug") > 3) {
@@ -1065,12 +1277,12 @@ define pow_ratio(len)
      * determine the 2nd algorithm rate
      */
     loops = max(1, ceil(loops * test_time / tlen));
-    if (loops < 8) {
+    if (loops < 16) {
 	if (config("user_debug") > 1) {
-	    printf("    we must expand loop test time to more than %d secs\n",
-		ceil(test_time * (8 / loops)));
+	    printf("    we must expand alg2 loop test time to about %d secs\n",
+		ceil(test_time * (16 / loops)));
 	}
-	loops = 8;
+	loops = 16;
     }
     tlen = pow_loop(loops, &x, ex);
     if (config("user_debug") > 3) {
@@ -1099,7 +1311,12 @@ define pow_ratio(len)
     /*
      * return alg1 / alg2 rate ratio
      */
-    return (alg1_rate / alg2_rate);
+    ret = alg1_rate / alg2_rate;
+    if (config("user_debug") > 2) {
+	printf("\tprecise ratio is: %.f pow_ratio will return: %.3f\n",
+		alg1_rate / alg2_rate, ret);
+    }
+    return ret;
 }
 
 
@@ -1121,17 +1338,25 @@ define best_pow2()
 {
     local ratio;	/* previously calculated alg1/alg2 ratio */
     local low;		/* low loop value tested */
-    local high;		/* low loop value tested */
+    local high;		/* high loop value tested */
+    local mid;		/* between low and high */
+    local best_val;	/* value found with ratio closest to unity */
+    local best_ratio;	/* cloest ratio found to unity */
     local expand;	/* how fast to expand the length */
     local looped;	/* 1 ==> we have expanded lengths before */
 
     /*
      * setup
      */
-    test_time = 15.0;
-    if (config("user_debug") > 0) {
-	printf("will start with loop test time of %d secs\n", test_time);
+    printf("WARNING: This tool may not be computing the correct best value\n");
+    test_time = 60.0;
+    printf("The best_pow2() function will take a LONG time to run!\n");
+    printf("It is important that best_pow2() run on an othwewise idle host!\n");
+    if (config("user_debug") <= 0) {
+	printf("To monitor progress, set user_debug to 2: "
+	       "config(\"user_debug\", 2)\n");
     }
+    printf("Starting with loop test time of %d secs\n", test_time);
 
     /*
      * firewall - must have a >1.02 ratio for the initial length
@@ -1142,16 +1367,27 @@ define best_pow2()
      */
     low = 4;
     high = 4;
+    best_val = high;
+    best_ratio = 1e10;	/* not a real value */
     do {
 	high *= 4;
 	if (config("user_debug") > 0) {
 	    printf("testing pmod alg1/alg2 ratio for len = %d\n", high);
 	}
 	ratio = pow_ratio(high);
+	if (abs(ratio - 1.0) < abs(best_ratio - 1.0)) {
+	    best_val = high;
+	    best_ratio = ratio;
+	    if (config("user_debug") > 1) {
+		printf("    len %d has a new cloest ratio to unity: %.6f\n",
+		       best_val, best_ratio);
+	    }
+	}
 	if (config("user_debug") > 1) {
 	    printf("    pmod alg1/alg2 ratio = %.3f\n", ratio);
 	    if (ratio > 1.0 && ratio <= 1.02) {
-	    printf("    while alg1 is slightly better than alg2, it is not clearly better\n");
+	    printf("    while alg1 is slightly better than alg2, "
+	    	   "it is not clearly better\n");
 	    }
 	}
     } while (ratio <= 1.02);
@@ -1193,20 +1429,27 @@ define best_pow2()
 	 * determine the alg1/alg2 test ratio for this new length
 	 */
 	if (high >= 2^31) {
-	    quit "best_pow2: tests imply config(\"pow2\") should be >= 2^31";
+	    quit "best_pow2: test implies pow2 >= 2^31, which seems bogus";
 	}
 	if (config("user_debug") > 0) {
 	    printf("testing pmod alg1/alg2 ratio for len = %d\n", high);
 	}
 	ratio = pow_ratio(high);
+	if (abs(ratio - 1.0) < abs(best_ratio - 1.0)) {
+	    best_val = high;
+	    best_ratio = ratio;
+	    if (config("user_debug") > 1) {
+		printf("    len %d has a new cloest ratio to unity: %.6f\n",
+		       best_val, best_ratio);
+	    }
+	}
 	if (config("user_debug") > 1) {
-	    printf("    pmod alg1/alg2 ratio = %.3f\n", ratio);
+	    printf("    pmod alg1/alg2 ratio = %.6f\n", ratio);
 	}
 	looped = 1;
-    } while (ratio >= 1.0);
+    } while (ratio > 1.0);
     if (config("user_debug") > 0) {
-	printf("alg1/alg2 ratio now < 1.0, starting binary search between %d and %d\n",
-	    low, high);
+	printf("Starting binary search between %d and %d\n", low, high);
     }
 
     /*
@@ -1215,39 +1458,69 @@ define best_pow2()
     while (low+1 < high) {
 
     	/* try the mid-point */
+	mid = int((low+high)/2);
 	if (config("user_debug") > 0) {
-	    printf("testing pmod alg1/alg2 ratio for len = %d\n",
-	    	int((low+high)/2));
+	    printf("testing pow2 alg1/alg2 ratio for len = %d\n", mid);
+	}
+	ratio = pow_ratio(mid);
+	if (abs(ratio - 1.0) < abs(best_ratio - 1.0)) {
+	    best_val = mid;
+	    best_ratio = ratio;
+	    if (config("user_debug") > 1) {
+		printf("    len %d has a new cloest ratio to unity: %.6f\n",
+		       best_val, best_ratio);
+	    }
 	}
-	ratio = pow_ratio(int((low+high)/2));
 	if (config("user_debug") > 1) {
-	    printf("    pmod alg1/alg2 ratio = %.3f\n", ratio);
+	    printf("    len %d pmod alg1/alg2 ratio = %.6f\n", mid, ratio);
+	}
+
+	/* stop search if near unity */
+	if (close_to_one(ratio)) {
+	    low = mid;
+	    high = mid;
+	    if (config("user_debug") > 0) {
+	    	printf("\twe are close enough to unity ratio at: %d\n", mid);
+	    }
+	    break;
 	}
 
 	/* bump lower range up if we went over */
-	if (ratio >= 1.0) {
+	if (ratio > 1.0) {
 	    if (config("user_debug") > 2) {
 	    	printf("\tmove low from %d up to %d\n",
-		    low, int((low+high)/2));
+		    low, mid);
 	    }
-	    low = int((low+high)/2);
+	    low = mid;
 
 	/* drop higher range down if we went under */
 	} else {
 	    if (config("user_debug") > 2) {
 	    	printf("\tmove high from %d down to %d\n",
-		     high, int((low+high)/2));
+		     high, mid);
 	    }
-	    high = int((low+high)/2);
+	    high = mid;
+	}
+
+	/* report on test loop progress */
+	if (config("user_debug") > 1) {
+	    printf("\tsetting low: %d high: %d diff: %d\n",
+	    	   low, high, high-low);
 	}
     }
 
     /*
      * return on the suggested config("pow2") value
      */
+    mid = int((low+high)/2);
     if (config("user_debug") > 0) {
-	printf("best value of config(\"pow2\") is %d\n",
-	    (ratio >= 1.0) ? high : low);
+	printf("Best value for pmod is near %d\n", best_val);
+	printf("Best pmod alg1/alg2 ratio is: %.6f\n", best_ratio);
+	printf("We suggest placing this line in your .calcrc:\n");
+	printf("config(\"pow2\", %d),;\n", best_val);
+	printf("WARNING: It is believed that the output "
+	       "of this resource file is bogus!\n");
+	printf("WARNING: You may NOT wish to follow the above suggeston.\n");
     }
-    return ((ratio >= 1.0) ? high : low);
+    return mid;
 }

cal/bernpoly.cal

@@ -0,0 +1,59 @@
+/*
+ * bernpoly - Bernoully polynomials B_n(z) for arbitrary n,z..
+ *
+ * Copyright (C) 2013 Christoph Zurnieden
+ *
+ * Calc is open software; you can redistribute it and/or modify it under
+ * the terms of the version 2.1 of the GNU Lesser General Public License
+ * as published by the Free Software Foundation.
+ *
+ * Calc is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+ * or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU Lesser General
+ * Public License for more details.
+ *
+ * A copy of version 2.1 of the GNU Lesser General Public License is
+ * distributed with calc under the filename COPYING-LGPL.  You should have
+ * received a copy with calc; if not, write to Free Software Foundation, Inc.
+ * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
+ *
+ * @(#) $Revision: 30.4 $
+ * @(#) $Id: bernpoly.cal,v 30.4 2013/08/18 20:01:53 chongo Exp $
+ * @(#) $Source: /usr/local/src/bin/calc/cal/RCS/bernpoly.cal,v $
+ *
+ * Under source code control:	2013/08/11 01:31:28
+ * File existed as early as:	2013
+ */
+
+
+static resource_debug_level;
+resource_debug_level = config("resource_debug", 0);
+
+
+read -once zeta2
+
+
+/* Idea by Don Zagier */
+define bernpoly(n,z){
+  local h s c k;
+  if(isint(n) && n>=0){
+    h=0;s=0;c=-1;
+    for(k=1;k<=n+1;k++){
+      c*=1-(n+2)/k;
+      s+=z^n;
+      z++;
+      h+=c*s/k;
+    }
+    return h;
+  }
+  else return -n*hurwitzzeta(1-n,z);
+}
+
+
+/*
+ * restore internal function from resource debugging
+ */
+config("resource_debug", resource_debug_level),;
+if (config("resource_debug") & 3) {
+    print "bernpoly(n,z)";
+}

cal/brentsolve.cal

@@ -0,0 +1,258 @@
+/*
+ * brentsolve - Root finding with the Brent-Dekker trick
+ *
+ * Copyright (C) 2013 Christoph Zurnieden
+ *
+ * Calc is open software; you can redistribute it and/or modify it under
+ * the terms of the version 2.1 of the GNU Lesser General Public License
+ * as published by the Free Software Foundation.
+ *
+ * Calc is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+ * or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU Lesser General
+ * Public License for more details.
+ *
+ * A copy of version 2.1 of the GNU Lesser General Public License is
+ * distributed with calc under the filename COPYING-LGPL.  You should have
+ * received a copy with calc; if not, write to Free Software Foundation, Inc.
+ * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
+ *
+ * @(#) $Revision: 30.4 $
+ * @(#) $Id: brentsolve.cal,v 30.4 2013/08/18 20:01:53 chongo Exp $
+ * @(#) $Source: /usr/local/src/bin/calc/cal/RCS/brentsolve.cal,v $
+ *
+ * Under source code control:	2013/08/11 01:31:28
+ * File existed as early as:	2013
+ */
+
+
+static resource_debug_level;
+resource_debug_level = config("resource_debug", 0);
+
+
+/*
+  A short explanation is at http://en.wikipedia.org/wiki/Brent%27s_method
+  I tried to follow the description at wikipedia as much as possible to make
+  the slight changes I did more visible.
+  You may give http://people.sc.fsu.edu/~jburkardt/cpp_src/brent/brent.html a
+  short glimpse (Brent's originl Fortran77 versions and some translations of
+  it).
+*/
+
+static true = 1;
+static false = 0;
+define brentsolve(low, high,eps){
+  local a b c d fa fb fc fa2 fb2 fc2 s fs  tmp tmp2  mflag i places;
+  a = low;
+  b = high;
+  c = 0;
+
+  if(isnull(eps))
+    eps = epsilon(epsilon()*1e-3);
+  places = highbit(1 + int( 1/epsilon() ) ) + 1;
+
+  d = 1/eps;
+
+  fa = f(a);
+  fb = f(b);
+
+  fc = 0;
+  s = 0;
+  fs = 0;
+
+  if(fa * fb >= 0){
+    if(fa < fb){
+      epsilon(eps);
+      return a;
+    }
+    else{
+      epsilon(eps);
+      return b;
+    }
+  }
+
+  if(abs(fa) < abs(fb)){
+    tmp = a; a = b; b = tmp;
+    tmp = fa; fa = fb; fb = tmp;
+  }
+
+  c = a;
+  fc = fa;
+  mflag = 1;
+  i = 0;
+
+  while(!(fb==0) && (abs(a-b) > eps)){
+    if((fa != fc) && (fb != fc)){
+      /* Inverse quadratic interpolation*/
+      fc2 = fc^2;
+      fa2 = fa^2;
+      s = bround(((fb^2*((fc*a)-(c*fa)))+(fb*((c*fa2)-(fc2*a)))+(b*((fc2*fa)
+                    -(fc*fa2))))/((fc - fb)*(fa - fb)*(fc - fa)),places++);
+    }
+    else{
+      /* Secant Rule*/
+      s =bround( b - fb * (b - a) / (fb - fa),places++);
+    }
+    tmp2 = (3 * a + b) / 4;
+    if( (!( ((s > tmp2) && (s < b))||((s < tmp2) && (s > b))))
+        || (mflag && (abs(s - b) >= (abs(b - c) / 2)))
+        || (!mflag && (abs(s - b) >= (abs(c - d) / 2)))) {
+      s = (a + b) / 2;
+      mflag = true;
+    }
+    else{
+      if( (mflag && (abs(b - c) < eps))
+          || (!mflag && (abs(c - d) < eps))) {
+        s = (a + b) / 2;
+        mflag = true;
+      }
+      else
+        mflag = false;
+    }
+    fs = f(s);
+    c = b;
+    fc = fb;
+    if (fa * fs < 0){
+      b = s;
+      fb = fs;
+    }
+    else {
+      a = s;
+      fa = fs;
+    }
+
+    if (abs(fa) < abs(fb)){
+      tmp = a; a = b; b = tmp;
+      tmp = fa; fa = fb; fb = tmp;
+    }
+    i++;
+    if (i > 1000){
+      epsilon(eps);
+      return newerror("brentsolve: does not converge");
+    }
+  }
+  epsilon(eps);
+  return b;
+}
+
+/*
+  A variation of the solver to accept functions named differently from "f". The
+  code should explain it.
+*/
+define brentsolve2(low, high,which,eps){
+  local a b c d fa fb fc fa2 fb2 fc2 s fs  tmp tmp2  mflag i places;
+  a = low;
+  b = high;
+  c = 0;
+
+  switch(param(0)){
+    case 0:
+    case 1: return newerror("brentsolve2: not enough argments");
+    case 2: eps = epsilon(epsilon()*1e-2);
+            which = 0;break;
+    case 3: eps = epsilon(epsilon()*1e-2);break;
+    default: break;
+  };
+  places = highbit(1 + int(1/epsilon())) + 1;
+
+  d = 1/eps;
+
+  switch(which){
+    case 1:  fa = __CZ__invbeta(a);
+             fb = __CZ__invbeta(b); break;
+    case 2:  fa = __CZ__invincgamma(a);
+             fb = __CZ__invincgamma(b); break;
+
+    default: fa = f(a);fb = f(b);   break;
+  };
+
+  fc = 0;
+  s = 0;
+  fs = 0;
+
+  if(fa * fb >= 0){
+    if(fa < fb)
+      return a;
+    else
+      return b;
+  }
+
+  if(abs(fa) < abs(fb)){
+    tmp = a; a = b; b = tmp;
+    tmp = fa; fa = fb; fb = tmp;
+  }
+
+  c = a;
+  fc = fa;
+  mflag = 1;
+  i = 0;
+
+  while(!(fb==0) && (abs(a-b) > eps)){
+
+    if((fa != fc) && (fb != fc)){
+      /* Inverse quadratic interpolation*/
+      fc2 = fc^2;
+      fa2 = fa^2;
+      s = bround(((fb^2*((fc*a)-(c*fa)))+(fb*((c*fa2)-(fc2*a)))+(b*((fc2*fa)
+                    -(fc*fa2))))/((fc - fb)*(fa - fb)*(fc - fa)),places);
+      places++;
+    }
+    else{
+      /* Secant Rule*/
+      s =bround( b - fb * (b - a) / (fb - fa),places);
+      places++;
+    }
+    tmp2 = (3 * a + b) / 4;
+    if( (!( ((s > tmp2) && (s < b))||((s < tmp2) && (s > b))))
+        || (mflag && (abs(s - b) >= (abs(b - c) / 2)))
+        || (!mflag && (abs(s - b) >= (abs(c - d) / 2)))) {
+      s = (a + b) / 2;
+      mflag = true;
+    }
+    else{
+      if( (mflag && (abs(b - c) < eps))
+          || (!mflag && (abs(c - d) < eps))) {
+        s = (a + b) / 2;
+        mflag = true;
+      }
+      else
+        mflag = false;
+    }
+    switch(which){
+      case 1:  fs = __CZ__invbeta(s); break;
+      case 2:  fs = __CZ__invincgamma(s); break;
+
+      default: fs = f(s);             break;
+    };
+    c = b;
+    fc = fb;
+    if (fa * fs < 0){
+      b = s;
+      fb = fs;
+    }
+    else {
+      a = s;
+      fa = fs;
+    }
+
+    if (abs(fa) < abs(fb)){
+      tmp = a; a = b; b = tmp;
+      tmp = fa; fa = fb; fb = tmp;
+    }
+    i++;
+    if (i > 1000){
+      return newerror("brentsolve2: does not converge");
+    }
+  }
+  return b;
+}
+
+
+/*
+ * restore internal function from resource debugging
+ */
+config("resource_debug", resource_debug_level),;
+if (config("resource_debug") & 3) {
+    print "brentsolve(low, high,eps)";
+    print "brentsolve2(low, high,which,eps)";
+}

cal/constants.cal

@@ -0,0 +1,104 @@
+/*
+ * constants - implementation of different constants to arbitrary precision
+ *
+ * Copyright (C) 2013 Christoph Zurnieden
+ *
+ * Calc is open software; you can redistribute it and/or modify it under
+ * the terms of the version 2.1 of the GNU Lesser General Public License
+ * as published by the Free Software Foundation.
+ *
+ * Calc is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+ * or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU Lesser General
+ * Public License for more details.
+ *
+ * A copy of version 2.1 of the GNU Lesser General Public License is
+ * distributed with calc under the filename COPYING-LGPL.  You should have
+ * received a copy with calc; if not, write to Free Software Foundation, Inc.
+ * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
+ *
+ * @(#) $Revision: 30.4 $
+ * @(#) $Id: constants.cal,v 30.4 2013/08/18 20:01:53 chongo Exp $
+ * @(#) $Source: /usr/local/src/bin/calc/cal/RCS/constants.cal,v $
+ *
+ * Under source code control:	2013/08/11 01:31:28
+ * File existed as early as:	2013
+ */
+
+
+static resource_debug_level;
+resource_debug_level = config("resource_debug", 0);
+
+
+static __CZ__euler_mascheroni = 0;
+static __CZ__euler_mascheroni_prec = 0;
+
+
+define e(){
+  local k temp1 temp2 ret eps factor upperlimit prec;
+
+  prec = digits(1/epsilon());
+  if(__CZ__euler_mascheroni != 0 &&  __CZ__euler_mascheroni_prec >= prec)
+    return __CZ__euler_mascheroni;
+  if(prec<=20) return 2.718281828459045235360287471;
+  if(prec<=1800){
+    __CZ__euler_mascheroni = exp(1);
+    __CZ__euler_mascheroni_prec = prec;
+  }
+
+  eps=epsilon(1e-20);
+  factor = 1;
+  k = 0;
+  upperlimit = prec * ln(10);
+  while(k<upperlimit){
+    k += ln(factor);
+    factor++;
+  }
+  epsilon(eps);
+  temp1 = 0;
+  ret   = 1;
+  for(k=3;k<=factor;k++){
+    temp2 = temp1;
+    temp1 = ret;
+    ret =  (k-1) *(temp1 + temp2);
+  }
+
+  ret = inverse( ret * inverse(factorial(factor) ) ) ;
+  __CZ__euler_mascheroni = ret;
+  __CZ__euler_mascheroni_prec = prec;
+  return ret;
+}
+
+
+/* Lupas' series */
+static __CZ__catalan = 0;
+static __CZ__catalan_prec = 0;
+define G(){
+  local eps a s t n;
+  eps = epsilon(epsilon()*1e-10);
+  if(__CZ__catalan != 0 &&  __CZ__catalan >= log(1/eps))
+    return __CZ__catalan;
+  a = 1;
+  s = 0;
+  t = 1;
+  n = 1;
+  while(abs(t)> eps){
+    a *= 32 * n^3 * (2*n-1);
+    a /=((3-16*n+16*n^2)^2);
+    t = a * (-1)^(n-1) * (40*n^2-24*n+3) / (n^3 * (2*n-1));
+    s += t;
+    n += 1;
+  }
+  s = s/64;
+  __CZ__catalan = s;
+  __CZ__catalan_prec = log(1/eps);
+  epsilon(eps);
+  return s;
+}
+
+
+config("resource_debug", resource_debug_level),;
+if (config("resource_debug") & 3) {
+    print "e()";
+    print "G()";
+}

cal/factorial.cal

@@ -0,0 +1,204 @@
+/*
+ * factorial - implementation of different algorithms for the factorial
+ *
+ * Copyright (C) 2013 Christoph Zurnieden
+ *
+ * Calc is open software; you can redistribute it and/or modify it under
+ * the terms of the version 2.1 of the GNU Lesser General Public License
+ * as published by the Free Software Foundation.
+ *
+ * Calc is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+ * or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU Lesser General
+ * Public License for more details.
+ *
+ * A copy of version 2.1 of the GNU Lesser General Public License is
+ * distributed with calc under the filename COPYING-LGPL.  You should have
+ * received a copy with calc; if not, write to Free Software Foundation, Inc.
+ * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
+ *
+ * @(#) $Revision: 30.4 $
+ * @(#) $Id: factorial.cal,v 30.4 2013/08/18 20:01:53 chongo Exp $
+ * @(#) $Source: /usr/local/src/bin/calc/cal/RCS/factorial.cal,v $
+ *
+ * Under source code control:	2013/08/11 01:31:28
+ * File existed as early as:	2013
+ */
+
+
+/*
+ * hide internal function from resource debugging
+ */
+static resource_debug_level;
+resource_debug_level = config("resource_debug", 0);
+
+
+/*
+  get dependencies
+*/
+read -once toomcook;
+
+
+/* A simple list to keep things...uhm...simple?*/
+static __CZ__primelist = list();
+
+/* Helper for primorial: fill list with primes in range a,b */
+define __CZ__fill_prime_list(a,b)
+{
+  local k;
+  k=a;
+  if(isprime(k))k--;
+  while(1){
+    k = nextprime(k);
+    if(k > b) break;
+    append(__CZ__primelist,k );
+  }
+}
+
+/* Helper for factorial: how often prime p divides the factorial of n */
+define __CZ__prime_divisors(n,p)
+{
+    local q,m;
+    q = n;
+    m = 0;
+    if (p > n) return 0;
+    if (p > n/2) return 1;
+    while (q >= p) {
+        q  = q//p;
+        m += q;
+    }
+    return m;
+}
+
+/*
+  Wrapper. Please set cut-offs to own taste and hardware.
+*/
+define factorial(n){
+  local prime result shift prime_list k k1 k2 expo_list pix cut primorial;
+
+  result = 1;
+  prime  = 2;
+
+  if(!isint(n))  {
+    return newerror("factorial(n): n is not an integer"); ## or gamma(n)?
+  }
+  if(n < 0)      return newerror("factorial(n): n < 0");
+  if(n < 9000 && !isdefined("test8900"))   {
+    ## builtin is implemented with splitting but only with
+    ## Toom-Cook 2 (by Karatsuba (the father))
+    return n!;
+  }
+
+  shift = __CZ__prime_divisors(n,prime);
+  prime = 3;
+  cut = n//2;
+  pix   = pix(cut);
+  prime_list = mat[pix];
+  expo_list  = mat[pix];
+
+  k = 0;
+/*
+  Peter Borwein's algorithm
+
+  @Article{journals/jal/Borwein85,
+  author = {Borwein, Peter B.},
+  title  = {On the Complexity of Calculating Factorials.},
+  journal = {J. Algorithms},
+  year   = {1985},
+  number = {3},
+  url    = {http://dblp.uni-trier.de/db/journals/jal/jal6.html#Borwein85}
+*/
+
+  do {
+    prime_list[k]  = prime;
+    expo_list[k++] = __CZ__prime_divisors(n,prime);
+    prime          = nextprime(prime);
+  }while(prime <= cut);
+
+  /* size of the largest exponent in bits  */
+  k1 = highbit(expo_list[0]);
+  k2 = size(prime_list)-1;
+
+  for(;k1>=0;k1--){
+    /*
+       the cut-off for T-C-4 ist still to low, using T-C-3 here
+       TODO: check cutoffs
+     */
+    result = toomcook3square(result);
+    /*
+      almost all time is spend in this loop, so cutting of the
+      upper half of the primes makes sense
+    */
+    for(k=0; k<=k2; k++) {
+      if((expo_list[k] & (1 << k1)) != 0) {
+        result *= prime_list[k];
+      }
+    }
+
+  }
+  primorial = primorial( cut, n);
+  result *= primorial;
+  result <<= shift;
+  return result;
+}
+
+/*
+  Helper for primorial: do the product with binary splitting
+  TODO: do it without the intermediate list
+*/
+define __CZ__primorial__lowlevel( a, b ,p)
+{
+  local  c;
+  if( b == a) return p ;
+  if( b-a > 1){
+    c= (b + a) >> 1;
+    return  __CZ__primorial__lowlevel( a   , c , __CZ__primelist[a] )
+         *  __CZ__primorial__lowlevel( c+1 , b , __CZ__primelist[b] ) ;
+  }
+  return  __CZ__primelist[a] * __CZ__primelist[b];
+}
+
+/*
+   Primorial, Product of consecutive primes in range a,b
+
+  Originally meant to do primorials with a start different from 2, but
+  found out that this is faster at about a=1,b>=10^5 than the builtin
+  function pfact().  With the moderately small list a=1,b=10^6 (78498
+  primes) it is 3 times faster.  A quick look-up showed what was already
+  guessed: pfact() does it linearly.  (BTW: what is the time complexity
+  of the primorial with the naive algorithm?)
+*/
+define primorial(a,b)
+{
+  local C1 C2;
+  if(!isint(a)) return newerror("primorial(a,b): a is not an integer");
+  else if(!isint(b)) return newerror("primorial(a,b): b is not an integer");
+  else if(a < 0) return newerror("primorial(a,b): a < 0");
+  else if( b < 2 ) return newerror("primorial(a,b): b < 2");
+  else if( b < a) return newerror("primorial(a,b): b < a");
+  else{
+    /* last prime < 2^32 is also  max. prime for nextprime()*/
+    if(b >= 4294967291) return newerror("primorial(a,b): max. prime exceeded");
+    if(b ==  2) return 2;
+    /*
+      Can be extended by way of pfact(b)/pfact(floor(a-1/2)) for small a
+     */
+    if(a<=2 && b < 10^5) return pfact(b);
+    /* TODO: use pix() and a simple array (mat[])instead*/
+     __CZ__primelist = list();
+     __CZ__fill_prime_list(a,b);
+    C1 = size(__CZ__primelist)-1;
+    return __CZ__primorial__lowlevel( 0, C1,1)
+  }
+}
+
+
+/*
+ * restore internal function from resource debugging
+ * report important interface functions
+ */
+config("resource_debug", resource_debug_level),;
+if (config("resource_debug") & 3) {
+    print "factorial(n)";
+    print "primorial(a, b)";
+}

cal/factorial2.cal

@@ -0,0 +1,723 @@
+/*
+ * factorial2 - implementation of different factorial related functions
+ *
+ * Copyright (C) 2013 Christoph Zurnieden
+ *
+ * Calc is open software; you can redistribute it and/or modify it under
+ * the terms of the version 2.1 of the GNU Lesser General Public License
+ * as published by the Free Software Foundation.
+ *
+ * Calc is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+ * or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU Lesser General
+ * Public License for more details.
+ *
+ * A copy of version 2.1 of the GNU Lesser General Public License is
+ * distributed with calc under the filename COPYING-LGPL.  You should have
+ * received a copy with calc; if not, write to Free Software Foundation, Inc.
+ * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
+ *
+ * @(#) $Revision: 30.4 $
+ * @(#) $Id: factorial2.cal,v 30.4 2013/08/18 20:01:53 chongo Exp $
+ * @(#) $Source: /usr/local/src/bin/calc/cal/RCS/factorial2.cal,v $
+ *
+ * Under source code control:	2013/08/11 01:31:28
+ * File existed as early as:	2013
+ */
+
+
+/*
+ * hide internal function from resource debugging
+ */
+static resource_debug_level;
+resource_debug_level = config("resource_debug", 0);
+
+
+/*
+  get dependencies
+*/
+read -once factorial toomcook specialfunctions;
+
+
+/*
+  Factorize a factorial and put the result in a 2-column matrix with pi(n) rows
+  mat[ primes , exponent ]
+  Result can be restricted to start at a prime different from 2 with the second
+  argument "start". That arguments gets taken at face value if it prime and
+  smaller than n, otherwise the next larger prime is taken if that prime is
+  smaller than n.
+*/
+
+define __CZ__factor_factorial(n,start){
+  local prime prime_list k pix stop;
+
+
+  if(!isint(n)) return
+    newerror("__CZ__factor_factorial(n,start): n is not integer");
+  if(n <  0)    return newerror("__CZ__factor_factorial(n,start): n < 0");
+  if(n == 1)    return newerror("__CZ__factor_factorial(n,start): n == 1");
+
+  if(start){
+    if(!isint(start) && start < 0 && start > n)
+      return newerror("__CZ__factor_factorial(n,start): value of "
+      		      "parameter 'start' out of range");
+    if(start == n && isprime(n)){
+      prime_list = mat[1 , 2];
+      prime_list[0,0] = n;
+      prime_list[0,1] = 1;
+    }
+    else if(!isprime(start) && nextprime(start) >n)
+      return newerror("__CZ__factor_factorial(n,start): value of parameter "
+      		      "'start' out of range");
+    else{
+      if(!isprime(start))  prime = nextprime(start);
+      else prime = start;
+    }
+  }
+  else
+    prime = 2;
+
+  pix   = pix(n);
+  if(start){
+    pix -= pix(prime) -1;
+  }
+  prime_list = mat[pix , 2];
+
+  k = 0;
+
+  do {
+    prime_list[k  ,0] = prime;
+    prime_list[k++,1] = __CZ__prime_divisors(n,prime);
+    prime          = nextprime(prime);
+  }while(prime <= n);
+
+  return prime_list;
+}
+
+/*
+
+  subtracts exponents of n_1! from exponents of n_2! with n_1<=n_2
+
+  Does not check for size or consecutiveness of the primes or a carry
+*/
+
+define __CZ__subtract_factored_factorials(matrix_2n,matrix_n){
+  local k ret len1,len2,tmp count p e;
+  len1 = size(matrix_n)/2;
+  len2 = size(matrix_2n)/2;
+  if(len2<len1){
+
+    swap(len1,len2);
+    tmp = matrix_n;
+    matrix_n = matrix_2n;
+    matrix_2n = tmp;
+  }
+  tmp = mat[len1,2];
+  k   = 0;
+
+  for(;k<len1;k++){
+     p = matrix_2n[k,0];
+     e = matrix_2n[k,1] - matrix_n[k,1];
+     if(e!=0){
+       tmp[count  ,0] = p;
+       tmp[count++,1] = e;
+     }
+  }
+  ret = mat[count + (len2-len1),2];
+  for(k=0;k<count;k++){
+     ret[k,0] = tmp[k,0];
+     ret[k,1] = tmp[k,1];
+  }
+
+  free(tmp);
+  for(k=len1;k<len2;k++){
+     ret[count,0] = matrix_2n[k,0];
+     ret[count++,1] = matrix_2n[k,1];
+  }
+  return ret;
+}
+
+/*
+
+  adds exponents of n_1! to exponents of n_2! with n_1<=n_2
+
+  Does not check for size or consecutiveness of the primes or a carry
+*/
+
+define __CZ__add_factored_factorials(matrix_2n,matrix_n){
+  local k ret len1,len2,tmp;
+  len1 = size(matrix_n)/2;
+  len2 = size(matrix_2n)/2;
+  if(len2<len1){
+    swap(len1,len2);
+    tmp = matrix_n;
+    matrix_n = matrix_2n;
+    matrix_2n = tmp;
+  }
+  ret = mat[len2,2];
+  k   = 0;
+  for(;k<len1;k++){
+     ret[k,0] = matrix_2n[k,0];
+     ret[k,1] = matrix_2n[k,1] + matrix_n[k,1];
+  }
+  for(;k<len2;k++){
+     ret[k,0] = matrix_2n[k,0];
+     ret[k,1] = matrix_2n[k,1];
+  }
+  return ret;
+}
+
+/*
+   Does not check if all exponents are positive
+
+
+                   timings
+                   this		  comb	    comb-this     rel. k/n
+;  benchmark_binomial(10,13)
+n=2^13 k=2^10 	 0.064004 	 0.016001	+ 	0.76923076923076923077
+n=2^13 k=2^11 	 0.064004 	 0.048003	+ 	0.84615384615384615385
+n=2^13 k=2^12 	 0.068004 	 0.124008	- 	0.92307692307692307692
+;  benchmark_binomial(10,15)
+n=2^15 k=2^10 	 0.216014 	 0.024001	+ 	0.66666666666666666667
+n=2^15 k=2^11 	 0.220014 	 0.064004	+ 	0.73333333333333333333
+n=2^15 k=2^12 	 0.228014 	 0.212014	+ 	0.8
+n=2^15 k=2^13 	 0.216013 	 0.664042	- 	0.86666666666666666667
+n=2^15 k=2^14 	 0.240015 	 1.868117	- 	0.93333333333333333333
+;  benchmark_binomial(11,15)
+n=2^15 k=2^11 	 0.216014 	 0.068004	+ 	0.73333333333333333333
+n=2^15 k=2^12 	 0.236015 	 0.212013	+ 	0.8
+n=2^15 k=2^13 	 0.216013 	 0.656041	- 	0.86666666666666666667
+n=2^15 k=2^14 	 0.244016 	 1.872117	- 	0.93333333333333333333
+;  benchmark_binomial(11,18)
+n=2^18 k=2^11 	 1.652103 	 0.100006	+ 	0.61111111111111111111
+n=2^18 k=2^12 	 1.608101 	 0.336021	+ 	0.66666666666666666667
+n=2^18 k=2^13 	 1.700106 	 1.140071	+ 	0.72222222222222222222
+n=2^18 k=2^14 	 1.756109 	 3.924245	- 	0.77777777777777777778
+n=2^18 k=2^15 	 2.036127 	 13.156822	- 	0.83333333333333333333
+n=2^18 k=2^16 	 2.172135 	 41.974624	- 	0.88888888888888888889
+n=2^18 k=2^17 	 2.528158 	 121.523594	- 	0.94444444444444444444
+;  benchmark_binomial(15,25)
+n=2^25 k=2^15 	 303.790985 	 38.266392	+ 	0.6
+;  benchmark_binomial(17,25)
+n=2^25 k=2^17 	 319.127944 	 529.025062	- 	0.68
+*/
+
+define benchmark_binomial(s,limit){
+  local ret k A B T1 T2 start end N K;
+    N = 2^(limit);
+  for(k=s;k<limit;k++){
+    K = 2^k;
+   start=usertime();A=binomial(N,K);end=usertime();
+   T1 = end-start;
+   start=usertime();B=comb(N,K);end=usertime();
+   T2 = end-start;
+   print "n=2^"limit,"k=2^"k,"	",T1,"	",T2,T1<T2?"-":"+","	"k/limit;
+   if(A!=B){
+     print "false";
+    break;
+   }
+  }
+}
+
+define __CZ__multiply_factored_factorial(matrix,stop){
+  local prime result shift prime_list k k1 k2 expo_list pix count start;
+  local hb flag;
+
+  result = 1;
+  shift  = 0;
+
+
+  if(!ismat(matrix))
+   return newerror("__CZ__multiply_factored_factorial(matrix): "
+   		   "argument matrix not a matrix ");
+  if(!matrix[0,0])
+    return
+      newerror("__CZ__multiply_factored_factorial(matrix): "
+      	       "matrix[0,0] is null/0");
+
+  if(!isnull(stop))
+    pix = stop;
+  else
+    pix   = size(matrix)/2-1;
+
+  if(matrix[0,0] == 2 && matrix[0,1] > 0){
+    shift = matrix[0,1];
+    if(pix-1 == 0)
+      return 2^matrix[0,1];
+  }
+
+  /*
+    This is a more general way to do the multiplication, so any optimization
+    must have been done by the caller.
+  */
+  k = 0;
+  /*
+     The size of the largest exponent in bits is calculated dynamically.
+     Can be done more elegantly and saves one run over the whole array if done
+     inside the main loop.
+   */
+  hb =0;
+  for(k=0;k<pix;k++){
+    k1=highbit(matrix[k,1]);
+    if(hb < k1)hb=k1;
+  }
+
+  k2 = pix;
+  start = 0;
+  if(shift) start++;
+
+  for(k1=hb;k1>=0;k1--){
+    /*
+       the cut-off for T-C-4 ist still too low, using T-C-3 here
+       TODO: check cutoffs
+     */
+    result = toomcook3square(result);
+
+    for(k=start; k<=k2; k++) {
+      if((matrix[k,1] & (1 << k1)) != 0) {
+        result *= matrix[k,0];
+      }
+    }
+  }
+
+  result <<= shift;
+  return result;
+}
+
+/*
+    Compute binomial coeficients n!/(k!(n-k)!)
+
+    One of the rare cases where a formula once meant to ease manual computation
+    is actually the (aymptotically) fastest way to do it (in July 2013) for
+    the extreme case binomial(2N,N) but for a high price, the memory
+    needed is pi(N)--theoretically.
+*/
+define binomial(n,k){
+  local ret factored_n factored_k factored_nk denom num quot K prime_list prime;
+  local pix diff;
+
+  if(!isint(n) || !isint(k))
+    return newerror("binomial(n,k): input is not integer");
+  if(n<0 || k<0)
+    return newerror("binomial(n,k): input is not >= 0"); ;
+  if(n<k )   return 0;
+  if(n==k)   return 1;
+  if(k==0)   return 1;
+  if(k==1)   return n;
+  if(n-k==1) return n;
+  /*
+      cut-off depends on real size of n,k and size of n/k
+      The current cut-off is to small for large n, e.g.:
+      for 2n=2^23, k=n-n/2 the quotient is q=2n/k=0.25. Empirical tests showed
+      that 2n=2^23 and k=2^16  with q=0.0078125 are still faster than the
+      builtin function.
+
+      The symmetry (n,k) = (n,n-k) is of not much advantage here. One way
+      might be to get closer to k=n/2 if k<n-k but only if the difference
+      is small and n very large.
+   */
+  if(n<2e4 && !isdefined("test8900"))  return comb(n,k);
+  if(n<2e4 && k< n-n/2 && !isdefined("test8900")) return comb(n,k);
+  /*
+    This should be done in parallel to save some memory, e.g. no temporary
+    arrays are needed, all can be done inline.
+    The theoretical memory needed is pi(k).
+    Which is still a lot.
+  */
+
+  prime = 2;
+  pix   = pix(n);
+  prime_list = mat[pix , 2];
+  K = 0;
+  do {
+    prime_list[K  ,0] = prime;
+    diff = __CZ__prime_divisors(n,prime)-
+	   ( __CZ__prime_divisors(n-k,prime)+__CZ__prime_divisors(k,prime));
+    if(diff != 0)
+      prime_list[K++,1] = diff;
+    prime          = nextprime(prime);
+  }while(prime <= k);
+
+  do {
+    prime_list[K  ,0] = prime;
+    diff = __CZ__prime_divisors(n,prime)-__CZ__prime_divisors(n-k,prime);
+    if(diff != 0)
+      prime_list[K++,1] = diff;
+    prime          = nextprime(prime);
+  }while(prime <= n-k);
+
+  do {
+    prime_list[K  ,0] = prime;
+    prime_list[K++,1] = __CZ__prime_divisors(n,prime);
+    prime          = nextprime(prime);
+  }while(prime <= n);
+  ##print K,pix(k),pix(n-k),pix(n);
+  ##factored_k  = __CZ__factor_factorial(k,1);
+  ##factored_nk = __CZ__factor_factorial(n-k,1);
+
+  ##denom = __CZ__add_factored_factorials(factored_k,factored_nk);
+    ##free(factored_k,factored_nk);
+  ##num  = __CZ__factor_factorial(n,1);
+  ##quot  = __CZ__subtract_factored_factorials( num , denom );
+    ##free(num,denom);
+
+  ret = __CZ__multiply_factored_factorial(`prime_list,K-1);
+
+  return ret;
+}
+
+/*
+    Compute large catalan numbers  C(n) = binomial(2n,n)/(n+1) with
+    cut-off: (n>5e4)
+    Needs a lot of memory.
+*/
+define bigcatalan(n){
+  if(!isint(n) )return newerror("bigcatalan(n): n is not integer");
+  if( n<0) return newerror("bigcatalan(n): n < 0");
+  if( n<5e4 && !isdefined("test8900") ) return catalan(n);
+  return binomial(2*n,n)/(n+1);
+}
+
+/*
+  df(-111) = -1/3472059605858239446587523014902616804783337112829102414124928
+  		7753332469144201839599609375
+
+ df(-3+1i) = 0.12532538977287649201-0.0502372106177184607i
+ df(2n + 1) = (2*n)!/(n!*2^n)
+*/
+define __CZ__double_factorial(n){
+  local n1 n2 diff prime pix K prime_list k;
+  prime = 3;
+  pix   = pix(2*n)+1;
+  prime_list = mat[pix , 2];
+  K = 0;
+  do {
+    prime_list[K  ,0] = prime;
+    diff = __CZ__prime_divisors(2*n,prime)-( __CZ__prime_divisors(n,prime));
+    if(diff != 0)
+      prime_list[K++,1] = diff;
+    prime          = nextprime(prime);
+  }while(prime <= n);
+  do {
+    prime_list[K  ,0] = prime;
+    prime_list[K++,1] = __CZ__prime_divisors(2*n,prime);
+    prime          = nextprime(prime);
+  }while(prime <= 2*n);
+  return __CZ__multiply_factored_factorial(prime_list,K);
+/*
+        n1=__CZ__factor_factorial(2*n,1);
+        n1[0,1] = n1[0,1]-n;
+        n2=__CZ__factor_factorial(n,1);
+        diff=__CZ__subtract_factored_factorials( n1 , n2 );
+        return __CZ__multiply_factored_factorial(diff);
+*/
+}
+
+##1, 1, 3, 15, 105, 945, 10395, 135135, 2027025, 34459425, 654729075,
+##13749310575, 316234143225, 7905853580625, 213458046676875,
+##6190283353629375, 191898783962510625, 6332659870762850625,
+##221643095476699771875, 8200794532637891559375
+
+## 1, 2, 8, 48, 384, 3840, 46080, 645120, 10321920, 185794560,
+##3715891200, 81749606400, 1961990553600, 51011754393600,
+##1428329123020800, 42849873690624000, 1371195958099968000,
+##46620662575398912000, 1678343852714360832000, 63777066403145711616000
+define doublefactorial(n){
+    local n1 n2 diff eps ret;
+    if(!isint(n) ){
+       /*
+           Probably one of the not-so-good ideas. See result of
+            http://www.wolframalpha.com/input/?i=doublefactorial%28a%2Bbi%29
+        */
+       eps=epsilon(epsilon()*1e-2);
+       ret =  2^(n/2-1/4 * cos(pi()* n)+1/4) * pi()^(1/4 *
+       	      cos(pi()* n)-1/4)* gamma(n/2+1);
+       epsilon(eps);
+       return ret;
+    }
+    if(n==2) return 2;
+    if(n==3) return 3;
+    switch(n){
+      case -1:
+      case 0 : return 1;break;
+      case 2 : return 2;break;
+      case 3 : return 3;break;
+      case 4 : return 8;break;
+      default: break;
+    }
+    if(isodd(n)){
+      /*
+        TODO: find reasonable cutoff
+        df(2n + 1) = (2*n)!/(n!*2^n)
+      */
+      if(n>0){
+         n = (n+1)//2;
+         return __CZ__double_factorial(n);
+      }
+      else{
+        if(n == -3 ) return -1;
+        n = ((-n)-1)/2;
+        return ((-1)^-n)/__CZ__double_factorial(n);
+       }
+    }
+    else{
+      /*
+         I'm undecided here. The formula for complex n is valid for the negative
+         integers, too.
+      */
+      n = n>>1;
+      if(n>0){
+        if(!isdefined("test8900"))
+          return factorial(n)<<n;
+        else
+          return n!<<n;
+      }
+      else
+        return newerror("doublefactorial(n): even(n) < 0");
+   }
+}
+
+/*
+    Algorithm 3.17,
+    Donald Kreher and Douglas Simpson,
+    Combinatorial Algorithms,
+    CRC Press, 1998, page 89.
+*/
+static __CZ__stirling1;
+static __CZ__stirling1_n = -1;
+static __CZ__stirling1_m = -1;
+
+define stirling1(n,m){
+  local i j k;
+  if(n<0)return newerror("stirling1(n,m): n <= 0");
+  if(m<0)return newerror("stirling1(n,m): m < 0");
+  if(n<m) return 0;
+  if(n==m) return 1;
+  if(m==0 || n==0) return 0;
+  /* We always use the list */
+  /*
+  if(m=1){
+    if(iseven(n)) return -factorial(n-1);
+    else return factorial(n-1);
+  }
+  if(m == n-1){
+    if(iseven(n)) return -binomial(n,2);
+    else return -binomial(n,2);
+  }
+  */
+  if(__CZ__stirling1_n >= n && __CZ__stirling1_m  >= m){
+    return __CZ__stirling1[n,m];
+  }
+  else{
+    __CZ__stirling1      = mat[n+1,m+1];
+    __CZ__stirling1[0,0] = 1;
+    for(i=1;i<=n;i++)
+      __CZ__stirling1[i,0] = 0;
+    for(i=1;i<=n;i++){
+      for(j=1;j<=m;j++){
+        if(j<=i){
+          __CZ__stirling1[i, j] =   __CZ__stirling1[i - 1, j - 1] - (i - 1)\
+                                  * __CZ__stirling1[i - 1, j];
+        }
+        else{
+         __CZ__stirling1[i, j] = 0;
+        }
+      }
+    }
+    __CZ__stirling1_n = n;
+    __CZ__stirling1_m = m;
+    return __CZ__stirling1[n,m];
+  }
+}
+
+define stirling2(n,m){
+  local k sum;
+  if(n<0)return newerror("stirling2(n,m): n < 0");
+  if(m<0)return newerror("stirling2(n,m): m < 0");
+  if(n<m) return 0;
+  if(n==0 && n!=m) return 0;
+  if(n==m) return 1;
+  if(m==0 )return 0;
+  if(m==1) return 1;
+  if(m==2) return 2^(n-1)-1;
+  /*
+    There are different methods to speed up alternating sums.
+    This one doesn't.
+   */
+  if(isdefined("test8900")){
+    for(k=0;k<=m;k++){
+      sum += (-1)^(m-k)*comb(m,k)*k^n;
+    }
+  return sum/(m!);
+  }
+  else{
+    for(k=0;k<=m;k++){
+      sum += (-1)^(m-k)*binomial(m,k)*k^n;
+    }
+  return sum/factorial(m);
+  }
+}
+
+static __CZ__stirling2;
+static __CZ__stirling2_n = -1;
+static __CZ__stirling2_m = -1;
+define stirling2caching(n,m){
+  local nm i j ;
+  if(n<0)return newerror("stirling2iter(n,m): n < 0");
+  if(m<0)return newerror("stirling2iter(n,m): m < 0");
+  /* no shortcuts here */
+
+  if(n<m) return 0;
+  if(n==0 && n!=m) return 0;
+  if(n==m) return 1;
+  if(m==0 )return 0;
+  if(m==1) return 1;
+  if(m==2) return 2^(n-1)-1;
+
+  nm = n-m;
+  if(__CZ__stirling2_n >= n && __CZ__stirling2_m >= m){
+    return __CZ__stirling2[n,m];
+  }
+  else{
+    __CZ__stirling2 = mat[n+1,m+1];
+    __CZ__stirling2[0,0] = 1;
+    for(i=1;i<=n;i++){
+      __CZ__stirling2[i,0] = 0;
+      for(j=1;j<=m;j++){
+        if(j<=i){
+          __CZ__stirling2[i, j] =   __CZ__stirling2[i -1, j -1] + (j )\
+                                  * __CZ__stirling2[i - 1, j];
+        }
+        else{
+         __CZ__stirling2[i, j] = 0;
+        }
+      }
+    }
+  }
+  __CZ__stirling2_n = (n);
+  __CZ__stirling2_m = (m);
+  return __CZ__stirling2[n,m];
+}
+
+define bell(n){
+  local sum s2list k A;
+
+  if(!isint(n)) return newerror("bell(n): n is not integer");
+  if(n < 0) return newerror("bell(n): n is not positive");
+  /* place some more shortcuts here?*/
+  if(n<=15){
+    mat A[16] = {
+                  1, 1, 2, 5, 15, 52, 203, 877, 4140, 21147, 115975, 678570,
+                  4213597, 27644437, 190899322, 1382958545
+                };
+    return A[n];
+  }
+  /* Start by generating the list of stirling numbers of the second kind */
+  s2list = stirling2caching(n,n//2);
+  if(iserror(s2list))
+    return newerror("bell(n): could not build stirling num. list");
+  sum = 0;
+  for(k=1;k<=n;k++){
+      sum += stirling2caching(n,k);
+  }
+  return sum;
+}
+
+define subfactorialrecursive(n){
+  if(n==0) return 1;
+  if(n==1) return 0;
+  if(n==2) return 1;
+  return n * subfactorialrecursive(n-1) + (-1)^n;
+}
+
+/* This is, quite amusingely, faster than the very same algorithm in
+   PARI/GP + GMP*/
+define subfactorialiterative(n){
+  local k temp1 temp2 ret;
+  if(n==0) return 1;
+  if(n==1) return 0;
+  if(n==2) return 1;
+  temp1 = 0;
+  ret   = 1;
+  for(k=3;k<=n;k++){
+    temp2 = temp1;
+    temp1 = ret;
+    ret =  (k-1) *(temp1 + temp2);
+  }
+  return ret;
+}
+
+define subfactorial(n){
+  local epsilon eps ret lnfact;
+  if(!isint(n))return  newerror("subfactorial(n): n is not integer.");
+  if(n < 0)return  newerror("subfactorial(n): n < 0");
+  return subfactorialiterative(n);
+}
+
+define risingfactorial(x,n){
+  local num denom quot ret;
+  if(n == 1) return x;
+  if(x==0) return newerror("risingfactorial(x,n): x == 0");
+  if(!isint(x) || !isint(n)){
+    return gamma(x+n)/gamma(x);
+  }
+  if(x<1)return newerror("risingfactorial(x,n): integer x and x < 1");
+  if(x+n < 1)return newerror("risingfactorial(x,n): integer x+n and x+n < 1");
+  if(x<9000&&n<9000){
+    return (x+n-1)!/(x-1)!;
+  }
+  else{
+    num   = __CZ__factor_factorial(x+n-1,1);
+    denom = __CZ__factor_factorial(x-1,1);
+    quot  = __CZ__subtract_factored_factorials( num , denom );
+      free(num,denom);
+    ret = __CZ__multiply_factored_factorial(quot);
+    return ret;
+  }
+}
+
+define fallingfactorial(x,n){
+  local num denom quot ret;
+  if(n == 0) return 1;
+
+  if(!isint(x) || !isint(n)){
+    if(x == n) return gamma(x+1);
+    return gamma(x+1)/gamma(x-n+1);
+  }
+  else{
+    if(x<0 || x-n < 0)
+     return newerror("fallingfactorial(x,n): integer x<0 or x-n < 0");
+    if(x == n) return factorial(x);
+    if(x<9000&&n<9000){
+      return (x)!/(x-n)!;
+    }
+    else{
+      num   = __CZ__factor_factorial(x,1);
+      denom = __CZ__factor_factorial(x-n,1);
+      quot  = __CZ__subtract_factored_factorials( num , denom );
+        free(num,denom);
+      ret = __CZ__multiply_factored_factorial(quot);
+      return ret;
+    }
+  }
+}
+
+
+/*
+ * restore internal function from resource debugging
+ * report important interface functions
+ */
+config("resource_debug", resource_debug_level),;
+if (config("resource_debug") & 3) {
+    print "binomial(n,k)";
+    print "bigcatalan(n)";
+    print "doublefactorial(n)";
+    print "subfactorial(n)";
+    print "stirling1(n,m)";
+    print "stirling2(n,m)";
+    print "stirling2caching(n,m)";
+    print "bell(n)";
+    print "subfactorial(n)";
+    print "risingfactorial(x,n)";
+    print "fallingfactorial(x,n)";
+}

cal/infinities.cal

@@ -0,0 +1,88 @@
+/*
+ * infinities - handle infinities symbolically, a little helper file
+ *
+ * Copyright (C) 2013 Christoph Zurnieden
+ *
+ * Calc is open software; you can redistribute it and/or modify it under
+ * the terms of the version 2.1 of the GNU Lesser General Public License
+ * as published by the Free Software Foundation.
+ *
+ * Calc is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+ * or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU Lesser General
+ * Public License for more details.
+ *
+ * A copy of version 2.1 of the GNU Lesser General Public License is
+ * distributed with calc under the filename COPYING-LGPL.  You should have
+ * received a copy with calc; if not, write to Free Software Foundation, Inc.
+ * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
+ */
+
+
+static resource_debug_level;
+resource_debug_level = config("resource_debug", 0);
+
+
+define isinfinite(x)
+{
+    if (isstr(x)) {
+	if (strncmp(x, "cinf", 4) == 0
+	    || strncmp(x, "pinf", 4) == 0 || strncmp(x, "ninf", 4) == 0)
+	    return 1;
+    }
+    return 0;
+}
+
+define iscinf(x)
+{
+    if (isstr(x)) {
+	if (strncmp(x, "cinf", 4) == 0)
+	    return 1;
+    }
+    return 0;
+}
+
+define ispinf(x)
+{
+    if (isstr(x)) {
+	if (strncmp(x, "pinf", 4) == 0)
+	    return 1;
+    }
+    return 0;
+}
+
+define isninf(x)
+{
+    if (isstr(x)) {
+	if (strncmp(x, "ninf", 4) == 0)
+	    return 1;
+    }
+    return 0;
+}
+
+define cinf()
+{
+    return "cinf";
+}
+
+define ninf()
+{
+    return "ninf";
+}
+
+define pinf()
+{
+    return "pinf";
+}
+
+
+config("resource_debug", resource_debug_level),;
+if (config("resource_debug") & 3) {
+    print "isinfinite(x)";
+    print "iscinf(x)";
+    print "ispinf(x)";
+    print "isninf(x)";
+    print "cinf()";
+    print "ninf()";
+    print "pinf()";
+}

cal/intnum.cal

@@ -0,0 +1,728 @@
+/*
+ * intnum - implementation of tanhsinh- and Gauss-Legendre quadrature
+ *
+ * Copyright (C) 2013 Christoph Zurnieden
+ *
+ * Calc is open software; you can redistribute it and/or modify it under
+ * the terms of the version 2.1 of the GNU Lesser General Public License
+ * as published by the Free Software Foundation.
+ *
+ * Calc is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+ * or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU Lesser General
+ * Public License for more details.
+ *
+ * A copy of version 2.1 of the GNU Lesser General Public License is
+ * distributed with calc under the filename COPYING-LGPL.  You should have
+ * received a copy with calc; if not, write to Free Software Foundation, Inc.
+ * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
+ */
+
+
+static resource_debug_level;
+resource_debug_level = config("resource_debug", 0);
+
+
+read -once infinities;
+
+static __CZ__tanhsinh_x;
+static __CZ__tanhsinh_w;
+static __CZ__tanhsinh_order;
+static __CZ__tanhsinh_prec;
+
+define quadtsdeletenodes()
+{
+    free(__CZ__tanhsinh_x);
+    free(__CZ__tanhsinh_w);
+    free(__CZ__tanhsinh_order);
+    free(__CZ__tanhsinh_prec);
+}
+
+define quadtscomputenodes(order, expo, eps)
+{
+    local t cht sht chp sum k PI places;
+    local h t0 x w;
+    if (__CZ__tanhsinh_order == order && __CZ__tanhsinh_prec == eps)
+	return 1;
+    __CZ__tanhsinh_order = order;
+    __CZ__tanhsinh_prec = eps;
+    __CZ__tanhsinh_x = list();
+    __CZ__tanhsinh_w = list();
+    /* The tanhsinh algorithm needs a slightly higher precision than G-L */
+    eps = epsilon(eps * 1e-2);
+    places = highbit(1 + int (1 / epsilon())) +1;
+    PI = pi();
+    sum = 0;
+    t0 = 2 ^ (-expo);
+    h = 2 * t0;
+    /*
+     * The author wanted to use the mpmath trick here which was
+     * advertised---and reasonably so!---to be faster. Didn't work out
+     * so well with calc.
+     * PI4 = PI/4;
+     * expt0 = bround(exp(t0),places);
+     * a = bround( PI4 * expt0,places);
+     * b = bround(PI4 / expt0,places);
+     * udelta = bround(exp(h),places);
+     * urdelta = bround(1/udelta,places);
+     */
+    /* make use of x(-t) = -x(t), w(-t) = w(t)  */
+    for (k = 0; k < 20 * order + 1; k++) {
+	/*
+	 * x = tanh(pi/2 * sinh(t))
+	 * w = pi/2 * cosh(t) / cosh(pi/2 * sinh(t))^2
+	 */
+	t = bround(t0 + k * h, places);
+
+	cht = bround(cosh(t), places);
+	sht = bround(sinh(t), places);
+	chp = bround(cosh(0.5 * PI * sht), places);
+	x = bround(tanh(0.5 * PI * sht), places);
+	w = bround((PI * h * cht) / (2 * chp ^ 2), places);
+	/*
+	 * c = bround(exp(a-b),places);
+	 * d = bround(1/c,places);
+	 * co =bround( (c+d)/2,places);
+	 * si =bround( (c-d)/2,places);
+	 * x = bround(si / co,places);
+	 * w = bround((a+b) / co^2,places);
+	 */
+	if (abs(x - 1) <= eps)
+	    break;
+
+	append(__CZ__tanhsinh_x, x);
+	append(__CZ__tanhsinh_w, w);
+	/*
+	 * a *= udelta;
+	 * b *= urdelta;
+	 */
+    }
+
+    /* Normalize the weights to make them add up to 2 (two) */
+    /*
+     * for(k=0;k < size(__CZ__tanhsinh_w);k++)
+     * sum = bround(sum + __CZ__tanhsinh_w[k],places);
+     * sum *= 2;
+     * for(k=0;k < size(__CZ__tanhsinh_w);k++)
+     * __CZ__tanhsinh_w[k] = bround(2.0 * __CZ__tanhsinh_w[k] / sum,places);
+     */
+
+    epsilon(eps);
+    return 1;
+}
+
+define quadtscore(a, b, n)
+{
+    local k c d order eps places sum ret x x1 x2 xm w w1 w2 m sizel;
+
+    eps = epsilon(epsilon() * 1e-2);
+    places = highbit(1 + int (1 / epsilon())) +1;
+    m = int (4 + max(0, ln(places / 30.0) / ln(2))) + 2;
+    if (!isnull(n)) {
+	order = n;
+	m = ilog(order / 3, 2) + 1;
+    } else
+	order = 3 * 2 ^ (m - 1);
+
+    quadtscomputenodes(order, m, epsilon());
+    sizel = size(__CZ__tanhsinh_w);
+
+    if (isinfinite(a) || isinfinite(b)) {
+	/*
+	 *           x
+	 * t =  ------------
+	 *                2
+	 *      sqrt(1 - y )
+	 */
+	if (isninf(a) && ispinf(b)) {
+	    for (k = 0; k < sizel; k++) {
+		x1 = __CZ__tanhsinh_x[k];
+		x2 = -__CZ__tanhsinh_x[k];
+		w1 = __CZ__tanhsinh_w[k];
+
+		x = bround(x1 * (1 - x1 ^ 2) ^ (-1 / 2), places);
+		xm = bround(x2 * (1 - x2 ^ 2) ^ (-1 / 2), places);
+		w = bround(w1 * (((1 - x1 ^ 2) ^ (-1 / 2)) / (1 - x1 ^ 2)),
+			   places);
+		w2 = bround(w1 * (((1 - x2 ^ 2) ^ (-1 / 2)) / (1 - x2 ^ 2)),
+			    places);
+		sum += bround(w * f(x), places);
+		sum += bround(w2 * f(xm), places);
+	    }
+	}
+	/*
+	 *        1
+	 * t =  - - + b + 1
+	 *        x
+	 */
+	else if (isninf(a) && !iscinf(b)) {
+	    for (k = 0; k < sizel; k++) {
+		x1 = __CZ__tanhsinh_x[k];
+		x2 = -__CZ__tanhsinh_x[k];
+		w1 = __CZ__tanhsinh_w[k];
+
+		x = bround((b + 1) - (2 / (x1 + 1)), places);
+		xm = bround((b + 1) - (2 / (x2 + 1)), places);
+		w = bround(w1 * (1 / 2 * (2 / (x1 + 1)) ^ 2), places);
+		w2 = bround(w1 * (1 / 2 * (2 / (x2 + 1)) ^ 2), places);
+		sum += bround(w * f(x), places);
+		sum += bround(w2 * f(xm), places);
+	    }
+	}
+	/*
+	 *      1
+	 * t =  - + a - 1
+	 *      x
+	 */
+	else if (!iscinf(a) && ispinf(b)) {
+	    for (k = 0; k < sizel; k++) {
+		x1 = __CZ__tanhsinh_x[k];
+		x2 = -__CZ__tanhsinh_x[k];
+		w1 = __CZ__tanhsinh_w[k];
+		x = bround((a - 1) + (2 / (x1 + 1)), places);
+		xm = bround((a - 1) + (2 / (x2 + 1)), places);
+		w = bround(w1 * (((1 / 2) * (2 / (x1 + 1)) ^ 2)), places);
+		w2 = bround(w1 * (((1 / 2) * (2 / (x2 + 1)) ^ 2)), places);
+		sum += bround(w * f(x), places);
+		sum += bround(w2 * f(xm), places);
+	    }
+	} else if (isninf(a) || isninf(b)) {
+	    /*TODO: swap(a,b) and negate(w)? Lookup! */
+	    return newerror("quadtscore: reverse limits?");
+	} else {
+	    return
+		newerror("quadtscore: complex infinity not yet implemented");
+        }
+	ret = sum;
+    } else {
+	/* Avoid rounding errors */
+	if (a == -1 && b == 1) {
+	    c = 1;
+	    d = 0;
+	} else {
+	    c = (b - a) / 2;
+	    d = (b + a) / 2;
+	}
+	sum = 0;
+	for (k = 0; k < sizel; k++) {
+	    sum +=
+		bround(__CZ__tanhsinh_w[k] * f(c * __CZ__tanhsinh_x[k] + d),
+		       places);
+	    sum +=
+		bround(__CZ__tanhsinh_w[k] * f(c * -__CZ__tanhsinh_x[k] + d),
+		       places);
+	}
+	ret = c * sum;
+    }
+    epsilon(eps);
+    return ret;
+}
+
+static __CZ__quadts_error;
+
+define quadts(a, b, points)
+{
+    local k sp results epsbits nsect interval length segment slope C ;
+    local x1 x2 y1 y2  sum D1 D2 D3 D4;
+    if (param(0) < 2)
+	return newerror("quadts: not enough arguments");
+    epsbits = highbit(1 + int (1 / epsilon())) +1;
+    if (param(0) < 3 || isnull(points)) {
+	/* return as given */
+	return quadtscore(a, b);
+    } else {
+	if ((isinfinite(a) || isinfinite(b))
+	    && (!ismat(points) && !islist(points)))
+	    return
+		newerror(strcat
+			 ("quadts: segments of infinite length ",
+			  "are not yet supported"));
+	if (ismat(points) || islist(points)) {
+	    sp = size(points);
+	    if (sp == 0)
+		return
+		    newerror(strcat
+			     ("quadts: variable 'points` must be a list or ",
+			      "1d-matrix of a length > 0"));
+	    /* check if all points are numbers */
+	    for (k = 0; k < sp; k++) {
+		if (!isnum(points[k]))
+		    return
+			newerror(strcat
+				 ("quadts: elements of 'points` must be",
+				  " numbers only"));
+	    }
+	    /* We have n-1 intervals and a and b, hence n-1 + 2 results */
+	    results = mat[sp + 1];
+            if (a != points[0]) {
+	        results[0] = quadtscore(a, points[0]);
+            } else {
+                results[0] = 0;
+            }
+	    if (sp == 1) {
+                if (b != points[0]) {
+		    results[1] = quadtscore(points[0], b);
+                } else {
+                    results[1] = 0;
+                }
+	    } else {
+		for (k = 1; k < sp; k++) {
+		    results[k] = quadtscore(points[k - 1], points[k]);
+		}
+                if (b != points[k - 1]) {
+		    results[k] = quadtscore(points[k - 1], b);
+                } else {
+                    results[k] = 0;
+                }
+	    }
+	} else {
+	    if (!isint(points) || points <= 0)
+		return newerror(strcat("quadts: variable 'points` must be a ",
+				       "list or a positive integer"));
+	    /* Taking "points" as the number of equally spaced intervals */
+	    results = mat[points + 1];
+	    /* It is easy if a,b lie on the real line */
+	    if (isreal(a) && isreal(b)) {
+		length = abs(a - b);
+		segment = length / points;
+
+		for (k = 1; k <= points; k++) {
+		    results[k - 1] =
+			quadtscore(a + (k - 1) * segment, a + k * segment);
+		}
+	    } else {
+		/* We have at least one complex limit but treat "points" still
+                 * as the number of equally spaced intervals on a straight line
+                 * connecting a and b. Computing the segments here is a bit
+                 * more complicated but not much, it should have been taught in
+                 * highschool.
+		 * Other contours by way of a list of points */
+		slope = (im(b) - im(a)) / (re(b) - re(a));
+		C = (im(a) + slope) * re(a);
+		length = abs(re(a) - re(b));
+		segment = length / points;
+
+		/* y = mx+C where m is the slope, x is the real part and y the
+		 * imaginary part  */
+                if(re(a)>re(b))swap(a,b);
+		for (k = re(a); k <= (re(b)); k+=segment) {
+		    x1 = slope*(k) +  C;
+		    results[k] = quadtscore(k + x1 * 1i);
+		}
+	    }			/* else of isreal */
+	}			/* else of ismat|islist */
+    }				/* else of isnull(points) */
+    /* With a bit of undeserved luck we have a result by now. */
+    sp = size(results);
+    for (k = 0; k < sp; k++) {
+	sum += results[k];
+    }
+    return sum;
+}
+
+static __CZ__gl_x;
+static __CZ__gl_w;
+static __CZ__gl_order;
+static __CZ__gl_prec;
+
+define quadglcomputenodes(N)
+{
+    local places k l x w t1 t2 t3 t4 t5 r tmp;
+
+    if (__CZ__gl_order == N && __CZ__gl_prec == epsilon())
+	return;
+
+    __CZ__gl_x = mat[N];
+    __CZ__gl_w = mat[N];
+    __CZ__gl_order = N;
+    __CZ__gl_prec = epsilon();
+
+    places = highbit(1 + int (1 / epsilon())) +1;
+
+    /*
+     * Compute roots and weights (doing it inline seems to be fastest)
+     * Trick shamelessly stolen from D. Bailey et .al (program "arprec")
+     */
+    for (k = 1; k <= N//2; k++) {
+	r = bround(cos(pi() * (k - .25) / (N + .5)), places);
+	while (1) {
+	    t1 = 1, t2 = 0;
+	    for (l = 1; l <= N; l++) {
+		t3 = t2;
+		t2 = t1;
+		t1 = bround(((2 * l - 1) * r * t2 - (l - 1) * t3) / l, places);
+	    }
+	    t4 = bround(N * (r * t1 - t2) / ((r ^ 2) - 1), places);
+	    t5 = r;
+	    tmp = t1 / t4;
+	    r = r - tmp;
+	    if (abs(tmp) <= epsilon())
+		break;
+	}
+	x = r;
+	w = bround(2 / ((1 - r ^ 2) * t4 ^ 2), places);
+
+	__CZ__gl_x[k - 1] = x;
+	__CZ__gl_w[k - 1] = w;
+	__CZ__gl_x[N - k] = -__CZ__gl_x[k - 1];
+	__CZ__gl_w[N - k] = __CZ__gl_w[k - 1];
+    }
+    return;
+}
+
+define quadgldeletenodes()
+{
+    free(__CZ__gl_x);
+    free(__CZ__gl_w);
+    free(__CZ__gl_order);
+    free(__CZ__gl_prec);
+}
+
+define quadglcore(a, b, n)
+{
+    local k c d digs order eps places sum ret err x x1 w w1 m;
+    local phalf x2 px1 spx1 u b1 a1 half;
+
+    eps = epsilon(epsilon() * 1e-2);
+    places = highbit(1 + int (1 / epsilon())) +1;
+    if (!isnull(n))
+	order = n;
+    else {
+	m = int (4 + max(0, ln(places / 30.0) / ln(2))) + 2;
+	order = 3 * 2 ^ (m - 1);
+    }
+
+
+    quadglcomputenodes(order, 1);
+
+    if (isinfinite(a) || isinfinite(b)) {
+	if (isninf(a) && ispinf(b)) {
+	    for (k = 0; k < order; k++) {
+		x1 = __CZ__gl_x[k];
+		w1 = __CZ__gl_w[k];
+
+		x = bround(x1 * (1 - x1 ^ 2) ^ (-1 / 2), places);
+		w = bround(w1 * (((1 - x1 ^ 2) ^ (-1 / 2)) / (1 - x1 ^ 2)),
+			   places);
+		sum += bround(w * f(x), places);
+	    }
+	} else if (isninf(a) && !iscinf(b)) {
+	    for (k = 0; k < order; k++) {
+		x1 = __CZ__gl_x[k];
+		w1 = __CZ__gl_w[k];
+
+		x = bround((b + 1) - (2 / (x1 + 1)), places);
+		w = bround(w1 * (1 / 2 * (2 / (x1 + 1)) ^ 2), places);
+		sum += bround(w * f(x), places);
+	    }
+	} else if (!iscinf(a) && ispinf(b)) {
+	    for (k = 0; k < order; k++) {
+		x1 = __CZ__gl_x[k];
+		w1 = __CZ__gl_w[k];
+		x = bround((a - 1) + (2 / (x1 + 1)), places);
+		w = bround(w1 * (((1 / 2) * (2 / (x1 + 1)) ^ 2)), places);
+		sum += bround(w * f(x), places);
+	    }
+	} else if (isninf(a) || isninf(b)) {
+	    /*TODO: swap(a,b) and negate(w)? Lookup! */
+	    return newerror("quadglcore: reverse limits?");
+	} else
+	    return
+		newerror("quadglcore: complex infinity not yet implemented");
+	ret = sum;
+    } else {
+	/* Avoid rounding errors */
+	if (a == -1 && b == 1) {
+	    c = 1;
+	    d = 0;
+	} else {
+	    c = (b - a) / 2;
+	    d = (b + a) / 2;
+	}
+	sum = 0;
+	for (k = 0; k < order; k++) {
+	    sum += bround(__CZ__gl_w[k] * f(c * __CZ__gl_x[k] + d), places);
+	}
+	ret = c * sum;
+    }
+    epsilon(eps);
+    return ret;
+}
+
+define quadgl(a, b, points)
+{
+    local k sp results epsbits nsect interval length segment slope C x1 y1 x2
+	y2;
+    local sum D1 D2 D3 D4;
+    if (param(0) < 2)
+	return newerror("quadgl: not enough arguments");
+    epsbits = highbit(1 + int (1 / epsilon())) +1;
+    if (isnull(points)) {
+	/* return as given */
+	return quadglcore(a, b);
+    } else {
+	/* But if we could half the time needed to execute a single operation
+	 * we could do all of it in just twice that time. */
+	if (isinfinite(a) || isinfinite(b)
+	    && (!ismat(points) && !islist(points)))
+	    return
+		newerror(strcat
+			 ("quadgl: multiple segments of infinite length ",
+			  "are not yet supported"));
+	if (ismat(points) || islist(points)) {
+	    sp = size(points);
+	    if (sp == 0)
+		return
+		    newerror(strcat
+			     ("quadgl: variable 'points` must be a list or ",
+			      "1d-matrix of a length > 0"));
+	    /* check if all points are numbers */
+	    for (k = 0; k < sp; k++) {
+		if (!isnum(points[k]))
+		    return
+			newerror(strcat
+				 ("quadgl: elements of 'points` must be ",
+				  "numbers only"));
+	    }
+	    /* We have n-1 intervals and a and b, hence n-1 + 2 results */
+	    results = mat[sp + 1];
+            if (a != points[0]) {
+	        results[0] = quadglcore(a, points[0]);
+            } else {
+                results[0] = 0;
+            }
+	    if (sp == 1) {
+		if (b != points[0]) {
+		    results[1] = quadglcore(points[0], b);
+                } else {
+                    results[1] = 0;
+                }
+	    } else {
+		for (k = 1; k < sp; k++) {
+		    results[k] = quadglcore(points[k - 1], points[k]);
+		}
+                if (b != points[k - 1]) {
+		    results[k] = quadglcore(points[k - 1], b);
+                } else {
+                    results[k] = 0;
+                }
+	    }
+	} else {
+	    if (!isint(points) || points <= 0)
+		return newerror(strcat("quadgl: variable 'points` must be a ",
+				       "list or a positive integer"));
+	    /* Taking "points" as the number of equally spaced intervals */
+	    results = mat[points + 1];
+	    /* It is easy if a,b lie on the real line */
+	    if (isreal(a) && isreal(b)) {
+		length = abs(a - b);
+		segment = length / points;
+
+		for (k = 1; k <= points; k++) {
+		    results[k - 1] =
+			quadglcore(a + (k - 1) * segment, a + k * segment);
+		}
+	    } else {
+		/* Other contours by way of a list of points */
+		slope = (im(b) - im(a)) / (re(b) - re(a));
+		C = (im(a) + slope) * re(a);
+		length = abs(re(a) - re(b));
+		segment = length / points;
+
+		/* y = mx+C where m is the slope, x is the real part and y the
+		 * imaginary part  */
+                if(re(a)>re(b))swap(a,b);
+		for (k = re(a); k <= (re(b)); k+=segment) {
+		    x1 = slope*(k) +  C;
+		    results[k] = quadglcore(k + x1 * 1i);
+		}
+	    }			/* else of isreal */
+	}			/* else of ismat|islist */
+    }				/* else of isnull(points) */
+    /* With a bit of undeserved luck we have a result by now. */
+    sp = size(results);
+    for (k = 0; k < sp; k++) {
+	sum += results[k];
+    }
+    return sum;
+}
+
+define quad(a, b, points = -1, method = "tanhsinh")
+{
+    if (isnull(a) || isnull(b) || param(0) < 2)
+	return newerror("quad: both limits must be given");
+    if (isstr(a)) {
+	if (strncmp(a, "cinf", 1) == 0)
+	    return
+		newerror(strcat
+			 ("quad: complex infinity not yet supported, use",
+			  " 'pinf' or 'ninf' respectively"));
+    }
+    if (isstr(b)) {
+	if (strncmp(b, "cinf", 1) == 0)
+	    return
+		newerror(strcat
+			 ("quad: complex infinity not yet supported, use",
+			  " 'pinf' or 'ninf' respectively"));
+    }
+
+    if (param(0) == 3) {
+	if (isstr(points))
+	    method = points;
+    }
+
+    if (strncmp(method, "tanhsinh", 1) == 0) {
+	if (!isstr(points)) {
+	    if (points == -1) {
+		return quadts(a, b);
+	    } else {
+		return quadts(a, b, points);
+	    }
+	} else {
+	    return quadts(a, b);
+	}
+    }
+
+    if (strncmp(method, "gausslegendre", 1) == 0) {
+	if (!isstr(points)) {
+	    if (points == -1) {
+		return quadgl(a, b);
+	    } else {
+		return quadgl(a, b, points);
+	    }
+	} else {
+	    return quadgl(a, b);
+	}
+    }
+}
+
+define makerange(start, end, steps)
+{
+    local ret k l step C length slope x1 x2 y1 y2;
+    local segment;
+    steps = int (steps);
+    if (steps < 1) {
+	return newerror("makerange: number of steps must be > 0");
+    }
+    if (!isnum(start) || !isnum(end)) {
+	return newerror("makerange: only numbers are supported yet");
+    }
+    if (isreal(start) && isreal(end)) {
+	step = (end - start) / (steps);
+	print step;
+	ret = mat[steps + 1];
+	for (k = 0; k <= steps; k++) {
+	    ret[k] = k * step + start;
+	}
+    } else {
+	ret = mat[steps + 1];
+	if (re(start) > re(end)) {
+	    swap(start, end);
+	}
+
+	slope = (im(end) - im(start)) / (re(end) - re(start));
+	C = im(start) - slope * re(start);
+	length = abs(re(start) - re(end));
+	segment = length / (steps);
+
+	for (k = re(start), l = 0; k <= (re(end)); k += segment, l++) {
+	    x1 = slope * (k) + C;
+	    ret[l] = k + x1 * 1i;
+	}
+
+    }
+    return ret;
+}
+
+define makecircle(radius, center, points)
+{
+    local ret k a b twopi centerx centery;
+    if (!isint(points) || points < 2) {
+	return
+	    newerror("makecircle: number of points is not a positive integer");
+    }
+    if (!isnum(center)) {
+	return newerror("makecircle: center does not lie on the complex plane");
+    }
+    if (!isreal(radius) || radius <= 0) {
+	return newerror("makecircle: radius is not a real > 0");
+    }
+    ret = mat[points];
+    twopi = 2 * pi();
+    centerx = re(center);
+    centery = im(center);
+    for (k = 0; k < points; k++) {
+	a = centerx + radius * cos(twopi * k / points);
+	b = centery + radius * sin(twopi * k / points);
+	ret[k] = a + b * 1i;
+    }
+    return ret;
+}
+
+define makeellipse(angle, a, b, center, points)
+{
+    local ret k x y twopi centerx centery;
+    if (!isint(points) || points < 2) {
+	return
+	    newerror("makeellipse: number of points is not a positive integer");
+    }
+    if (!isnum(center)) {
+	return
+	    newerror("makeellipse: center does not lie on the complex plane");
+    }
+    if (!isreal(a) || a <= 0) {
+	return newerror("makecircle: a is not a real > 0");
+    }
+    if (!isreal(b) || b <= 0) {
+	return newerror("makecircle: b is not a real > 0");
+    }
+    if (!isreal(angle)) {
+	return newerror("makecircle: angle is not a real");
+    }
+    ret = mat[points];
+    twopi = 2 * pi();
+    centerx = re(center);
+    centery = im(center);
+    for (k = 0; k < points; k++) {
+	x = centerx + a * cos(twopi * k / points) * cos(angle)
+	    - b * sin(twopi * k / points) * sin(angle);
+	y = centerx + a * cos(twopi * k / points) * sin(angle)
+	    + b * sin(twopi * k / points) * cos(angle);
+	ret[k] = x + y * 1i;
+    }
+    return ret;
+}
+
+define makepoints()
+{
+    local ret k;
+    ret = mat[param(0)];
+    for (k = 0; k < param(0); k++) {
+	if (!isnum(param(k + 1))) {
+	    return
+		newerror(strcat
+			 ("makepoints: parameter number \"", str(k + 1),
+			  "\" is not a number"));
+	}
+	ret[k] = param(k + 1);
+    }
+    return ret;
+}
+
+
+config("resource_debug", resource_debug_level),;
+if (config("resource_debug") & 3) {
+    print "quadtsdeletenodes()";
+    print "quadtscomputenodes(order, expo, eps)";
+    print "quadtscore(a,b,n)";
+    print "quadts(a,b,points)";
+    print "quadglcomputenodes(N)";
+    print "quadgldeletenodes()";
+    print "quadglcore(a,b,n)";
+    print "quadgl(a,b,points)";
+    print "quad(a,b,points=-1,method=\"tanhsinh\")";
+    print "makerange(start, end, steps)";
+    print "makecircle(radius, center, points)";
+    print "makeellipse(angle, a, b, center, points)";
+    print "makepoints(a1,[...])";
+}

cal/lambertw.cal

@@ -0,0 +1,288 @@
+/*
+ * lambertw - Lambert's W-function
+ *
+ * Copyright (C) 2013 Christoph Zurnieden
+ *
+ * Calc is open software; you can redistribute it and/or modify it under
+ * the terms of the version 2.1 of the GNU Lesser General Public License
+ * as published by the Free Software Foundation.
+ *
+ * Calc is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+ * or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU Lesser General
+ * Public License for more details.
+ *
+ * A copy of version 2.1 of the GNU Lesser General Public License is
+ * distributed with calc under the filename COPYING-LGPL.  You should have
+ * received a copy with calc; if not, write to Free Software Foundation, Inc.
+ * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
+ *
+ * @(#) $Revision: 30.4 $
+ * @(#) $Id: lambertw.cal,v 30.4 2013/08/18 20:01:53 chongo Exp $
+ * @(#) $Source: /usr/local/src/bin/calc/cal/RCS/lambertw.cal,v $
+ *
+ * Under source code control:	2013/08/11 01:31:28
+ * File existed as early as:	2013
+ */
+
+
+static resource_debug_level;
+resource_debug_level = config("resource_debug", 0);
+
+
+/*
+
+     R. M. Corless and G. H. Gonnet and D. E. G. Hare and D. J. Jeffrey and
+     D. E. Knuth, "On the Lambert W Function", Advances n Computational
+     Mathematics, 329--359, (1996)
+     http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.112.6117
+
+     D. J. Jeffrey, D. E. G. Hare, R. M. Corless, "Unwinding the branches of the
+     Lambert W function", The Mathematical Scientist, 21, pp 1-7, (1996)
+     http://www.apmaths.uwo.ca/~djeffrey/Offprints/wbranch.pdf
+
+     Darko Verebic, "Having Fun with Lambert W(x) Function"
+     arXiv:1003.1628v1, March 2010, http://arxiv.org/abs/1003.1628
+
+     Winitzki, S. "Uniform Approximations for Transcendental Functions",
+     In Part 1 of Computational Science and its Applications - ICCSA 2003,
+     Lecture Notes in Computer Science, Vol. 2667, Springer-Verlag,
+     Berlin, 2003, 780-789. DOI 10.1007/3-540-44839-X_82
+     A copy may be found by Google.
+
+
+*/
+static true = 1;
+static false = 0;
+
+/* Branch 0, Winitzki (2003) , the well known Taylor series*/
+define __CZ__lambertw_0(z,eps){
+  local a=2.344e0, b=0.8842e0, c=0.9294e0, d=0.5106e0, e=-1.213e0;
+  local y=sqrt(2*exp(1)*z+2);
+  return (2*ln(1+b*y)-ln(1+c*ln(1+d*y))+e)/(1+1/(2*ln(1+b*y)+2*a));
+}
+/* branch -1 */
+define __CZ__lambertw_m1(z,eps){
+  local wn k;
+  /* Cut-off found in Maxima */
+  if(z < 0.3) return __CZ__lambertw_app(z,eps);
+  wn = z;
+  /*  Verebic (2010) eqs. 16-18*/
+  for(k=0;k<10;k++){
+    wn = ln(-z)-ln(-wn);
+  }
+  return wn;
+}
+
+/*
+  generic approximation
+
+  series for 1+W((z-2)/(2 e))
+
+  Corless et al (1996) (4.22)
+  Verebic (2010) eqs. 35-37; more coefficients given at the end of sect. 3.1
+  or online
+     http://www.wolframalpha.com/input/?
+       i=taylor+%28+1%2Bproductlog%28+%28z-2%29%2F%282*e%29+%29+%29
+  or by using the function lambertw_series_print() after running
+  lambertw_series(z,eps,branch,terms) at least once with the wanted number of
+  terms and z = 1 (which might throw an error because the series will not
+  converge in anybodies lifetime for something that far from the branchpoint).
+
+
+*/
+define __CZ__lambertw_app(z,eps){
+  local  b0=-1, b1=1, b2=-1/3, b3=11/72;
+  local  y=sqrt(2*exp(1)*z+2);
+  return b0 + ( y * (b1 + (y * (b2 + (b3 * y)))));
+}
+
+static __CZ__Ws_a;
+static __CZ__Ws_c;
+static __CZ__Ws_len=0;
+
+define lambertw_series_print(){
+  local k;
+  for(k=0;k<__CZ__Ws_len;k++){
+    print num(__CZ__Ws_c[k]):"/":den(__CZ__Ws_c[k]):"*p^":k;
+  }
+}
+
+/*
+   The series is fast but only if _very_ close to the branchpoint
+   The exact branch must be given explicitly, e.g.:
+
+; lambertw(-exp(-1)+.001)-lambertw_series(-exp(-1)+.001,epsilon()*1e-10,0)
+	-0.14758879113205794065490184399030194122136720202792-
+                  0.00000000000000000000000000000000000000000000000000i
+; lambertw(-exp(-1)+.001)-lambertw_series(-exp(-1)+.001,epsilon()*1e-10,1)
+	0.00000000000000000000000000000000000000000000000000-
+                  0.00000000000000000000000000000000000000000000000000i
+ */
+define lambertw_series(z,eps,branch,terms){
+  local k l limit tmp sum A C P PP epslocal;
+  if(!isnull(terms))
+    limit = terms;
+  else
+    limit = 100;
+
+  if(isnull(eps))
+    eps = epsilon(epsilon()*1e-10);
+  epslocal = epsilon(eps);
+
+  P = sqrt(2*(exp(1)*z+1));
+  if(branch != 0) P = -P;
+  tmp=0;sum=0;PP=P;
+
+    __CZ__Ws_a   = mat[limit+1];
+    __CZ__Ws_c   = mat[limit+1];
+    __CZ__Ws_len = limit;
+   /*
+      c0 = -1; c1 = 1
+      a0 =  2; a1 =-1
+    */
+    __CZ__Ws_c[0] = -1; __CZ__Ws_c[1] =  1;
+    __CZ__Ws_a[0] =  2; __CZ__Ws_a[1] = -1;
+    sum += __CZ__Ws_c[0];
+    sum += __CZ__Ws_c[1] * P;
+    PP *= P;
+    for(k=2;k<limit;k++){
+      for(l=2;l<k;l++){
+         __CZ__Ws_a[k] += __CZ__Ws_c[l]*__CZ__Ws_c[k+1-l];
+      }
+
+      __CZ__Ws_c[k] = (k-1) * (  __CZ__Ws_c[k-2]/2
+                            +__CZ__Ws_a[k-2]/4)/
+                      (k+1)-__CZ__Ws_a[k]/2-__CZ__Ws_c[k-1]/(k+1);
+      tmp = __CZ__Ws_c[k] * PP;
+      sum += tmp;
+      if(abs(tmp) <= eps){
+        epsilon(epslocal);
+        return sum;
+      }
+      PP *= P;
+    }
+    epsilon(epslocal);
+    return
+      newerror(strcat("lambertw_series: does not converge in ",
+                        str(limit)," terms" ));
+}
+
+/* */
+define lambertw(z,branch){
+  local eps epslarge ret branchpoint bparea w we ew w1e wn k places m1e;
+  local closeness;
+
+  eps = epsilon();
+  if(branch == 0){
+    if(!im(z)){
+      if(abs(z) <= eps) return 0;
+      if(abs(z-exp(1)) <= eps) return 1;
+      if(abs(z - (-ln(2)/2)) <= eps ) return -ln(2);
+      if(abs(z - (-pi()/2)) <= eps ) return 1i*pi()/2;
+    }
+  }
+
+  branchpoint = -exp(-1);
+  bparea = .2;
+  if(branch == 0){
+    if(!im(z) && abs(z-branchpoint) == 0) return -1;
+    ret = __CZ__lambertw_0(z,eps);
+   /* Yeah, C&P, I know, sorry */
+   ##ret = ln(z) + 2*pi()*1i*branch - ln(ln(z)+2*pi()*1i*branch);
+  }
+  else if(branch == 1){
+    if(im(z)<0 && abs(z-branchpoint) <= bparea)
+      ret = __CZ__lambertw_app(z,eps);
+   /* Does calc have a goto? Oh, it does!  */
+    ret =ln(z) + 2*pi()*1i*branch - ln(ln(z)+2*pi()*1i*branch);
+  }
+  else if(branch == -1){##print "-1";
+    if(!im(z) && abs(z-branchpoint) == 0) return -1;
+    if(!im(z) && z>branchpoint && z < 0){##print "0";
+      ret = __CZ__lambertw_m1(z,eps);}
+    if(im(z)>=0 && abs(z-branchpoint) <= bparea){##print "1";
+      ret = __CZ__lambertw_app(z,eps);}
+    ret =ln(z) + 2*pi()*1i*branch - ln(ln(z)+2*pi()*1i*branch);
+  }
+  else
+    ret = ln(z) + 2*pi()*1i*branch - ln(ln(z)+2*pi()*1i*branch);
+
+  /*
+     Such a high precision is only needed _very_ close to the branchpoint
+     and might even be insufficient if z has not been computed with
+     sufficient precision itself (M below was calculated by Mathematica and also
+     with the series above with epsilon(1e-200)):
+    ; epsilon(1e-50)
+	0.00000000000000000001
+    ; display(50)
+	20
+    ; M=-0.9999999999999999999999997668356018402875796636464119050387
+    ; lambertw(-exp(-1)+1e-50,0)-M
+	-0.00000000000000000000000002678416515423276355643684
+    ; epsilon(1e-60)
+	0.0000000000000000000000000000000000000000000000000
+    ; A=-exp(-1)+1e-50
+    ; epsilon(1e-50)
+	0.00000000000000000000000000000000000000000000000000
+    ; lambertw(A,0)-M
+	-0.00000000000000000000000000000000000231185460220585
+    ; lambertw_series(A,epsilon(),0)-M
+	-0.00000000000000000000000000000000000132145133161626
+    ; epsilon(1e-100)
+	0.00000000000000000000000000000000000000000000000001
+    ; A=-exp(-1)+1e-50
+    ; epsilon(1e-65)
+	0.00000000000000000000000000000000000000000000000000
+    ; lambertw_series(A,epsilon(),0)-M
+	0.00000000000000000000000000000000000000000000000000
+    ; lambertw_series(-exp(-1)+1e-50,epsilon(),0)-M
+	-0.00000000000000000000000000000000000000002959444084
+    ; epsilon(1e-74)
+	0.00000000000000000000000000000000000000000000000000
+    ; lambertw_series(-exp(-1)+1e-50,epsilon(),0)-M
+	-0.00000000000000000000000000000000000000000000000006
+   */
+  closeness = abs(z-branchpoint);
+  if( closeness< 1){
+    if(closeness != 0)
+      eps = epsilon(epsilon()*( closeness));
+    else
+      eps = epsilon(epsilon()^2);
+  }
+  else
+    eps = epsilon(epsilon()*1e-2);
+
+
+  epslarge =epsilon();
+
+  places = highbit(1 + int(1/epslarge)) + 1;
+  w = ret;
+  for(k=0;k<100;k++){
+    ew = exp(w);
+    we = w*ew;
+    if(abs(we-z)<= 4*epslarge*abs(z))break;
+    w1e = (1+w)*ew;
+    wn = bround(w- ((we - z) / ( w1e - ( (w+2)*(we-z) )/(2*w+2)  ) ),places++) ;
+    if( abs(wn - w) <= epslarge*abs(wn)) break;
+    else  w = wn;
+  }
+
+  if(k==100){
+    epsilon(eps);
+    return newerror("lambertw: Halley iteration does not converge");
+  }
+  /* The Maxima coders added a check if the iteration converged to the correct
+     branch. This coder deems it superfluous. */
+
+  epsilon(eps);
+  return wn;
+}
+
+
+config("resource_debug", resource_debug_level),;
+if (config("resource_debug") & 3) {
+    print "lambertw(z,branch)";
+    print "lambertw_series(z,eps,branch,terms)";
+    print "lambertw_series_print()";
+}

cal/lnseries.cal

@@ -0,0 +1,112 @@
+/*
+ * lnseries - special functions (e.g.: gamma, zeta, psi)
+ *
+ * Copyright (C) 2013 Christoph Zurnieden
+ *
+ * Calc is open software; you can redistribute it and/or modify it under
+ * the terms of the version 2.1 of the GNU Lesser General Public License
+ * as published by the Free Software Foundation.
+ *
+ * Calc is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+ * or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU Lesser General
+ * Public License for more details.
+ *
+ * A copy of version 2.1 of the GNU Lesser General Public License is
+ * distributed with calc under the filename COPYING-LGPL.  You should have
+ * received a copy with calc; if not, write to Free Software Foundation, Inc.
+ * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
+ *
+ * @(#) $Revision: 30.4 $
+ * @(#) $Id: lnseries.cal,v 30.4 2013/08/18 20:01:53 chongo Exp $
+ * @(#) $Source: /usr/local/src/bin/calc/cal/RCS/lnseries.cal,v $
+ *
+ * Under source code control:	2013/08/11 01:31:28
+ * File existed as early as:	2013
+ */
+
+
+/*
+ * hide internal function from resource debugging
+ */
+static resource_debug_level;
+resource_debug_level = config("resource_debug", 0);
+
+
+static __CZ__int_logs;
+static __CZ__int_logs_limit;
+static __CZ__int_logs_prec;
+
+
+define deletelnseries(){
+  free(__CZ__int_logs,__CZ__int_logs_limit,__CZ__int_logs_prec);
+}
+
+define lnfromseries(n){
+  if(    isnull(__CZ__int_logs)
+      || __CZ__int_logs_limit < n
+      || __CZ__int_logs_prec < log(1/epsilon())){
+
+    lnseries(n+1);
+  }
+  return __CZ__int_logs[n,0];
+}
+
+define lnseries(limit){
+  local k j eps ;
+  if(    isnull(__CZ__int_logs)
+      || __CZ__int_logs_limit < limit
+      || __CZ__int_logs_prec < log(1/epsilon())){
+    __CZ__int_logs = mat[limit+1,2];
+    __CZ__int_logs_limit = limit;
+    __CZ__int_logs_prec = log(1/epsilon());
+
+    /* probably still too much */
+    eps = epsilon(epsilon()*10^(-(5+log(limit))));
+    k =2;
+    while(1){
+      /* the prime itself, compute logarithm */
+      __CZ__int_logs[k,0] = ln(k);
+      __CZ__int_logs[k,1] = k;
+
+      for(j = 2*k;j<=limit;j+=k){
+       /* multiples of prime k, add logarithm of k computed earlier */
+       __CZ__int_logs[j,0] += __CZ__int_logs[k,0];
+       /* First hit, set counter to number */
+       if(__CZ__int_logs[j,1] ==0)
+         __CZ__int_logs[j,1]=j;
+       /* reduce counter by prime added */
+       __CZ__int_logs[j,1] //= __CZ__int_logs[k,1];
+      }
+
+      k++;
+      if(k>=limit) break;
+      /* Erastothenes-sieve: look for next prime. */
+      while(__CZ__int_logs[k,0]!=0){
+        k++;
+        if(k>=limit) break;
+      }
+    }
+    /* Second run to include the last factor */
+    for(k=1;k<=limit;k++){
+      if(__CZ__int_logs[k,1] != k){
+        __CZ__int_logs[k,0] +=__CZ__int_logs[ __CZ__int_logs[k,1],0];
+        __CZ__int_logs[k,1] = 0;
+      }
+    }
+
+    epsilon(eps);
+  }
+  return 1;
+}
+
+
+/*
+ * restore internal function from resource debugging
+ */
+config("resource_debug", resource_debug_level),;
+if (config("resource_debug") & 3) {
+    print "lnseries(limit)";
+    print "lnfromseries(n)";
+    print "deletelnseries()";
+}

cal/lucas.cal

@@ -1,7 +1,7 @@
 /*
  * lucas - perform a Lucas primality test on h*2^n-1
  *
- * Copyright (C) 1999  Landon Curt Noll
+ * Copyright (C) 1999,2017  Landon Curt Noll
  *
  * Calc is open software; you can redistribute it and/or modify it under
  * the terms of the version 2.1 of the GNU Lesser General Public License
@@ -17,9 +17,9 @@
  * received a copy with calc; if not, write to Free Software Foundation, Inc.
  * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
  *
- * @(#) $Revision: 30.1 $
- * @(#) $Id: lucas.cal,v 30.1 2007/03/16 11:09:54 chongo Exp $
- * @(#) $Source: /usr/local/src/bin/calc/cal/RCS/lucas.cal,v $
+ * @(#) $Revision: 30.4 $
+ * @(#) $Id: lucas.cal,v 30.4 2017/05/20 21:54:16 chongo Exp $
+ * @(#) $Source: /usr/local/src/bin/calc-RHEL7/cal/RCS/lucas.cal,v $
  *
  * Under source code control:	1990/05/03 16:49:51
  * File existed as early as:	1990
@@ -28,6 +28,12 @@
  * Share and enjoy!  :-)	http://www.isthe.com/chongo/tech/comp/calc/
  */
 
+/*
+ * For a general tutorial on how to find a new largest known prime, see:
+ *
+ *	http://www.isthe.com/chongo/tech/math/prime/prime-tutorial.pdf
+ */
+
 /*
  * NOTE: This is a standard calc resource file.  For information on calc see:
  *
@@ -71,10 +77,15 @@
  * NOTE: Both largest known and largest known twin prime records have been
  *	 broken.  Rather than update this file each time, I'll just
  *	 congratulate the finders and encourage others to try for
- *	 larger finds.	Records were made to be broken afterall!
+ *	 larger finds.	Records were made to be broken after all!
  */
 
-/* ON GAINING A WORLD RECORD:
+/*
+ * ON GAINING A WORLD RECORD:
+ *
+ * For a general tutorial on how to find a new largest known prime, see:
+ *
+ *	http://www.isthe.com/chongo/tech/math/prime/prime-tutorial.pdf
  *
  * The routines in calc were designed to be portable, and to work on
  * numbers of 'sane' size.  The Amdahl 6 team used a 'ultra-high speed
@@ -83,6 +94,13 @@
  * The heart of the package was a multiplication and square routine that
  * was based on the PFA Fast Fourier Transform and on Winograd's radix FFTs.
  *
+ * NOTE: While the PFA Fast Fourier Transform and Winograd's radix FFTs
+ *	 might have been optimal for the Amdahl 6 team at the time,
+ *	 they might not be optimal for your CPU architecture.  See
+ *	 the above mentioned tutorial for information on better
+ *	 methods of performing multiplications and squares of very
+ *	 large numbers.
+ *
  * Having a fast computer, and a good multi-precision package are
  * critical, but one also needs to know where to look in order to have
  * a good chance at a record.  Knowing what to test is beyond the scope
@@ -139,12 +157,10 @@
  * be the factors of another candidate.
  *
  * Finally, one should eliminate all values of 'h*2^n-1' where
- * 'h*2^n+1' is divisible by a small primes.  The ideas behind this
- * point is beyond the scope of this program.
+ * 'h*2^n+1' is divisible by a small primes.
  */
 
-
-global pprod256;	/* product of  "primes up to 256" / "primes up to 46" */
+pprod256 = 0;	/* product of  "primes up to 256" / "primes up to 46" */
 
 /*
  * lucas - lucas primality test on h*2^n-1
@@ -171,15 +187,33 @@ global pprod256;	/* product of  "primes up to 256" / "primes up to 46" */
  *	"Introduction to Analytic Number Theory", by Tom A. Apostol,
  *	Springer-Verlag, 1984, p 188.
  *
+ * An excellent 5-page paper by Oystein J. Rodseth (we apologize that the
+ * ASCII character set does not allow us to spell his name with the
+ * umlaut marks on the O's):
+ *
+ * NOTE: The original Amdahl 6 method predates the publication of Ref4.
+ *	 The gen_v1() function used by lucas() uses the Ref4 method.
+ *	 See the 'Amdahl 6 legacy code' section below for the original
+ *	 method of generating v(1).
+ *
+ * Ref4:
+ *
+ * 	"A note on primality tests for N = h*2^n-1", by Oystein J. Rodseth,
+ *	Department of Mathematics, University of Bergen, BIT Numerical
+ *	Mathematics. 34 (3): pp 451-454.
+ *
+ *	http://folk.uib.no/nmaoy/papers/luc.pdf
+ *
  * This test is performed as follows:	(see Ref1, Theorem 5)
  *
- *	a) generate u(0)		(see the function gen_u0() below)
+ *	a) generate u(2)		(see the function gen_u2() below)
+ *					(NOTE: some call this u(0))
  *
- *	b) generate u(n-2) according to the rule:
+ *	b) generate u(n) according to the rule:
  *
  *		u(i+1) = u(i)^2-2 mod h*2^n-1
  *
- *	c) h*2^n-1 is prime if and only if u(n-2) == 0		Q.E.D. :-)
+ *	c) h*2^n-1 is prime if and only if u(n) == 0		Q.E.D. :-)
  *
  * Now the following conditions must be true for the test to work:
  *
@@ -188,7 +222,7 @@ global pprod256;	/* product of  "primes up to 256" / "primes up to 46" */
  *	h < 2^n
  *	h mod 2 == 1
  *
- * A few misc notes:
+ * A few miscellaneous notes:
  *
  * In order to reduce the number of tests, as attempt to eliminate
  * any number that is divisible by a prime less than 257.  Valid prime
@@ -222,7 +256,7 @@ lucas(h, n)
 	local testval;		/* h*2^n-1 */
 	local shiftdown;	/* the power of 2 that divides h */
 	local u;		/* the u(i) sequence value */
-	local v1;		/* the v(1) generator of u(0) */
+	local v1;		/* the v(1) generator of u(2) */
 	local i;		/* u sequence cycle number */
 	local oldh;		/* pre-reduced h */
 	local oldn;		/* pre-reduced n */
@@ -364,18 +398,17 @@ lucas(h, n)
 	}
 
 	/*
-	 * try to compute u(0)
+	 * try to compute u(2)		(NOTE: some call this u(0))
 	 *
 	 * We will use gen_v1() to give us a v(1) using the values
-	 * of 'h' and 'n'.  We will then use gen_u0() to convert
-	 * the v(1) into u(0).
+	 * of 'h' and 'n'.  We will then use gen_u2() to convert
+	 * the v(1) into u(2).
 	 *
 	 * If gen_v1() returns a negative value, then we failed to
-	 * generate a test for h*2^n-1.	 This is because h mod 3 == 0
-	 * is hard to do, and in rare cases, exceed the tables found
-	 * in this program.  We will generate an message and assume
-	 * the number is not prime, even though if we had a larger
-	 * table, we might have been able to show that it is prime.
+	 * generate a test for h*2^n-1.	 The legacy function,
+	 * legacy_gen_v1() used by the Amdahl 6 could have returned
+	 * -1. The new gen_v1() based on the method outlined in Ref4
+	 * will never return -1.
 	 */
 	v1 = gen_v1(h, n);
 	if (v1 < 0) {
@@ -384,10 +417,10 @@ lucas(h, n)
 		ldebug("lucas", "unknown: no v(1)");
 		return -1;
 	}
-	u = gen_u0(h, n, v1);
+	u = gen_u2(h, n, v1);
 
 	/*
-	 * compute u(n-2)
+	 * compute u(n)			(NOTE: some call this u(n-2))
 	 */
 	for (i=3; i <= n; ++i) {
 		/* u = (u^2 - 2) % testval; */
@@ -407,11 +440,19 @@ lucas(h, n)
 }
 
 /*
- * gen_u0 - determine the initial Lucas sequence for h*2^n-1
+ * gen_u2 - determine the initial Lucas sequence for h*2^n-1
+ *
+ * Historically many start the Lucas sequence with u(0).
+ * Some, like the author of this code, prefer to start
+ * with U(2).  This is so one may say:
+ *
+ *	2^p-1 is prime if u(p) = 0 mod 2^p-1
+ * or:
+ *	h*2^p-1 is prime if u(p) = 0 mod h*2^p-1
  *
  * According to Ref1, Theorem 5:
  *
- *	u(0) = alpha^h + alpha^(-h)
+ *	u(2) = alpha^h + alpha^(-h)	(NOTE: Ref1 calls it u(0))
  *
  * Now:
  *
@@ -419,7 +460,7 @@ lucas(h, n)
  *
  * Therefore:
  *
- *	u(0) = v(h)
+ *	u(2) = v(h)			(NOTE: Ref1 calls it u(0))
  *
  * We calculate v(h) as follows:	(Ref1, top of page 873)
  *
@@ -442,16 +483,16 @@ lucas(h, n)
  * See the function gen_v1() for details on the value of v(1).
  *
  * input:
- *	h	- h as in h*2^n-1	(h mod 2 != 0)
+ *	h	- h as in h*2^n-1
  *	n	- n as in h*2^n-1
  *	v1	- gen_v1(h,n)		(see function below)
  *
  * returns:
- *	u(0)	- initial value for Lucas test on h*2^n-1
- *	-1	- failed to generate u(0)
+ *	u(2)	- initial value for Lucas test on h*2^n-1
+ *	-1	- failed to generate u(2)
  */
 define
-gen_u0(h, n, v1)
+gen_u2(h, n, v1)
 {
 	local shiftdown;	/* the power of 2 that divides h */
 	local r;		/* low value: v(n) */
@@ -475,13 +516,6 @@ gen_u0(h, n, v1)
 		quit "bogus arg: v1 is <= 0";
 	}
 
-	/*
-	 * enforce the h mod rules
-	 */
-	if (h%2 == 0) {
-		quit "h must not be even";
-	}
-
 	/*
 	 * enforce the h > 0 and n >= 2 rules
 	 */
@@ -507,7 +541,7 @@ gen_u0(h, n, v1)
 	 * at least 2 bits long for the loop below to work.
 	 */
 	if (h == 1) {
-		ldebug("gen_u0", "quick h == 1 case");
+		ldebug("gen_u2", "quick h == 1 case");
 		/* return r%(h*2^n-1); */
 		return hnrmod(r, h, n, -1);
 	}
@@ -547,21 +581,502 @@ gen_u0(h, n, v1)
 	return r;
 }
 
+/*
+ * gen_u0 - determine the initial Lucas sequence for h*2^n-1
+ *
+ * Historically many start the Lucas sequence with u(0).
+ * Some, like the author of this code, prefer to start
+ * with u(2).  This is so one may say:
+ *
+ *	2^p-1 is prime if u(p) = 0 mod 2^p-1
+ * or:
+ *	h*2^n-1 is prime if U(n) = 0 mod h*2^n-1
+ *
+ * For those using the old code with gen_u0(), we
+ * simply call gen_u2() instead.
+ *
+ * See the function gen_u2() for details.
+ *
+ * input:
+ *	h	- h as in h*2^n-1
+ *	n	- n as in h*2^n-1
+ *	v1	- gen_v1(h,n)		(see function below)
+ *
+ * returns:
+ *	u(2)	- initial value for Lucas test on h*2^n-1
+ *	-1	- failed to generate u(2)
+ */
+define
+gen_u0(h, n, v1)
+{
+	return gen_u2(h, n, v1);
+}
+
+/*
+ * rodseth_xhn - determine if v(1) == x for h*2^n-1
+ *
+ * For a given h*2^n-1, v(1) == x if:
+ *
+ *	jacobi(x-2, h*2^n-1) == 1		(Ref4, condition 1) part 1
+ *	jacobi(x+2, h*2^n-1) == -1		(Ref4, condition 1) part 2
+ *
+ * Now when x-2 <= 0:
+ *
+ *	jacobi(x-2, h*2^n-1) == 0
+ *
+ * because:
+ *
+ *	jacobi(x,y) == 0                        if x <= 0
+ *
+ * So for (Ref4, condition 1) part 1 to be true:
+ *
+ *	x-2 > 0
+ *
+ * And therefore:
+ *
+ *	x > 2
+ *
+ * input:
+ *	x	- potential v(1) value
+ *	h	- h as in h*2^n-1
+ *	n	- n as in h*2^n-1
+ *
+ * returns:
+ *	1	if v(1) == x for h*2^n-1
+ *	0	otherwise
+ */
+define
+rodseth_xhn(x, h, n)
+{
+	local testval;		/* h*2^n-1 */
+
+	/*
+	 * check arg types
+	 */
+	if (!isint(h)) {
+		quit "bad args: h must be an integer";
+	}
+	if (!isint(n)) {
+		quit "bad args: n must be an integer";
+	}
+	if (!isint(x)) {
+		quit "bad args: x must be an integer";
+	}
+
+	/*
+	 * firewall
+	 */
+	if (x <= 2) {
+		return 0;
+	}
+
+ 	/*
+	 * Check for jacobi(x-2, h*2^n-1) == 1	(Ref4, condition 1) part 1
+	 */
+	testval = h*2^n-1;
+	if (jacobi(x-2, testval) != 1) {
+		return 0;
+	}
+
+ 	/*
+	 * Check for jacobi(x+2, h*2^n-1) == -1	(Ref4, condition 1) part 2
+	 */
+	if (jacobi(x+2, testval) != -1) {
+		return 0;
+	}
+
+	/*
+	 * v(1) == x for this h*2^n-1
+	 */
+	return 1;
+}
+
 /*
  * Trial tables used by gen_v1()
  *
- * When h mod 3 == 0, one needs particular values of D, a and b (see gen_v1
- * documentation) in order to find a value of v(1).
+ * When h mod 3 == 0, according to Ref4 we need to find the first value X where:
  *
- * This table defines 'quickmax' possible tests to be taken in ascending
- * order.  The v1_qval[x] refers to a v(1) value from Ref1, Table 1.  A
- * related D value is found in d_qval[x].  All D values expect d_qval[1]
- * are also taken from Ref1, Table 1.  The case of D == 21 as listed in
- * Ref1, Table 1 can be changed to D == 7 for the sake of the test because
- * of {note 6}.
+ *	jacobi(X-2, h*2^n-1) == 1		(Ref4, condition 1) part 1
+ *	jacobi(X+2, h*2^n-1) == -1		(Ref4, condition 1) part 2
+ *
+ * We can show that X > 2.  See the comments in the rodseth_xhn(x,h,n) above.
+ *
+ * Some values of X satisfy more often than others. For example a large sample
+ * of odd h, h multiple of 3 and large n (some around 1e4, some near 1e6, others
+ * near 3e7) where the sample size was 66 973 365, here is the count of the
+ * smallest value of X that satisfies conditions in Ref4, condition 1:
+ *
+ *      count  X
+ *    ----------
+ *    26791345 3
+ *    17223016 5
+ *     7829600 9
+ *     6988774 11
+ *     3301093 15
+ *     1517149 17
+ *      910346 21
+ *      711791 29
+ *      573403 20
+ *      390395 27
+ *      288637 35
+ *      149751 36
+ *      107733 39
+ *       58743 41
+ *       35619 45
+ *       25052 32
+ *       17775 51
+ *       13031 44
+ *        7563 56
+ *        7540 49
+ *        7060 59
+ *        4407 57
+ *        2948 65
+ *        2502 55
+ *        2388 69
+ *        2094 71
+ *         689 77
+ *         626 81
+ *         491 66
+ *         426 95
+ *         219 80
+ *         203 67
+ *         185 84
+ *         152 99
+ *         127 72
+ *         102 74
+ *          98 87
+ *          67 90
+ *          55 104
+ *          48 101
+ *          32 105
+ *          17 109
+ *          16 116
+ *          15 111
+ *          13 92
+ *          12 125
+ *           7 129
+ *           3 146
+ *           2 140
+ *           2 120
+ *           1 165
+ *           1 161
+ *           1 155
+ *
+ * The above distribution was found to hold fairly well over many values of
+ * odd h that are a multiple of 3 and for many values of n where h < 2^n.
+ *
+ * Given this information, when odd h is a multiple of 3 we try, in order,
+ * these values of X:
+ *
+ *      3, 5, 9, 11, 15, 17, 21, 29, 20, 27, 35, 36, 39, 41, 45, 32, 51, 44,
+ *      56, 49, 59, 57, 65, 55, 69, 71, 77, 81, 66, 95, 80, 67, 84, 99, 72,
+ *      74, 87, 90, 104, 101, 105, 109, 116, 111, 92
+ *
+ * And stop on the first value of X where:
+ *
+ *      jacobi(X-2, h*2^n-1) == 1
+ *      jacobi(X+2, h*2^n-1) == -1
+ *
+ * If no value in that list works, we start simple search starting with X = 120
+ * and incrementing by 1 until a value of X is found.
+ *
+ * The x_tbl[] matrix contains those common values of X to try in order.
+ * If all x_tbl_len fail to satisfy Ref4 condition 1, then we begin a
+ * linear search at next_x until we find a proper X value.
+ *
+ * IMPORTANT NOTE: Using this table will not find the smallest possible v(1)
+ *		   for a given h and n.  This is not a problem because using
+ *		   a larger value of v(1) does not impact the primality test.
+ *		   Furthermore after lucas(h, n) generates a few u(n) terms,
+ *		   the values will wrap (due to computing mod h*2^n-1).
+ *		   Finally on average, about 1/4 of the values of X work as
+ *		   v(1) for a given n when h is a multiple of 3.  Skipping
+ *		   rarely used v(1) will not doom gen_v1() to a long search.
+ */
+x_tbl_len = 45;
+mat x_tbl[x_tbl_len];
+x_tbl = {
+    3, 5, 9, 11, 15, 17, 21, 29, 20, 27, 35, 36, 39, 41, 45, 32, 51, 44,
+   56, 49, 59, 57, 65, 55, 69, 71, 77, 81, 66, 95, 80, 67, 84, 99, 72,
+   74, 87, 90, 104, 101, 105, 109, 116, 111, 92
+};
+next_x = 120;
+
+/*
+ * gen_v1 - compute the v(1) for a given h*2^n-1 if we can
+ *
+ * This function assumes:
+ *
+ *	n > 2			(n==2 has already been eliminated)
+ *	h mod 2 == 1
+ *	h < 2^n
+ *	h*2^n-1 mod 3 != 0	(h*2^n-1 has no small factors, such as 3)
+ *
+ * The generation of v(1) depends on the value of h.  There are two cases
+ * to consider, h mod 3 != 0, and h mod 3 == 0.
+ *
+ ***
+ *
+ * Case 1:	(h mod 3 != 0)
+ *
+ * This case is easy.
+ *
+ * In Ref1, page 869, one finds that if:	(or see Ref2, page 131-132)
+ *
+ *	h mod 6 == +/-1
+ *	h*2^n-1 mod 3 != 0
+ *
+ * which translates, gives the functions assumptions, into the condition:
+ *
+ *	h mod 3 != 0
+ *
+ * If this case condition is true, then:
+ *
+ *	u(2) = (2+sqrt(3))^h + (2-sqrt(3))^h	 (see Ref1, page 869)
+ *	     = (2+sqrt(3))^h + (2+sqrt(3))^(-h)	 (NOTE: some call this u(2))
+ *
+ * and since Ref1, Theorem 5 states:
+ *
+ *	u(2) = alpha^h + alpha^(-h)		 (NOTE: some call this u(2))
+ *	r = abs(2^2 - 1^2*3) = 1
+ *
+ * and the bottom of Ref1, page 872 states:
+ *
+ *	v(x) = alpha^x + alpha^(-x)
+ *
+ * If we let:
+ *
+ *	alpha = (2+sqrt(3))
+ *
+ * then
+ *
+ *	u(2) = v(h)				 (NOTE: some call this u(2))
+ *
+ * so we simply return
+ *
+ *	v(1) = alpha^1 + alpha^(-1)
+ *	     = (2+sqrt(3)) + (2-sqrt(3))
+ *	     = 4
+ *
+ ***
+ *
+ * Case 2:	(h mod 3 == 0)
+ *
+ * For the case where h is a multiple of 3, we turn to Ref4.
+ *
+ * The central theorem on page 3 of that paper states that
+ * we may set v(1) to the first value X that satisfies:
+ *
+ *	jacobi(X-2, h*2^n-1) == 1		(Ref4, condition 1)
+ *	jacobi(X+2, h*2^n-1) == -1		(Ref4, condition 1)
+ *
+ *	NOTE: Ref4 uses P, which we shall refer to as X.
+ *	      Ref4 uses N, which we shall refer to as h*2^n-1.
+ *
+ *	NOTE: Ref4 uses the term Legendre-Jacobi symbol, which
+ *	      we shall refer to as the Jacobi symbol.
+ *
+ * Before we address the two conditions, we need some background information
+ * on two symbols, Legendre and Jacobi.	 In Ref 2, pp 278, 284-285, we find
+ * the following definitions of jacobi(a,b) and L(a,p):
+ *
+ * The Legendre symbol L(a,p) takes the value:
+ *
+ *	L(a,p) == 1	=> a is a quadratic residue of p
+ *	L(a,p) == -1	=> a is NOT a quadratic residue of p
+ *
+ * when:
+ *
+ *	p is prime
+ *	p mod 2 == 1
+ *	gcd(a,p) == 1
+ *
+ * The value a is a quadratic residue of b if there exists some integer z
+ * such that:
+ *
+ *	z^2 mod b == a
+ *
+ * The Jacobi symbol jacobi(a,b) takes the value:
+ *
+ *	jacobi(a,b) == 1	=> b is not prime,
+ *				   or a is a quadratic residue of b
+ *	jacobi(a,b) == -1	=> a is NOT a quadratic residue of b
+ *
+ * when
+ *
+ *	b mod 2 == 1
+ *	gcd(a,b) == 1
+ *
+ * It is worth noting for the Legendre symbol, in order for L(X+/-2,
+ * h*2^n-1) to be defined, we must ensure that neither X-2 nor X+2 are
+ * factors of h*2^n-1.  This is done by pre-screening h*2^n-1 to not
+ * have small factors and keeping X+2 less than that small factor
+ * limit.  It is worth noting that in lucas(h, n), we first verify
+ * that h*2^n-1 does not have a factor < 257 before performing the
+ * primality test.  So while X+/-2 < 257, we know that
+ * gcd(X+/-2, h*2^n-1) == 1.
+ *
+ * Returning to the testing of conditions in Ref4, condition 1:
+ *
+ *	jacobi(X-2, h*2^n-1) == 1
+ *	jacobi(X+2, h*2^n-1) == -1
+ *
+ * When such an X is found, we set:
+ *
+ *	v(1) = X
+ *
+ ***
+ *
+ * In conclusion, we can compute v,(1) by attempting to do the following:
+ *
+ * h mod 3 != 0
+ *
+ *     we return:
+ *
+ *	   v(1) == 4
+ *
+ * h mod 3 == 0
+ *
+ *     we return:
+ *
+ *	    v(1) = X
+ *
+ *     where X > 2 in a integer such that:
+ *
+ *	    jacobi(X-2, h*2^n-1) == 1
+ *	    jacobi(X+2, h*2^n-1) == -1
+ *
+ ***
+ *
+ * input:
+ *	h	h as in h*2^n-1
+ *	n	n as in h*2^n-1
+ *
+ * output:
+ *	returns v(1), or -1 is there is no quick way
+ */
+define
+gen_v1(h, n)
+{
+	local x;	/* potential v(1) to test */
+	local i;	/* x_tbl index */
+
+	/*
+	 * check arg types
+	 */
+	if (!isint(h)) {
+		quit "bad args: h must be an integer";
+	}
+	if (!isint(n)) {
+		quit "bad args: n must be an integer";
+	}
+
+	/*
+	 * check for Case 1:      (h mod 3 != 0)
+	 */
+	if (h % 3 != 0) {
+		/* v(1) is easy to compute */
+		return 4;
+	}
+
+	/*
+	 * What follow is Case 2:      (h mod 3 == 0)
+	 */
+
+	/*
+	 * We will look for x that satisfies conditions in Ref4, condition 1:
+	 *
+	 *	jacobi(X-2, h*2^n-1) == 1		part 1
+	 *	jacobi(X+2, h*2^n-1) == -1		part 2
+	 */
+	for (i=0; i < x_tbl_len; ++i) {
+
+		/*
+		 * test Ref4 condition 1:
+		 */
+		x = x_tbl[i];
+		if (rodseth_xhn(x, h, n) == 1) {
+
+			/*
+			 * found a x that satisfies Ref4 condition 1
+			 */
+			ldebug("gen_v1", "h= " + str(h) + " n= " + str(n) +
+				" v1= " + str(x) + " using tbl[ " +
+				str(i) + " ]");
+			return x;
+		}
+	}
+
+	/*
+	 * We are in that rare case (about 1 in 2 300 000) where none of the
+	 * common X values satisfy Ref4 condition 1.  We start a linear search
+	 * at next_x from here on.
+	 *
+	 * However, we also need to keep in mind that when x+2 >= 257, we
+	 * need to verify that gcd(x-2, h*2^n-1) == 1 and
+	 * and to verify that gcd(x+2, h*2^n-1) == 1.
+	 */
+	x = next_x;
+	while (rodseth_xhn(x, h, n) != 1) {
+		++x;
+	}
+	/* finally found a v(1) value */
+	ldebug("gen_v1", "h= " + str(h) + " n= " + str(n) +
+	       " v1= " + str(x) + " beyond tbl");
+	return x;
+}
+
+/*
+ * ldebug - print a debug statement
+ *
+ * input:
+ *	funct	name of calling function
+ *	str	string to print
+ */
+define
+ldebug(funct, str)
+{
+	if (config("resource_debug") & 8) {
+		print "DEBUG:", funct:":", str;
+	}
+	return;
+}
+
+/*
+ ************************
+ * Amdahl 6 legacy code *
+ ************************
+ *
+ * NOTE: What follows is legacy code based on the method used by the
+ *	 Amdahl 6 group:
+ *
+ *	    John Brown, Landon Curt Noll, Bodo Parady, Gene Smith,
+ *	    Joel Smith and Sergio Zarantonello
+ *
+ * This method generated v(1) for nearly all values, except for a
+ * few rare cases when h mod 3 == 0.  The code is NOT used by lucas.cal
+ * above.  The gen_v1() function above is based on an improved method
+ * outlined in Ref4.  That method generated v(1) for all h.
+ *
+ * The code below is kept for historical purposes only.  The functions
+ * and global variables of the Amdahl 6 legacy code all begin with legacy_.
+ */
+
+/*
+ * Trial tables used by legacy_gen_v1()
+ *
+ * When h mod 3 == 0, one needs particular values of D, a and b (see
+ * legacy_gen_v1 documentation) in order to find a value of v(1).
+ *
+ * This table defines 'legacy_quickmax' possible tests to be taken in ascending
+ * order.  The legacy_v1_qval[x] refers to a v(1) value from Ref1, Table 1.  A
+ * related D value is found in legacy_d_qval[x].  All D values expect
+ * legacy_d_qval[1] are also taken from Ref1, Table 1.  The case of D == 21 as
+ * listed in Ref1, Table 1 can be changed to D == 7 for the sake of the test
+ * because of {note 6}.
  *
  * It should be noted that the D values all satisfy the selection values
- * as outlined in the gen_v1() function comments.  That is:
+ * as outlined in the legacy_gen_v1() function comments.  That is:
  *
  *	   D == P*(2^f)*(3^g)
  *
@@ -578,20 +1093,20 @@ gen_u0(h, n, v1)
  * where Q == 1.  No further processing is needed to compute v(1) when r
  * is of this form.
  */
-quickmax = 8;
-mat d_qval[quickmax];
-mat v1_qval[quickmax];
-d_qval[0] = 5;		v1_qval[0] = 3;		/* a=1	 b=1  r=4  */
-d_qval[1] = 7;		v1_qval[1] = 5;		/* a=3	 b=1  r=12  D=21 */
-d_qval[2] = 13;		v1_qval[2] = 11;	/* a=3	 b=1  r=4  */
-d_qval[3] = 11;		v1_qval[3] = 20;	/* a=3	 b=1  r=2  */
-d_qval[4] = 29;		v1_qval[4] = 27;	/* a=5	 b=1  r=4  */
-d_qval[5] = 53;		v1_qval[5] = 51;	/* a=53	 b=1  r=4  */
-d_qval[6] = 17;		v1_qval[6] = 66;	/* a=17	 b=1  r=1  */
-d_qval[7] = 19;		v1_qval[7] = 74;	/* a=38	 b=1  r=2  */
+legacy_quickmax = 8;
+mat legacy_d_qval[legacy_quickmax];
+mat legacy_v1_qval[legacy_quickmax];
+legacy_d_qval[0] = 5;	legacy_v1_qval[0] = 3;	/* a=1	 b=1  r=4  */
+legacy_d_qval[1] = 7;	legacy_v1_qval[1] = 5;	/* a=3	 b=1  r=12  D=21 */
+legacy_d_qval[2] = 13;	legacy_v1_qval[2] = 11;	/* a=3	 b=1  r=4  */
+legacy_d_qval[3] = 11;	legacy_v1_qval[3] = 20;	/* a=3	 b=1  r=2  */
+legacy_d_qval[4] = 29;	legacy_v1_qval[4] = 27;	/* a=5	 b=1  r=4  */
+legacy_d_qval[5] = 53;	legacy_v1_qval[5] = 51;	/* a=53	 b=1  r=4  */
+legacy_d_qval[6] = 17;	legacy_v1_qval[6] = 66;	/* a=17	 b=1  r=1  */
+legacy_d_qval[7] = 19;	legacy_v1_qval[7] = 74;	/* a=38	 b=1  r=2  */
 
 /*
- * gen_v1 - compute the v(1) for a given h*2^n-1 if we can
+ * legacy_gen_v1 - compute the v(1) for a given h*2^n-1 if we can
  *
  * This function assumes:
  *
@@ -620,12 +1135,12 @@ d_qval[7] = 19;		v1_qval[7] = 74;	/* a=38	 b=1  r=2  */
  *
  * If this case condition is true, then:
  *
- *	u(0) = (2+sqrt(3))^h + (2-sqrt(3))^h		(see Ref1, page 869)
- *	     = (2+sqrt(3))^h + (2+sqrt(3))^(-h)
+ *	u(2) = (2+sqrt(3))^h + (2-sqrt(3))^h		(see Ref1, page 869)
+ *	     = (2+sqrt(3))^h + (2+sqrt(3))^(-h)		(some call this u(0))
  *
  * and since Ref1, Theorem 5 states:
  *
- *	u(0) = alpha^h + alpha^(-h)
+ *	u(2) = alpha^h + alpha^(-h)
  *	r = abs(2^2 - 1^2*3) = 1
  *
  * and the bottom of Ref1, page 872 states:
@@ -638,7 +1153,7 @@ d_qval[7] = 19;		v1_qval[7] = 74;	/* a=38	 b=1  r=2  */
  *
  * then
  *
- *	u(0) = v(h)
+ *	u(2) = v(h)
  *
  * so we simply return
  *
@@ -673,7 +1188,7 @@ d_qval[7] = 19;		v1_qval[7] = 74;	/* a=38	 b=1  r=2  */
  *
  * where L(x,y) is the Legendre symbol (see below), then:
  *
- *	u(0) = alpha^h + alpha^(-h)
+ *	u(2) = alpha^h + alpha^(-h)
  *
  * The bottom of Ref1, page 872 states:
  *
@@ -681,7 +1196,7 @@ d_qval[7] = 19;		v1_qval[7] = 74;	/* a=38	 b=1  r=2  */
  *
  * thus since:
  *
- *	u(0) = v(h)
+ *	u(2) = v(h)
  *
  * so we want to return:
  *
@@ -936,7 +1451,7 @@ d_qval[7] = 19;		v1_qval[7] = 74;	/* a=38	 b=1  r=2  */
  *	returns v(1), or -1 is there is no quick way
  */
 define
-gen_v1(h, n)
+legacy_gen_v1(h, n)
 {
 	local d;	/* the 'D' value to try */
 	local val_mod;	/* h*2^n-1 mod 'D' */
@@ -954,10 +1469,10 @@ gen_v1(h, n)
 	 * We will try all 'D' values until we find a proper v(1)
 	 * or run out of 'D' values.
 	 */
-	for (i=0; i < quickmax; ++i) {
+	for (i=0; i < legacy_quickmax; ++i) {
 
 		/* grab our 'D' value */
-		d = d_qval[i];
+		d = legacy_d_qval[i];
 
 		/* compute h*2^n-1 mod 'D' quickly */
 		val_mod = (h*pmod(2,n%(d-1),d)-1) % d;
@@ -972,13 +1487,13 @@ gen_v1(h, n)
 			/* D mod 4 == 1, so check for J(D, h*2^n-1) == -1 */
 			if (jacobi(val_mod, d) == -1) {
 				/* it worked, return the related v(1) value */
-				return v1_qval[i];
+				return legacy_v1_qval[i];
 			}
 		} else {
 			/* D mod 4 == -1, so check for J(D, h*2^n-1) == 1 */
 			if (jacobi(val_mod, d) == 1) {
 				/* it worked, return the related v(1) value */
-				return v1_qval[i];
+				return legacy_v1_qval[i];
 			}
 		}
 	}
@@ -1036,19 +1551,3 @@ gen_v1(h, n)
 	/* no quick and dirty v(1), so return -1 */
 	return -1;
 }
-
-/*
- * ldebug - print a debug statement
- *
- * input:
- *	funct	name of calling function
- *	str	string to print
- */
-define
-ldebug(funct, str)
-{
-	if (config("resource_debug") & 8) {
-		print "DEBUG:", funct:":", str;
-	}
-	return;
-}

cal/lucas_tbl.cal

@@ -1,165 +0,0 @@
-/*
- * lucas_tbl - lucasian criteria for primality tables
- *
- * Copyright (C) 1999  Landon Curt Noll
- *
- * Calc is open software; you can redistribute it and/or modify it under
- * the terms of the version 2.1 of the GNU Lesser General Public License
- * as published by the Free Software Foundation.
- *
- * Calc is distributed in the hope that it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
- * or FITNESS FOR A PARTICULAR PURPOSE.	 See the GNU Lesser General
- * Public License for more details.
- *
- * A copy of version 2.1 of the GNU Lesser General Public License is
- * distributed with calc under the filename COPYING-LGPL.  You should have
- * received a copy with calc; if not, write to Free Software Foundation, Inc.
- * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
- *
- * @(#) $Revision: 30.1 $
- * @(#) $Id: lucas_tbl.cal,v 30.1 2007/03/16 11:09:54 chongo Exp $
- * @(#) $Source: /usr/local/src/bin/calc/cal/RCS/lucas_tbl.cal,v $
- *
- * Under source code control:	1991/01/26 02:43:43
- * File existed as early as:	1991
- *
- * chongo <was here> /\oo/\	http://www.isthe.com/chongo/
- * Share and enjoy!  :-)	http://www.isthe.com/chongo/tech/comp/calc/
- */
-
-/*
- * Lucasian criteria for primality
- *
- * The following table is taken from:
- *
- *	"Lucasian Criteria for the Primality of N=h*2^n-1", by Hans Riesel,
- *	Mathematics of Computation, Vol 23 #108, p 872.
- *
- * The index of the *_val[] arrays correspond to the v(1) values found
- * in the table.  That is, for v(1) == x:
- *
- *	D == d_val[x]
- *	a == a_val[x]
- *	b == b_val[x]
- *	r == r_val[x]		(r == abs(a^2 - b^2*D))
- *
- *
- * Note that when *_val[i] is not a number, the related v(1) value
- * is not found in Table 1.
- */
-
-
-trymax = 100;
-mat d_val[trymax+1];
-mat a_val[trymax+1];
-mat b_val[trymax+1];
-mat r_val[trymax+1];
-/* v1= 0	    INVALID */
-/* v1= 1	    INVALID */
-/* v1= 2	    INVALID */
-d_val[ 3]=   5;	 a_val[ 3]= 1;	b_val[ 3]=1;  r_val[ 3]=4;
-d_val[ 4]=   3;	 a_val[ 4]= 1;	b_val[ 4]=1;  r_val[ 4]=2;
-d_val[ 5]=  21;	 a_val[ 5]= 3;	b_val[ 5]=1;  r_val[ 5]=12;
-d_val[ 6]=   2;	 a_val[ 6]= 1;	b_val[ 6]=1;  r_val[ 6]=1;
-/* v1= 7	    INVALID */
-d_val[ 8]=  15;	 a_val[ 8]= 3;	b_val[ 8]=1;  r_val[ 8]=6;
-d_val[ 9]=  77;	 a_val[ 9]= 7;	b_val[ 9]=1;  r_val[ 9]=28;
-d_val[10]=   6;	 a_val[10]= 2;	b_val[10]=1;  r_val[10]=2;
-d_val[11]=  13;	 a_val[11]= 3;	b_val[11]=1;  r_val[11]=4;
-d_val[12]=  35;	 a_val[12]= 5;	b_val[12]=1;  r_val[12]=10;
-d_val[13]= 165;	 a_val[13]=11;	b_val[13]=1;  r_val[13]=44;
-/* v1=14	    INVALID */
-d_val[15]= 221;	 a_val[15]=13;	b_val[15]=1;  r_val[15]=52;
-d_val[16]=   7;	 a_val[16]= 3;	b_val[16]=1;  r_val[16]=2;
-d_val[17]= 285;	 a_val[17]=15;	b_val[17]=1;  r_val[17]=60;
-/* v1=18	    INVALID */
-d_val[19]= 357;	 a_val[19]=17;	b_val[19]=1;  r_val[19]=68;
-d_val[20]=  11;	 a_val[20]= 3;	b_val[20]=1;  r_val[20]=2;
-d_val[21]= 437;	 a_val[21]=19;	b_val[21]=1;  r_val[21]=76;
-d_val[22]=  30;	 a_val[22]= 5;	b_val[22]=1;  r_val[22]=5;
-/* v1=23	    INVALID */
-d_val[24]= 143;	 a_val[24]=11;	b_val[24]=1;  r_val[24]=22;
-d_val[25]=  69;	 a_val[25]= 9;	b_val[25]=1;  r_val[25]=12;
-d_val[26]=  42;	 a_val[26]= 6;	b_val[26]=1;  r_val[26]=6;
-d_val[27]=  29;	 a_val[27]= 5;	b_val[27]=1;  r_val[27]=4;
-d_val[28]= 195;	 a_val[28]=13;	b_val[28]=1;  r_val[28]=26;
-d_val[29]=  93;	 a_val[29]= 9;	b_val[29]=1;  r_val[29]=12;
-d_val[30]=  14;	 a_val[30]= 4;	b_val[30]=1;  r_val[30]=2;
-d_val[31]= 957;	 a_val[31]=29;	b_val[31]=1;  r_val[31]=116;
-d_val[32]= 255;	 a_val[32]=15;	b_val[32]=1;  r_val[32]=30;
-d_val[33]=1085;	 a_val[33]=31;	b_val[33]=1;  r_val[33]=124;
-/* v1=34	    INVALID */
-d_val[35]=1221;	 a_val[35]=33;	b_val[35]=1;  r_val[35]=132;
-d_val[36]= 323;	 a_val[36]=17;	b_val[36]=1;  r_val[36]=34;
-d_val[37]=1365;	 a_val[37]=35;	b_val[37]=1;  r_val[37]=140;
-d_val[38]=  10;	 a_val[38]= 3;	b_val[38]=1;  r_val[38]=1;
-d_val[39]=1517;	 a_val[39]=37;	b_val[39]=1;  r_val[39]=148;
-d_val[40]= 399;	 a_val[40]=19;	b_val[40]=1;  r_val[40]=38;
-d_val[41]=1677;	 a_val[41]=39;	b_val[41]=1;  r_val[41]=156;
-d_val[42]= 110;	 a_val[42]=10;	b_val[42]=1;  r_val[42]=10;
-d_val[43]= 205;	 a_val[43]=15;	b_val[43]=1;  r_val[43]=20;
-d_val[44]= 483;	 a_val[44]=21;	b_val[44]=1;  r_val[44]=42;
-d_val[45]=2021;	 a_val[45]=43;	b_val[45]=1;  r_val[45]=172;
-d_val[46]=  33;	 a_val[46]= 6;	b_val[46]=1;  r_val[46]=3;
-/* v1=47	    INVALID */
-d_val[48]=  23;	 a_val[48]= 5;	b_val[48]=1;  r_val[48]=2;
-d_val[49]=2397;	 a_val[49]=47;	b_val[49]=1;  r_val[49]=188;
-d_val[50]=  39;	 a_val[50]= 6;	b_val[50]=1;  r_val[50]=3;
-d_val[51]=  53;	 a_val[51]= 7;	b_val[51]=1;  r_val[51]=4;
-/* v1=52	    INVALID */
-d_val[53]=2805;	 a_val[53]=51;	b_val[53]=1;  r_val[53]=204;
-d_val[54]= 182;	 a_val[54]=13;	b_val[54]=1;  r_val[54]=13;
-d_val[55]=3021;	 a_val[55]=53;	b_val[55]=1;  r_val[55]=212;
-d_val[56]=  87;	 a_val[56]= 9;	b_val[56]=1;  r_val[56]=6;
-d_val[57]=3245;	 a_val[57]=55;	b_val[57]=1;  r_val[57]=220;
-d_val[58]= 210;	 a_val[58]=14;	b_val[58]=1;  r_val[58]=14;
-d_val[59]=3477;	 a_val[59]=57;	b_val[59]=1;  r_val[59]=228;
-d_val[60]= 899;	 a_val[60]=29;	b_val[60]=1;  r_val[60]=58;
-d_val[61]= 413;	 a_val[61]=21;	b_val[61]=1;  r_val[61]=28;
-/* v1=62	    INVALID */
-d_val[63]=3965;	 a_val[63]=61;	b_val[63]=1;  r_val[63]=244;
-d_val[64]=1023;	 a_val[64]=31;	b_val[64]=1;  r_val[64]=62;
-d_val[65]= 469;	 a_val[65]=21;	b_val[65]=1;  r_val[65]=28;
-d_val[66]=  17;	 a_val[66]= 4;	b_val[66]=1;  r_val[66]=1;
-d_val[67]=4485;	 a_val[67]=65;	b_val[67]=1;  r_val[67]=260;
-d_val[68]=1155;	 a_val[68]=33;	b_val[68]=1;  r_val[68]=66;
-d_val[69]=4757;	 a_val[69]=67;	b_val[69]=1;  r_val[69]=268;
-d_val[70]=  34;	 a_val[70]= 6;	b_val[70]=1;  r_val[70]=2;
-d_val[71]=5037;	 a_val[71]=69;	b_val[71]=1;  r_val[71]=276;
-d_val[72]=1295;	 a_val[72]=35;	b_val[72]=1;  r_val[72]=70;
-d_val[73]= 213;	 a_val[73]=15;	b_val[73]=1;  r_val[73]=12;
-d_val[74]=  38;	 a_val[74]= 6;	b_val[74]=1;  r_val[74]=2;
-d_val[75]=5621;	 a_val[75]=73;	b_val[75]=1;  r_val[75]=292;
-d_val[76]=1443;	 a_val[76]=37;	b_val[76]=1;  r_val[76]=74;
-d_val[77]= 237;	 a_val[77]=15;	b_val[77]=1;  r_val[77]=12;
-d_val[78]=  95;	 a_val[78]=10;	b_val[78]=1;  r_val[78]=5;
-/* v1=79	    INVALID */
-d_val[80]=1599;	 a_val[80]=39;	b_val[80]=1;  r_val[80]=78;
-d_val[81]=6557;	 a_val[81]=79;	b_val[81]=1;  r_val[81]=316;
-d_val[82]= 105;	 a_val[82]=10;	b_val[82]=1;  r_val[82]=5;
-d_val[83]=  85;	 a_val[83]= 9;	b_val[83]=1;  r_val[83]=4;
-d_val[84]=1763;	 a_val[84]=41;	b_val[84]=1;  r_val[84]=82;
-d_val[85]=7221;	 a_val[85]=83;	b_val[85]=1;  r_val[85]=332;
-d_val[86]= 462;	 a_val[86]=21;	b_val[86]=1;  r_val[86]=21;
-d_val[87]=7565;	 a_val[87]=85;	b_val[87]=1;  r_val[87]=340;
-d_val[88]= 215;	 a_val[88]=15;	b_val[88]=1;  r_val[88]=10;
-d_val[89]=7917;	 a_val[89]=87;	b_val[89]=1;  r_val[89]=348;
-d_val[90]= 506;	 a_val[90]=22;	b_val[90]=1;  r_val[90]=22;
-d_val[91]=8277;	 a_val[91]=89;	b_val[91]=1;  r_val[91]=356;
-d_val[92]= 235;	 a_val[92]=15;	b_val[92]=1;  r_val[92]=10;
-d_val[93]=8645;	 a_val[93]=91;	b_val[93]=1;  r_val[93]=364;
-d_val[94]= 138;	 a_val[94]=12;	b_val[94]=1;  r_val[94]=6;
-d_val[95]=9021;	 a_val[95]=93;	b_val[95]=1;  r_val[95]=372;
-d_val[96]=  47;	 a_val[96]= 7;	b_val[96]=1;  r_val[96]=2;
-d_val[97]=1045;	 a_val[97]=33;	b_val[97]=1;  r_val[97]=44;
-/* v1=98	    INVALID */
-d_val[99]=9797;	 a_val[99]=97;	b_val[99]=1;  r_val[99]=388;
-d_val[100]=  51; a_val[100]= 7; b_val[100]=1; r_val[100]=2;
-
-if (config("resource_debug") & 3) {
-    print "d_val[100] defined";
-    print "a_val[100] defined";
-    print "b_val[100] defined";
-    print "r_val[100] defined";
-}

cal/quat.cal

@@ -17,8 +17,8 @@
  * received a copy with calc; if not, write to Free Software Foundation, Inc.
  * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
  *
- * @(#) $Revision: 30.1 $
- * @(#) $Id: quat.cal,v 30.1 2007/03/16 11:09:54 chongo Exp $
+ * @(#) $Revision: 30.2 $
+ * @(#) $Id: quat.cal,v 30.2 2013/08/11 08:41:38 chongo Exp $
  * @(#) $Source: /usr/local/src/bin/calc/cal/RCS/quat.cal,v $
  *
  * Under source code control:	1990/02/15 01:50:35
@@ -55,7 +55,8 @@ define quat(a,b,c,d)
 
 define quat_print(a)
 {
-	print "quat(" : a.s : ", " : a.v[0] : ", " : a.v[1] : ", " : a.v[2] : ")" :;
+	print "quat(" : a.s : ", " : a.v[0] : ", " :
+	      a.v[1] : ", " : a.v[2] : ")" :;
 }
 
 

cal/regress.cal

@@ -1,7 +1,7 @@
 /*
  * regress - calc regression and correctness test suite
  *
- * Copyright (C) 1999-2006  David I. Bell and Landon Curt Noll
+ * Copyright (C) 1999-2017  David I. Bell and Landon Curt Noll
  *
  * Calc is open software; you can redistribute it and/or modify it under
  * the terms of the version 2.1 of the GNU Lesser General Public License
@@ -17,9 +17,9 @@
  * received a copy with calc; if not, write to Free Software Foundation, Inc.
  * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
  *
- * @(#) $Revision: 30.6 $
- * @(#) $Id: regress.cal,v 30.6 2010/09/02 06:09:06 chongo Exp $
- * @(#) $Source: /usr/local/src/bin/calc/cal/RCS/regress.cal,v $
+ * @(#) $Revision: 30.14 $
+ * @(#) $Id: regress.cal,v 30.14 2017/05/19 16:09:14 chongo Exp $
+ * @(#) $Source: /usr/local/src/bin/calc-RHEL7/cal/RCS/regress.cal,v $
  *
  * Under source code control:	1990/02/15 01:50:36
  * File existed as early as:	before 1990
@@ -403,14 +403,14 @@ define test_config()
 					'512: config("trace") == 0');
 	vrfy(config("maxprint") == 16,
 					'513: config("maxprint") == 16');
-	vrfy(config("mul2") == 1780,
-					'514: config("mul2") == 1780');
-	vrfy(config("sq2") == 3388,
-					'515: config("sq2") == 3388');
-	vrfy(config("pow2") == 176,
-					'516: config("pow2") == 176');
-	vrfy(config("redc2") == 220,
-					'517: config("redc2") == 220');
+	vrfy(config("mul2") == 28,
+					'514: config("mul2") == 28');
+	vrfy(config("sq2") == 28,
+					'515: config("sq2") == 28');
+	vrfy(config("pow2") == 20,
+					'516: config("pow2") == 20');
+	vrfy(config("redc2") == 25,
+					'517: config("redc2") == 25');
 	vrfy(config("tilde"),
 					'518: config("tilde")');
 	vrfy(config("tab"),
@@ -768,6 +768,8 @@ print '016: parsed test_bignums()';
 
 /*
  * Test many of the built-in functions.
+ *
+ * See test_functionss() starting at test 9000 for more built-in function tests.
  */
 define test_functions()
 {
@@ -1397,7 +1399,8 @@ define test_functions()
 	vrfy(quomod(10,-3,a,b,12) == 1, '1193: vrfy(quomod(10,-3,a,b,12) == 1');
 	vrfy(a == -3,			'1194: a == -3');
 	vrfy(b == 1,			'1195: b == 1');
-	vrfy(quomod(-10,-3,a,b,13) == 1,'1196: vrfy(quomod(-10,-3,a,b,13) == 1');
+	vrfy(quomod(-10,-3,a,b,13) == 1,
+	    '1196: vrfy(quomod(-10,-3,a,b,13) == 1');
 	vrfy(a == 4,			'1197: a == 4');
 	vrfy(b == 2,			'1198: b == 2');
 	vrfy(quomod(10,3,a,b,14) == 1,	'1199: vrfy(quomod(10,3,a,b,14) == 1');
@@ -1444,7 +1447,12 @@ define test_functions()
 	vrfy(jacobi(-1,-1) == 0,	'1236: jacobi(-1,-1) == 0');
 	vrfy(jacobi(0,-1) == 0,		'1237: jacobi(0,-1) == 0');
 
-	print '1238: Ending test_functions';
+	/*
+	 * NOTE: Function tests are continued in test_functionss()
+	 *	 starting at test 9000.
+	 */
+
+	print '1293: Ending test_functions';
 }
 print '017: parsed test_functions()';
 
@@ -1461,14 +1469,14 @@ define _test_underscore()
 	local _a = 27;
 	local __a = 23209;
 
-	print "1290: Beginning _test_underscore";
+	print "1294: Beginning _test_underscore";
 
-	vrfy(_a == 27,			    '1291: _a == 27');
-	vrfy(_ == 49,			    '1292: _ == 49');
-	vrfy(__ == 63,			    '1293: __ == 63');
-	vrfy(__a == 23209,		    '1294: __a == 23209');
+	vrfy(_a == 27,			    '1295: _a == 27');
+	vrfy(_ == 49,			    '1296: _ == 49');
+	vrfy(__ == 63,			    '1297: __ == 63');
+	vrfy(__a == 23209,		    '1298: __a == 23209');
 
-	print "1295: Ending _test_underscore";
+	print "1299: Ending _test_underscore";
 }
 print '020: parsed _test_underscore';
 
@@ -1509,8 +1517,10 @@ define test_assoc()
 	vrfy(isnull(search(a,16)),	'1312: isnull(search(a,16))');
 	a["curds","whey"] = "spider";
 	print				'1313: a["curds","whey"] = "spider"';
-	vrfy(a["curds","whey"] == "spider", '1314: a["curds","whey"] == "spider"');
-	vrfy(a[[rsearch(a,"spider")]] == "spider", '1315: a[[rsearch(a,"spider")]] == "spider"');
+	vrfy(a["curds","whey"] == "spider",
+		'1314: a["curds","whey"] == "spider"');
+	vrfy(a[[rsearch(a,"spider")]] == "spider",
+		'1315: a[[rsearch(a,"spider")]] == "spider"');
 	b = a;
 	print				'1316: b = a';
 	vrfy(b[17] == 19,		'1317: b[17] == 19');
@@ -4892,7 +4902,8 @@ define test_newsyn()
 		vrfy(s5500 == 55,		  '5510: s5500 == 45');
 		vrfy(i == 11,		  '5511: i == 11');
 	}
-	print "5512: { local i; for (s5500 = 0, i = 0; i < 10; i++) s5500 += i; ... }";
+	print "5512: { local i; for (s5500 = 0, i = 0; i < 10; i++) ":
+	      "s5500 += i; ... }";
 	vrfy(s5500 == 55,			  '5513: s5500 == 55');
 	vrfy(i == 11,			  '5514: i == 11');
 
@@ -5027,7 +5038,7 @@ vrfy(i == 9,		'151: i == 9');
 
 
 /*
- * test_commaeq - test changes to , and =
+ * test_commaeq - test changes to = and ,
  */
 obj xx5600 {} xx5600;
 print '152: obj xx5600 {} xx5600';
@@ -6506,7 +6517,8 @@ define test_blk()
 	/* A second named block */
 
 	B1 = blk("+++6700", 15, 10) = {1,2,3,4,5};
-	print				  '6746: B1 = blk("+++6700", 15 , 10) = {1,2,3,4,5};';
+	print
+	    '6746: B1 = blk("+++6700", 15 , 10) = {1,2,3,4,5};';
 	vrfy(size(B1) == 15,		  '6747: size(B1) == 15');
 	vrfy(sizeof(B1) == 20,		  '6748: sizeof(B1) == 20');
 	vrfy(test(B1) == 1,		  '6749: test(B1) == 1');
@@ -6871,7 +6883,8 @@ define test_sha1()
 
 
 	z = sha1(list(1,2,3), "curds and whey", 2^21701-1, pi(1e-100));
-	print '7210: z = sha1(list(1,2,3), "curds and whey", 2^21701-1, pi(1e-100));';
+	print '7210: z = sha1(list(1,2,3), "curds and whey",',
+	      '2^21701-1, pi(1e-100));';
 	vrfy(sha1(z) == 0x158cc87deeb9dd478ca14e3ab359205b0fb15b83,
 		'7211: sha1(z) == 0x158cc87deeb9dd478ca14e3ab359205b0fb15b83');
 
@@ -7914,6 +7927,137 @@ return test_exponentiation();
 /* 88xx: test exponentiation */
 
 
+/*
+ * calc resource functions by Christoph Zurnieden
+ */
+print;
+print '8900: Starting test of calc resource functions by Christoph Zurnieden';
+print '8901: read -once "test8900"';
+read -once "test8900";
+print '8902: about to run test8900(1,,8903)';
+testnum = test8900(1,,8903);
+print '8999: ecnt = 203;'
+ecnt = 203;
+/* 89xx: test calc resource functions by Christoph Zurnieden */
+
+
+/*
+ * Test more of the built-in functions.
+ *
+ * See test_functions() (test 700 - 1238) for other built-in function tests.
+ */
+define test_functions2()
+{
+	print '9001: Beginning test_functions2';
+
+	/* ctype function tests */
+	vrfy(isalnum("A") == 1,		'9002: isalnum("A") == 1');
+	vrfy(isalnum("a") == 1,		'9003: isalnum("a") == 1');
+	vrfy(isalnum("2") == 1,		'9004: isalnum("2") == 1');
+	vrfy(isalnum("\t") == 0,	'9005: isalnum("\\t") == 0');
+
+	vrfy(isalpha("A") == 1,		'9006: isalpha("A") == 1');
+	vrfy(isalpha("a") == 1,		'9007: isalpha("a") == 1');
+	vrfy(isalpha("2") == 0,		'9008: isalpha("2") == 0');
+	vrfy(isalpha("\t") == 0,	'9009: isalpha("\\t") == 0');
+
+	vrfy(iscntrl("A") == 0,		'9010: iscntrl("A") == 0');
+	vrfy(iscntrl("a") == 0,		'9011: iscntrl("a") == 0');
+	vrfy(iscntrl("2") == 0,		'9012: iscntrl("2") == 0');
+	vrfy(iscntrl("\t") == 1,	'9013: iscntrl("\\t") == 1');
+
+	vrfy(isdigit("A") == 0,		'9014: isdigit("A") == 0');
+	vrfy(isdigit("a") == 0,		'9015: isdigit("a") == 0');
+	vrfy(isdigit("2") == 1,		'9016: isdigit("2") == 1');
+	vrfy(isdigit("\t") == 0,	'9017: isdigit("\\t") == 0');
+
+	vrfy(isgraph("A") == 1,		'9018: isgraph("A") == 1');
+	vrfy(isgraph("a") == 1,		'9019: isgraph("a") == 1');
+	vrfy(isgraph("2") == 1,		'9020: isgraph("2") == 1');
+	vrfy(isgraph("\t") == 0,	'9021: isgraph("\\t") == 0');
+
+	vrfy(islower("A") == 0,		'9022: islower("A") == 0');
+	vrfy(islower("a") == 1,		'9023: islower("a") == 1');
+	vrfy(islower("1") == 0,		'9024: islower("1") == 0');
+
+	vrfy(isprint("A") == 1,		'9025: isprint("A") == 1');
+	vrfy(isprint("a") == 1,		'9026: isprint("a") == 1');
+	vrfy(isprint(" ") == 1,		'9027: isprint(" ") == 1');
+	vrfy(isprint("\t") == 0,	'9028: isprint("\\t") == 0');
+
+	vrfy(ispunct("A") == 0,		'9029: ispunct("A") == 0');
+	vrfy(ispunct("a") == 0,		'9030: ispunct("a") == 0');
+	vrfy(ispunct(" ") == 0,		'9031: ispunct(" ") == 0');
+	vrfy(ispunct("?") == 1,		'9032: ispunct("?") == 1');
+
+	vrfy(isspace("A") == 0,		'9033: isspace("A") == 0');
+	vrfy(isspace("Krik") == 0,	'9034: isspace("Krik") == 0');
+	vrfy(isspace(" ") == 1,		'9035: isspace(" ") == 1');
+	vrfy(isspace("?") == 0,		'9036: isspace("?") == 0');
+
+	vrfy(isupper("A") == 1,		'9037: isupper("A") == 1');
+	vrfy(isupper("a") == 0,		'9038: isupper("a") == 0');
+	vrfy(isupper("1") == 0,		'9039: isupper("1") == 0');
+
+	vrfy(isxdigit("A") == 1,	'9040: isxdigit("A") == 1');
+	vrfy(isxdigit("f") == 1,	'9041: isxdigit("f") == 1');
+	vrfy(isxdigit("2") == 1,	'9042: isxdigit("2") == 1');
+	vrfy(isxdigit("x") == 0,	'9043: isxdigit("x") == 0');
+
+	vrfy(strcasecmp("ab", "aBc") == -1,
+			     '9044: strcasecmp("ab", "aBc") == -1');
+	vrfy(strcasecmp("abc", "aBb") == 1,
+			     '9045: strcasecmp("abc", "aBb") == 1');
+	vrfy(strcasecmp("abc", "abc") == 0,
+			     '9046: strcasecmp("abc", "abc") == 0');
+	vrfy(strcasecmp("abc", "aBc") == 0,
+			     '9047: strcasecmp("abc", "aBc") == 0');
+	vrfy(strcasecmp("abc", "aBd") == -1,
+			     '9048: strcasecmp("abc", "aBd") == -1');
+	vrfy(strcasecmp("abc\0", "aBc") == 1,
+			     '9049: strcasecmp("a8c\\0", "aBc") == 1');
+	vrfy(strcasecmp("a\0b", "A\0c") == -1,
+			     '9050: strcasecmp("a\\0b", "A\\0c") == -1');
+
+	vrfy(strncasecmp("abc", "xyz", 0) == 0,
+			     '9051: strncasecmp("abc", "xyz", 0) == 0');
+	vrfy(strncasecmp("abc", "xyz", 1) == -1,
+			     '9052: strncasecmp("abc", "xyz", 1) == -1');
+	vrfy(strncasecmp("abc", "", 1) == 1,
+			     '9053: strncasecmp("abc", "", 1) == 1');
+	vrfy(strncasecmp("a", "b", 2) == -1,
+			     '9054: strncasecmp("a", "b", 2) == -1');
+	vrfy(strncasecmp("ab", "Ac", 2) == -1,
+			     '9055: strncasecmp("ab", "Ac", 2) == -1');
+	vrfy(strncasecmp("\0ac", "\0b", 2) == -1,
+			     '9056: strncasecmp("\\0ac", "\\0b", 2) == -1');
+	vrfy(strncasecmp("ab", "aBc", 2) == 0,
+			     '9057: strncasecmp("ab", "aBc", 2) == 0');
+	vrfy(strncasecmp("abc", "abd", 2) == 0,
+			     '9058: strncasecmp("abc", "abd", 2) == 0');
+
+	local s1 = " gnu lesser general public license";
+	print '9059: local s1 = " gnu lesser general public license";';
+
+	vrfy(strcmp(strtolower(" GNU Lesser General Public License"), s1) == 0,
+	    '9060: strcmp(strtolower(" GNU Lesser General Public License"),' +
+	    ' s1) == 0');
+
+	local s2 = " GNU LESSER GENERAL PUBLIC LICENSE";
+	print '9061: local s2 = " GNU LESSER GENERAL PUBLIC LICENSE";';
+
+	vrfy(strcmp(strtoupper(" GNU Lesser General Public License"), s2) == 0,
+	    '9062: strcmp(strtoupper(" GNU Lesser General Public License"),' +
+	    ' s2) == 0');
+
+	print '9063: Ending test_functions2';
+}
+print;
+print '9000: parsed test_functions2()';
+print;
+return test_functions2();
+
+
 /*
  * read various calc resource files
  *
@@ -7924,7 +8068,6 @@ return test_exponentiation();
  *	hello.cal	- designed to go into an infinite loop
  *	lucal.cal	- already read by this file
  *	lucas_chk.cal	- already read by this file
- *	lucas_tbl.cal	- duplicatres code already read by another file
  *	regress.cal	- this file
  *	surd.cal	- already read by this file
  *	test9999.cal	- files of this form are already read by this file
@@ -7997,8 +8140,6 @@ read -once intfile;
 print	'9830: read -once intfile';
 read -once lucas;
 print	'9831: read -once lucas';
-read -once lucas_tbl;
-print	'9832: read -once lucas_tbl';
 read -once natnumset;
 print	'9833: read -once natnumset';
 read -once repeat;

cal/smallfactors.cal

@@ -0,0 +1,71 @@
+/*
+ * smallfactors - find the factors of a number < 2^32
+ *
+ * Copyright (C) 2013 Christoph Zurnieden
+ *
+ * Calc is open software; you can redistribute it and/or modify it under
+ * the terms of the version 2.1 of the GNU Lesser General Public License
+ * as published by the Free Software Foundation.
+ *
+ * Calc is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+ * or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU Lesser General
+ * Public License for more details.
+ *
+ * A copy of version 2.1 of the GNU Lesser General Public License is
+ * distributed with calc under the filename COPYING-LGPL.  You should have
+ * received a copy with calc; if not, write to Free Software Foundation, Inc.
+ * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
+ */
+
+
+static resource_debug_level;
+resource_debug_level = config("resource_debug", 0);
+
+
+define smallfactors(x0)
+{
+    local d q x flist tuple w;
+
+    if (x >= (2 ^ 32) - 1)
+	return newerror("smallfactors: number must be < 2^32 -1");
+
+    tuple = mat[2];
+    flist = list();
+    x = x0;
+    d = 2;
+    q = 0;
+    tuple[0] = d;
+    if (x < 2)
+	return 0;
+    do {
+	q = x // d;
+	    while (x == (q * d)) {
+	    tuple[0] = d;
+	    tuple[1]++;
+	    x = floor(q);
+	    q = x // d;
+	}
+	d = nextprime(d);
+	if (tuple[1] > 0)
+	    append(flist, tuple);
+	tuple = mat[2];
+    } while (d <= x);
+    return flist;
+}
+
+define printsmallfactors(flist)
+{
+    local k;
+    for (k = 0; k < size(flist); k++) {
+	print flist[k][0]:"^":flist[k][1];
+    }
+}
+
+
+config("resource_debug", resource_debug_level),;
+if (config("resource_debug") & 3) {
+    print "smallfactors(x0)";
+    print "printsmallfactors(flist)";
+
+}

cal/solve.cal

@@ -17,8 +17,8 @@
  * received a copy with calc; if not, write to Free Software Foundation, Inc.
  * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
  *
- * @(#) $Revision: 30.2 $
- * @(#) $Id: solve.cal,v 30.2 2008/05/10 13:30:00 chongo Exp $
+ * @(#) $Revision: 30.3 $
+ * @(#) $Id: solve.cal,v 30.3 2013/08/11 08:41:38 chongo Exp $
  * @(#) $Source: /usr/local/src/bin/calc/cal/RCS/solve.cal,v $
  *
  * Under source code control:	1990/02/15 01:50:37
@@ -52,7 +52,8 @@ define solve(low, high, epsilon)
 	if (sgn(flow) == sgn(fhigh))
 		quit "Non-opposite signs";
 	while (1) {
-		mid = bround(high - fhigh * (high - low) / (fhigh - flow), places);
+		mid = bround(high - fhigh * (high - low) / (fhigh - flow),
+			     places);
 		if ((mid == low) || (mid == high))
 			places++;
 		fmid = f(mid);

cal/statistics.cal

@@ -0,0 +1,502 @@
+/*
+ * statistics - Some assorted statistics functions.
+ *
+ * Copyright (C) 2013 Christoph Zurnieden
+ *
+ * Calc is open software; you can redistribute it and/or modify it under
+ * the terms of the version 2.1 of the GNU Lesser General Public License
+ * as published by the Free Software Foundation.
+ *
+ * Calc is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+ * or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU Lesser General
+ * Public License for more details.
+ *
+ * A copy of version 2.1 of the GNU Lesser General Public License is
+ * distributed with calc under the filename COPYING-LGPL.  You should have
+ * received a copy with calc; if not, write to Free Software Foundation, Inc.
+ * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
+ *
+ * @(#) $Revision: 30.4 $
+ * @(#) $Id: statistics.cal,v 30.4 2013/08/18 20:01:53 chongo Exp $
+ * @(#) $Source: /usr/local/src/bin/calc/cal/RCS/statistics.cal,v $
+ *
+ * Under source code control:	2013/08/11 01:31:28
+ * File existed as early as:	2013
+ */
+
+
+static resource_debug_level;
+resource_debug_level = config("resource_debug", 0);
+
+
+/*
+ *  get dependencies
+ */
+read -once factorial2 brentsolve
+
+
+/*******************************************************************************
+ *
+ *
+ *                Continuous distributions
+ *
+ *
+ ******************************************************************************/
+
+/* regularized incomplete gamma function like in Octave, hence the name */
+define gammaincoctave(z,a){
+  local tmp;
+  tmp = gamma(z);
+  return (tmp-gammainc(a,z))/tmp;
+}
+
+/* Inverse incomplete beta function. Old and slow. */
+static __CZ__invbeta_a;
+static __CZ__invbeta_b;
+static __CZ__invbeta_x;
+define __CZ__invbeta(x){
+   return __CZ__invbeta_x-__CZ__ibetaas63(x,__CZ__invbeta_a,__CZ__invbeta_b);
+}
+
+define invbetainc_slow(x,a,b){
+  local flag ret eps;
+  /* place checks and balances here */
+  eps = epsilon();
+  if(.5 < x){
+    __CZ__invbeta_x = 1 - x;
+    __CZ__invbeta_a = b;
+    __CZ__invbeta_b = a;
+    flag = 1;
+  }
+  else{
+    __CZ__invbeta_x = x;
+    __CZ__invbeta_a = a;
+    __CZ__invbeta_b = b;
+    flag = 0;
+  }
+
+  ret =  brentsolve2(0,1,1);
+
+  if(flag == 1)
+    ret = 1-ret;
+  epsilon(eps);
+  return ret;
+}
+
+/* Inverse incomplete beta function. Still old but not as slow as the function
+   above. */
+/*
+  Purpose:
+
+    invbetainc computes inverse of the incomplete Beta function.
+
+  Licensing:
+
+    This code is distributed under the GNU LGPL license.
+
+  Modified:
+
+    10 August 2013
+
+  Author:
+
+    Original FORTRAN77 version by GW Cran, KJ Martin, GE Thomas.
+    C version by John Burkardt.
+    Calc version by Christoph Zurnieden
+
+  Reference:
+
+    GW Cran, KJ Martin, GE Thomas,
+    Remark AS R19 and Algorithm AS 109:
+    A Remark on Algorithms AS 63: The Incomplete Beta Integral
+    and AS 64: Inverse of the Incomplete Beta Integeral,
+    Applied Statistics,
+    Volume 26, Number 1, 1977, pages 111-114.
+
+  Parameters:
+
+    Input,  P, Q, the parameters of the incomplete
+    Beta function.
+
+    Input,  BETA, the logarithm of the value of
+    the complete Beta function.
+
+    Input,  ALPHA, the value of the incomplete Beta
+    function.  0 <= ALPHA <= 1.
+
+    Output,  the argument of the incomplete
+    Beta function which produces the value ALPHA.
+
+  Local Parameters:
+
+    Local, SAE, the most negative decimal exponent
+    which does not cause an underflow.
+*/
+define invbetainc(x,a,b){
+  return __CZ__invbetainc(a,b,lnbeta(a,b),x);
+}
+
+define __CZ__invbetainc(p,q,beta,alpha){
+  local a acu adj fpu g h iex indx pp prev qq r s sae sq t tx value;
+  local w xin y yprev places eps;
+
+  /* Dirty trick, don't try at home */
+  eps= epsilon(epsilon()^2);
+  sae = -((log(1/epsilon())/log(2))//2);
+  fpu = 10.0^sae;
+
+  places = highbit(1 + int(1/epsilon())) + 1;
+  value = alpha;
+  if( p <= 0.0 ){
+    epsilon(eps);
+    return newerror("invbeta: argument p <= 0");
+  }
+  if( q <= 0.0 ){
+    epsilon(eps);
+    return newerror("invbeta: argument q <= 0");
+  }
+
+  if( alpha < 0.0 || 1.0 < alpha ){
+    epsilon(eps);
+    return newerror("invbeta: argument alpha out of domain");
+  }
+  if( alpha == 0.0 ){
+    epsilon(eps);
+    return 0;
+  }
+  if( alpha == 1.0 ){
+    epsilon(eps);
+    return 1;
+  }
+  if ( 0.5 < alpha ){
+    a = 1.0 - alpha;
+    pp = q;
+    qq = p;
+    indx = 1;
+  }
+  else{
+    a = alpha;
+    pp = p;
+    qq = q;
+    indx = 0;
+  }
+  r = sqrt ( - ln ( a * a ) );
+
+  y = r-(2.30753+0.27061*r)/(1.0+(0.99229+0.04481*r)*r);
+
+  if ( 1.0 < pp && 1.0 < qq ){
+    r = ( y * y - 3.0 ) / 6.0;
+    s = 1.0 / ( pp + pp - 1.0 );
+    t = 1.0 / ( qq + qq - 1.0 );
+    h = 2.0 / ( s + t );
+    w = y*sqrt(h+r)/h-(t-s)*(r+5.0/6.0-2.0/(3.0*h));
+    value = pp / ( pp + qq * exp ( w + w ) );
+  }
+  else{
+    r = qq + qq;
+    t = 1.0 / ( 9.0 * qq );
+    t = r *  ( 1.0 - t + y * sqrt ( t )^ 3 );
+
+    if ( t <= 0.0 ){
+      value = 1.0 - exp ( ( ln ( ( 1.0 - a ) * qq ) + beta ) / qq );
+    }
+    else{
+      t = ( 4.0 * pp + r - 2.0 ) / t;
+
+      if ( t <= 1.0 ) {
+        value = exp ( ( ln ( a * pp ) + beta ) / pp );
+      }
+      else{
+        value = 1.0 - 2.0 / ( t + 1.0 );
+      }
+    }
+  }
+  r = 1.0 - pp;
+  t = 1.0 - qq;
+  yprev = 0.0;
+  sq = 1.0;
+  prev = 1.0;
+
+  if ( value < 0.0001 )
+    value = 0.0001;
+
+  if ( 0.9999 < value )
+    value = 0.9999;
+
+  acu = 10^sae;
+
+  for ( ; ; ){
+    y = bround(__CZ__ibetaas63( value, pp, qq, beta),places);
+    xin = value;
+    y = bround(exp(ln(y-a)+(beta+r*ln(xin)+t*ln(1.0- xin ) )),places);
+
+    if ( y * yprev <= 0.0 ) {
+      prev = max ( sq, fpu );
+    }
+
+    g = 1.0;
+
+    for ( ; ; ){
+      for ( ; ; ){
+        adj = g * y;
+        sq = adj * adj;
+        if ( sq < prev ){
+          tx = value - adj;
+          if ( 0.0 <= tx && tx <= 1.0 )  break;
+        }
+        g = g / 3.0;
+      }
+      if ( prev <= acu ){
+        if ( indx )
+          value = 1.0 - value;
+        epsilon(eps);
+        return value;
+      }
+      if ( y * y <= acu ){
+        if ( indx )
+          value = 1.0 - value;
+        epsilon(eps);
+        return value;
+      }
+      if ( tx != 0.0 && tx != 1.0 )
+        break;
+      g = g / 3.0;
+    }
+    if ( tx == value )  break;
+    value = tx;
+    yprev = y;
+  }
+  if ( indx )
+    value = 1.0 - value;
+
+  epsilon(eps);
+  return value;
+}
+
+/*******************************************************************************
+ *
+ *
+ *                Beta distribution
+ *
+ *
+ ******************************************************************************/
+
+define betapdf(x,a,b){
+  if(x<0 || x>1) return newerror("betapdf: parameter x out of domain");
+  if(a<=0) return newerror("betapdf: parameter a out of domain");
+  if(b<=0) return newerror("betapdf: parameter b out of domain");
+
+  return 1/beta(a,b) *x^(a-1)*(1-x)^(b-1);
+}
+
+define betacdf(x,a,b){
+  if(x<0 || x>1) return newerror("betacdf: parameter x out of domain");
+  if(a<=0) return newerror("betacdf: parameter a out of domain");
+  if(b<=0) return newerror("betacdf: parameter b out of domain");
+
+  return betainc(x,a,b);
+}
+
+define betacdfinv(x,a,b){
+  return invbetainc(x,a,b);
+}
+
+define betamedian(a,b){
+  local t106 t104 t103 t105 approx ret;
+  if(a == b) return 1/2;
+  if(a == 1 && b > 0) return 1-(1/2)^(1/b);
+  if(a > 0 && b == 1) return   (1/2)^(1/a);
+  if(a == 3 && b == 2){
+    /* Yes, the author is not ashamed to ask Maxima for the exact solution
+       of a quartic equation. */
+    t103 = ( (2^(3/2))/27 +4/27  )^(1/3);
+    t104 = sqrt( ( 9*t103^2 + 4*t103 + 2  )/(t103)  )/3;
+    t105 = -t103-2/(9*t103) +8/9;
+    t106 = sqrt( (27*t104*t105+16)/(t104)  )/(2*3^(3/2));
+    return -t106+t104/2+1/3;
+  }
+  if(a == 2 && b == 3){
+    t103 = ( (2^(3/2))/27 +4/27  )^(1/3);
+    t104 = sqrt( ( 9*t103^2 + 4*t103 + 2  )/(t103)  )/3;
+    t105 = -t103-2/(9*t103) +8/9;
+    t106 = sqrt( (27*t104*t105+16)/(t104)  )/(2*3^(3/2));
+    return 1-(-t106+t104/2+1/3);
+  }
+  return invbetainc(1/2,a,b);
+}
+
+define betamode(a,b){
+  if(a + b == 2) return newerror("betamod: a + b = 2 = division by zero");
+  return (a-1)/(a+b-2);
+}
+
+define betavariance(a,b){
+  return (a*b)/( (a+b)^2*(a+b+1) );
+}
+
+define betalnvariance(a,b){
+  return polygamma(1,a)-polygamma(a+b);
+}
+
+define betaskewness(a,b){
+  return (2*(b-a)*sqrt(a+b+1))/( (a+b+1)*sqrt(a*b) );
+}
+
+define betakurtosis(a,b){
+  local num denom;
+
+  num = 6*( (a-b)^2*(a+b+1)-a*b*(a+b+2));
+  denom = a*b*(a+b+2)*(a+b+3);
+  return num/denom;
+}
+
+define betaentropy(a,b){
+  return lnbeta(a,b)-(a-1)*psi(a)-(b-1)*psi(b)+(a+b+1)*psi(a+b);
+
+}
+
+/*******************************************************************************
+ *
+ *
+ *                Normal (Gaussian) distribution
+ *
+ *
+ ******************************************************************************/
+
+
+define normalpdf(x,mu,sigma){
+  return 1/(sqrt(2*pi()*sigma^2))*exp( ( (x-mu)^2 )/( 2*sigma^2 ) );
+}
+
+define normalcdf(x,mu,sigma){
+  return 1/2*(1+erf( ( x-mu )/( sqrt(2*sigma^2) )  )  );
+}
+
+define probit(p){
+  if(p<0 || p > 1) return newerror("probit: p out of domain 0<=p<=1");
+  return sqrt(2)*ervinv(2*p-1);
+}
+
+define normalcdfinv(p,mu,sigma){
+  if(p<0 || p > 1) return newerror("normalcdfinv: p out of domain 0<=p<=1");
+  return mu+ sigma*probit(p);
+}
+
+define normalmean(mu,sigma){return mu;}
+
+define normalmedian(mu,sigma){return mu;}
+
+define normalmode(mu,sigma){return mu;}
+
+define normalvariance(mu,sigma){return sigma^2;}
+
+define normalskewness(mu,sigma){return 0;}
+
+define normalkurtosis(mu,sigma){return 0;}
+
+define normalentropy(mu,sigma){
+  return 1/3*ln( 2*pi()*exp(1)*sigma^2 );
+}
+
+/* moment generating f. */
+define normalmgf(mu,sigma,t){
+  return exp(mu*t+1/2*sigma^2*t^2);
+}
+
+/* characteristic f. */
+define normalcf(mu,sigma,t){
+  return exp(mu*t-1/2*sigma^2*t^2);
+}
+
+
+/*******************************************************************************
+ *
+ *
+ *                Chi-squared distribution
+ *
+ *
+ ******************************************************************************/
+
+define chisquaredpdf(x,k){
+  if(!isint(k) || k<0) return newerror("chisquaredpdf: k not in N");
+  if(im(x) || x<0) return newerror("chisquaredpdf: x not in +R");
+  /* The gamma function does not check for half integers, do it here? */
+  return 1/(2^(k/2)*gamma(k/2))*x^((k/2)-1)*exp(-x/2);
+}
+
+define chisquaredpcdf(x,k){
+  if(!isint(k) || k<0) return newerror("chisquaredcdf: k not in N");
+  if(im(x) || x<0) return newerror("chisquaredcdf: x not in +R");
+
+  return 1/(gamma(k/2))*gammainc(k/2,x/2);
+}
+
+define chisquaredmean(x,k){return k;}
+
+define chisquaredmedian(x,k){
+  /* TODO: implement a FAST inverse incomplete gamma-{q,p} function */
+  return k*(1-2/(9*k))^3;
+}
+
+define chisquaredmode(x,k){return max(k-2,0);}
+define chisquaredvariance(x,k){return 2*k;}
+define chisquaredskewness(x,k){return sqrt(8/k);}
+define chisquaredkurtosis(x,k){return 12/k;}
+define chisquaredentropy(x,k){
+  return k/2+ln(2*gamma(k/2)) + (1-k/2)*psi(k/2);
+}
+
+define chisquaredmfg(k,t){
+  if(t>=1/2)return newerror("chisquaredmfg: t >= 1/2");
+  return (1-2*t)^(k/2);
+}
+
+define chisquaredcf(k,t){
+  return (1-2*1i*t)^(k/2);
+}
+
+
+/*
+ * restore internal function from resource debugging
+ */
+config("resource_debug", resource_debug_level),;
+if (config("resource_debug") & 3) {
+    print "gammaincoctave(z,a)";
+    print "invbetainc(x,a,b)";
+    print "betapdf(x,a,b)";
+    print "betacdf(x,a,b)";
+    print "betacdfinv(x,a,b)";
+    print "betamedian(a,b)";
+    print "betamode(a,b)";
+    print "betavariance(a,b)";
+    print "betalnvariance(a,b)";
+    print "betaskewness(a,b)";
+    print "betakurtosis(a,b)";
+    print "betaentropy(a,b)";
+    print "normalpdf(x,mu,sigma)";
+    print "normalcdf(x,mu,sigma)";
+    print "probit(p)";
+    print "normalcdfinv(p,mu,sigma)";
+    print "normalmean(mu,sigma)";
+    print "normalmedian(mu,sigma)";
+    print "normalmode(mu,sigma)";
+    print "normalvariance(mu,sigma)";
+    print "normalskewness(mu,sigma)";
+    print "normalkurtosis(mu,sigma)";
+    print "normalentropy(mu,sigma)";
+    print "normalmgf(mu,sigma,t)";
+    print "normalcf(mu,sigma,t)";
+    print "chisquaredpdf(x,k)";
+    print "chisquaredpcdf(x,k)";
+    print "chisquaredmean(x,k)";
+    print "chisquaredmedian(x,k)";
+    print "chisquaredmode(x,k)";
+    print "chisquaredvariance(x,k)";
+    print "chisquaredskewness(x,k)";
+    print "chisquaredkurtosis(x,k)";
+    print "chisquaredentropy(x,k)";
+    print "chisquaredmfg(k,t)";
+    print "chisquaredcf(k,t)";
+}
+

cal/strings.cal

@@ -0,0 +1,41 @@
+/*
+ * strings - implementation of some of the macros in ctype.h
+ *
+ * Copyright (C) 2013 Christoph Zurnieden
+ *
+ * Calc is open software; you can redistribute it and/or modify it under
+ * the terms of the version 2.1 of the GNU Lesser General Public License
+ * as published by the Free Software Foundation.
+ *
+ * Calc is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+ * or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU Lesser General
+ * Public License for more details.
+ *
+ * A copy of version 2.1 of the GNU Lesser General Public License is
+ * distributed with calc under the filename COPYING-LGPL.  You should have
+ * received a copy with calc; if not, write to Free Software Foundation, Inc.
+ * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
+ */
+
+
+static resource_debug_level;
+resource_debug_level = config("resource_debug", 0);
+
+define isascii(c){
+  c = ord(c);
+  return (c >= 0 && c< 128);
+}
+
+define isblank(c){
+  c = ord(c);
+  return ( c == 32 || c == 9 );
+}
+
+
+config("resource_debug", resource_debug_level),;
+if (config("resource_debug") & 3) {
+    print "isascii(c)";
+    print "isblank(c)";
+}
+

cal/test2600.cal

@@ -19,8 +19,8 @@
  * received a copy with calc; if not, write to Free Software Foundation, Inc.
  * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
  *
- * @(#) $Revision: 30.2 $
- * @(#) $Id: test2600.cal,v 30.2 2007/07/11 22:57:23 chongo Exp $
+ * @(#) $Revision: 30.3 $
+ * @(#) $Id: test2600.cal,v 30.3 2013/08/11 08:41:38 chongo Exp $
  * @(#) $Source: /usr/local/src/bin/calc/cal/RCS/test2600.cal,v $
  *
  * Under source code control:	1995/10/13 00:13:14
@@ -91,7 +91,8 @@ define testismult(str, n, verbose)
 		if (!ismult(c,a)) {
 			m++;
 			if (verbose > 1) {
-				printf("*** Failure with:\na = %d\nb = %d\n", a,b);
+				printf("*** Failure with:\na = %d\nb = %d\n",
+				       a,b);
 			}
 		}
 	}
@@ -133,7 +134,8 @@ define testsqrt(str, n, eps, verbose)
 		if (abs(c) > 1) {
 			m++;
 			if (verbose > 1) {
-				printf("*** Failure with:\na = %d\neps = %d\n", a,eps);
+				printf("*** Failure with:\na = %d\neps = %d\n",
+				       a,eps);
 			}
 		}
 	}
@@ -178,7 +180,8 @@ define testexp(str, n, eps, verbose)
 		if (abs(c) > 0.02) {
 			m++;
 			if (verbose > 1) {
-				printf("*** Failure with:\na = %d\neps = %d\n", a,eps);
+				printf("*** Failure with:\na = %d\neps = %d\n",
+				       a,eps);
 			}
 		}
 	}
@@ -235,7 +238,8 @@ define testln(str, n, eps, verbose)
 		if (abs(c) > 0.5) {
 			m++;
 			if (verbose > 1) {
-				printf("*** Failure with:\na = %d\neps = %d\n", a,eps);
+				printf("*** Failure with:\na = %d\neps = %d\n",
+				       a,eps);
 			}
 		}
 	}

cal/test2700.cal

@@ -19,8 +19,8 @@
  * received a copy with calc; if not, write to Free Software Foundation, Inc.
  * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
  *
- * @(#) $Revision: 30.1 $
- * @(#) $Id: test2700.cal,v 30.1 2007/03/16 11:09:54 chongo Exp $
+ * @(#) $Revision: 30.2 $
+ * @(#) $Id: test2700.cal,v 30.2 2013/08/11 08:41:38 chongo Exp $
  * @(#) $Source: /usr/local/src/bin/calc/cal/RCS/test2700.cal,v $
  *
  * Under source code control:	1995/11/01 22:52:25
@@ -127,7 +127,8 @@ define testcsqrt(str, n, verbose)
 			if (p) {
 			if (verbose > 0)
 				printf(
-				 "*** Type %d failure for x = %r, y = %r, z = %d\n",
+				 "*** Type %d failure for x = %r, "
+				 "y = %r, z = %d\n",
 				    p, x, y, z);
 				m++;
 			}

cal/test4000.cal

@@ -19,8 +19,8 @@
  * received a copy with calc; if not, write to Free Software Foundation, Inc.
  * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
  *
- * @(#) $Revision: 30.1 $
- * @(#) $Id: test4000.cal,v 30.1 2007/03/16 11:09:54 chongo Exp $
+ * @(#) $Revision: 30.2 $
+ * @(#) $Id: test4000.cal,v 30.2 2013/08/11 08:41:38 chongo Exp $
  * @(#) $Source: /usr/local/src/bin/calc/cal/RCS/test4000.cal,v $
  *
  * Under source code control:	1996/03/13 02:38:45
@@ -199,7 +199,8 @@ define ctimes(str, N, n, count, skip, verbose)
 		p = ptest(A[i], count, skip);
 		if (p) {
 			if (verbose > 0) {
-				printf("*** Error, what should be rare has occurred for x = %d \n", A[i]);
+				printf("*** Error, what should be rare "
+				       "has occurred for x = %d \n", A[i]);
 				m++;
 			}
 		}
@@ -306,7 +307,8 @@ define ntimes(str, N, n, count, skip, residue, modulus, verbose)
 	}
 	tprev = round(usertime() - t, 4);
 	if (verbose > 0) {
-		printf("%d evaluations, nextcand: %d, prevcand: %d\n", n, tnext, tprev);
+		printf("%d evaluations, nextcand: %d, "
+		       "prevcand: %d\n", n, tnext, tprev);
 	}
 }
 

cal/test8500.cal

@@ -19,8 +19,8 @@
  * received a copy with calc; if not, write to Free Software Foundation, Inc.
  * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
  *
- * @(#) $Revision: 30.1 $
- * @(#) $Id: test8500.cal,v 30.1 2007/03/16 11:09:54 chongo Exp $
+ * @(#) $Revision: 30.2 $
+ * @(#) $Id: test8500.cal,v 30.2 2013/08/11 08:41:38 chongo Exp $
  * @(#) $Source: /usr/local/src/bin/calc/cal/RCS/test8500.cal,v $
  *
  * Under source code control:	1999/11/12 20:59:59
@@ -134,8 +134,8 @@ define onetest_8500(a,b,rnd) {
  * The rounding parameter is randomly chosen.
  *
  * After a run of divmod_8500 the a, b, rnd values which gave failure are
- * stored in the list L_8500.  L_8500[0], L_8500[1], L_8500[2] are a, b, rnd for the first
- * test, etc.
+ * stored in the list L_8500.  L_8500[0], L_8500[1], L_8500[2] are a, b,
+ * rnd for the first* test, etc.
  */
 define divmod_8500(N = 10, M1 = 2^128, M2 = 2^64, testnum = 0)
 {

cal/toomcook.cal

@@ -0,0 +1,362 @@
+/*
+ * toomcook - implementation of Toom-Cook(3,4) multiplication algorithm
+ *
+ * Copyright (C) 2013 Christoph Zurnieden
+ *
+ * Calc is open software; you can redistribute it and/or modify it under
+ * the terms of the version 2.1 of the GNU Lesser General Public License
+ * as published by the Free Software Foundation.
+ *
+ * Calc is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+ * or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU Lesser General
+ * Public License for more details.
+ *
+ * A copy of version 2.1 of the GNU Lesser General Public License is
+ * distributed with calc under the filename COPYING-LGPL.  You should have
+ * received a copy with calc; if not, write to Free Software Foundation, Inc.
+ * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
+ *
+ * @(#) $Revision: 30.4 $
+ * @(#) $Id: toomcook.cal,v 30.4 2013/08/18 20:01:53 chongo Exp $
+ * @(#) $Source: /usr/local/src/bin/calc/cal/RCS/toomcook.cal,v $
+ *
+ * Under source code control:	2013/08/11 01:31:28
+ * File existed as early as:	2013
+ */
+
+/*
+ * hide internal function from resource debugging
+ */
+static resource_debug_level;
+resource_debug_level = config("resource_debug", 0);
+
+
+/* */
+define toomcook3(a,b){
+  local alen blen  a0 a1 a2  b0 b1 b2 m ret sign mask;
+  local S0 S1 S2 S3 S4 T1 T2;
+
+  if(!isint(a) || !isint(b))
+      return newerror("toomcook3(a,b): a and/or b is not an integer");
+
+  alen = digits(a,2);
+  blen = digits(b,2);
+
+  sign = sgn(a) * sgn(b);
+  /* sgn(x) returns 0 if x = 0 */
+  if(sign == 0) return 0;
+
+  m = min(alen,blen)//3;
+  mask = ~-(1<<m);
+
+  /*
+    Cut-off at about 4,000 dec. digits
+    TODO: check
+  */
+  if(isdefined("test8900")){
+    if(m < 20) return a*b;
+  }
+  else{
+    if(m < 4096 ) return a*b;
+  }
+  a = abs(a);
+  b = abs(b);
+
+  a0 = a & mask;
+  a1 = (a>>m) & mask;
+  a2 = (a>>(2*m));
+
+  b0 =  b         & mask;
+  b1 = (b>>m)     & mask;
+  b2 = (b>>(2*m));
+
+  /*
+    Zimmermann
+  */
+
+  S0 = toomcook3(a0 , b0);
+  S1 = toomcook3((a2+a1+a0) , (b2+b1+b0));
+  S2 = toomcook3(((a2<<2)+(a1<<1)+a0) , ((b2<<2)+(b1<<1)+b0));
+  S3 = toomcook3((a2-a1+a0) , (b2-b1+b0));
+  S4 = toomcook3(a2,b2);
+  T1 = (S3<<1) + S2;
+  T1 /= 3;
+  T1 += S0;
+  T1 >>= 1;
+  T1 -= S4<<1;
+  T2 = (S1 + S3)>>1;
+  S1 -= T1;
+  S2 = T2 - S0 - S4;
+  S3 = T1 - T2;
+
+  ret =  (S4<<(4*m)) + (S3<<(3*m)) + (S2<<(2*m)) + (S1<<(1*m)) + S0;
+
+
+  ret = sign *ret;
+
+  return ret;
+}
+
+define toomcook3square(a){
+  local alen  a0 a1 a2  m tmp tmp2 ret sign S0 S1 S2 S3 S4 T1 mask;
+
+  if(!isint(a))return newerror("toomcook3square(a): a is not integer");
+
+  alen = digits(a,2);
+
+  sign = sgn(a) * sgn(a);
+  if(sign == 0) return 0;
+
+  m = alen//3;
+  mask = ~-(1<<m);
+  /*
+    Cut-off at about 5,000 dec. digits
+    TODO: check
+  */
+
+  if(isdefined("test8900")){
+    if(m < 20) return a^2;
+  }
+  else{
+    if(m < 5000 ) return a^2;
+  }
+
+  a = abs(a);
+
+  a0 = a & mask;
+  a1 = (a>>m) & mask;
+  a2 = (a>>(2*m));
+
+  /*
+   Bodrato/Zanoni
+   */
+  S0 = toomcook3square(a0);
+  S1 = toomcook3square(a2+a1+a0);
+  S2 = toomcook3square(a2-a1+a0);
+  S3 = toomcook3(a1<<1,a2);
+  S4 = toomcook3square(a2);
+
+  T1 = (S1 + S2)>>1;
+  S1 = S1 - T1 - S3;
+  S2 = T1 - S4 -S0;
+
+
+  S1 = S1<<(1*m);
+  S2 = S2<<(2*m);
+  S3 = S3<<(3*m);
+  S4 = S4<<(4*m);
+
+  ret = S0 + S1 + S2 + S3 + S4;
+  ret = sign *ret;
+
+  return ret;
+}
+
+define toomcook4(a,b)
+{
+
+  local  a0 a1 a2 a3 b0 b1 b2 b3 b4 ret tmp tmp2 tmp3 sign;
+  local  m alen blen mask;
+  local w1, w2, w3, w4, w5, w6, w7;
+
+  if(!isint(a) || !isint(b))
+    return newerror("toomcook4(a,b): a and/or b is not integer");
+
+  alen = digits(a,2);
+  blen = digits(b,2);
+
+  sign = sgn(a) * sgn(b);
+
+  if(sign == 0) return 0;
+
+  m = min(alen//4,blen//4);
+  mask = ~-(1<<m);
+
+  if(isdefined("test8900")){
+    if(m < 100) return toomcook3(a,b);
+  }
+  else{
+    if(m < 256*3072) return toomcook3(a,b);
+  }
+
+  a = abs(a);
+  b = abs(b);
+
+
+  a0 = a & mask;
+  a1 = (a>>m) & mask;
+  a2 = (a>>(2*m)) & mask;
+  a3 = (a>>(3*m));
+
+  b0 =  b         & mask;
+  b1 = (b>>m)     & mask;
+  b2 = (b>>(2*m)) & mask;
+  b3 = (b>>(3*m));
+
+  /*
+    Bodrato / Zanoni
+  */
+
+  w3 = a3 + (a1 + (a2 + a0));
+  w7 = b3 + (b1 + (b2 + b0));
+
+  w4 = -a3 + (-a1 + (a2 + a0));
+  w5 = -b3 + (-b1 + (b2 + b0));
+
+  w3 = toomcook4(w3, w7);
+  w4 = toomcook4(w4, w5);
+
+  w5 = a3 + ((a1<<2) + ((a2<<1) + (a0<<3)));
+  w2 = b3 + ((b1<<2) + ((b2<<1) + (b0<<3)));
+
+  w6 = -a3 + (-(a1<<2) + ((a2<<1) + (a0<<3)));
+  w7 = -b3 + (-(b1<<2) + ((b2<<1) + (b0<<3)));
+
+  w5 = toomcook4(w5, w2);
+  w6 = toomcook4(w6, w7);
+
+
+  w2 = (a3<<3) + ((a1<<1) + ((a2<<2) + a0));
+  w7 = (b3<<3) + ((b1<<1) + ((b2<<2) + b0));
+
+
+  w2 = toomcook4(w2, w7);
+
+  w1 = toomcook4(a3, b3);
+  w7 = toomcook4(a0, b0);
+
+  w2 = w2 + w5;
+  w6 = w5 - w6;
+  w4 = w3 - w4;
+  w5 = w5 - w1;
+  w5 -= w7 << 6;
+  w4 = w4>>1;
+  w3 = w3 - w4;
+  w5 = w5<<1;
+  w5 = w5 - w6;
+  w2 -= w3 * 65;
+  w3 = w3 - w7;
+  w3 = w3 - w1;
+  w2 += w3 * 45;
+  w5 -= w3<<3;
+  w5 = w5//24;
+  w6 =  w6 - w2;
+  w2 -= w4<<4;
+  w2 = w2//18;
+  w3 = w3 - w5;
+  w4 = w4 - w2;
+  w6 += w2 * 30;
+  w6 = w6//60;
+  w2 = w2 - w6;
+
+
+  ret = w7 + (w6<<m) + (w5<<(2*m)) + (w4<<(3*m))+ (w3<<(4*m))+
+        (w2<<(5*m))+ (w1<<(6*m));
+
+  ret = sign *ret;
+
+  return ret;
+}
+
+define toomcook4square(a){
+  local  a0 a1 a2 a3 ret S0 S1 S2 S3 S4 S5 S6 S7 tmp tmp2 tmp3;
+  local  sign m alen mask;
+  local T0 T1 T2 T3 T4 T5 T6 T7 T8;
+
+  if(!isint(a) )return newerror("toomcook3square(a): a is not integer");
+
+  alen = digits(a,2);
+
+  sign = sgn(a) * sgn(a);
+  /* sgn(x) returns 0 if x = 0 */
+  if(sign == 0) return 0;
+
+  m = (alen)//4;
+  mask = ~-( 1 << m );
+
+  /*
+    cut-off at about 2 mio. dec. digits
+    TODO: check!
+  */
+
+  if(isdefined("test8900")){
+    if(m < 100) return toomcook3square(a);
+  }
+  else{
+    if(m < 512*3072) return toomcook3square(a);
+  }
+
+  a = abs(a);
+
+  a0 = a          & mask;
+  a1 = (a>>m)     & mask;
+  a2 = (a>>(2*m)) & mask;
+  a3 = (a>>(3*m)) ;
+
+  /*
+    Bodrato / Zanoni
+  */
+
+  S1 = toomcook4square(a0);
+  S2 = toomcook4(a0<<1,a1);
+  S3 = toomcook4((a0 + a1 - a2 - a3 ) , (a0 - a1 - a2 + a3 ));
+  S4 = toomcook4square(a0 + a1 + a2 + a3 );
+  S5 = toomcook4( (a0 - a2 )<<1 , (a1 - a3 ));
+  S6 = toomcook4(a3<<1 , a2);
+  S7 = toomcook4square(a3);
+
+  T1 = S3 + S4;
+  T2 = (T1 + S5 )>>1;
+  T3 = S2 + S6;
+  T4 = T2 - T3;
+  T5 = T3 - S5;
+  T6 = T4 - S3;
+  T7 = T4 - S1;
+  T8 = T6 - S7;
+
+  ret =   (S7<<(6*m)) + (S6<<(5*m)) + (T7<<(4*m))
+        + (T5<<(3*m)) + (T8<<(2*m)) + (S2<<(1*m))  + S1;
+
+  ret = sign *ret;
+
+  return ret;
+}
+
+/*
+  TODO: Implement the asymmetric variations
+*/
+
+/*
+  produce_long_random_number(n) returns large pseudorandom numbers. Really large
+  numbers, e.g.:
+      produce_long_random_number(16)
+  is ca 4,128,561 bits (ca 1,242,821 dec. digits) large. Exact length is not
+  predeterminable because of the chaotic output of the function random().
+*/
+define __CZ__produce_long_random_number(n)
+{
+  local ret k;
+  ret = 1;
+  if(!isint(n) || n<1)
+    return newerror("__CZ__produce_long_random_number(n): "
+    		    "n is not an integer >=1");
+  for(k=0;k<n;k++){
+    ret += random();
+    ret  = toomcook4square(ret);
+  }
+  return ret;
+}
+
+
+/*
+ * restore internal function from resource debugging
+ * report important interface functions
+ */
+config("resource_debug", resource_debug_level),;
+if (config("resource_debug") & 3) {
+    print "toomcook3(a,b)";
+    print "toomcook3square(a)";
+    print "toomcook4(a,b)";
+    print "toomcook4square(a)";
+}

cal/zeta2.cal

@@ -0,0 +1,116 @@
+/*
+ * zeta2 - Hurwitz Zeta function
+ *
+ * Calc is open software; you can redistribute it and/or modify it under
+ * the terms of the version 2.1 of the GNU Lesser General Public License
+ * as published by the Free Software Foundation.
+ *
+ * Calc is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+ * or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU Lesser General
+ * Public License for more details.
+ *
+ * A copy of version 2.1 of the GNU Lesser General Public License is
+ * distributed with calc under the filename COPYING-LGPL.  You should have
+ * received a copy with calc; if not, write to Free Software Foundation, Inc.
+ * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
+ *
+ * @(#) $Revision: 30.4 $
+ * @(#) $Id: zeta2.cal,v 30.4 2013/08/18 20:01:53 chongo Exp $
+ * @(#) $Source: /usr/local/src/bin/calc/cal/RCS/zeta2.cal,v $
+ *
+ * Under source code control:	2013/08/11 01:31:28
+ * File existed as early as:	2013
+ */
+
+
+/*
+ * hide internal function from resource debugging
+ */
+static resource_debug_level;
+resource_debug_level = config("resource_debug", 0);
+
+
+define hurwitzzeta(s,a){
+   local realpart_a  imagpart_s  tmp tmp1 tmp2 tmp3;
+   local sum1 sum2 sum3 i k n precision result limit;
+   local limit_function offset offset_squared rest_sum eps;
+  /*
+    According to Linas Vepstas' "An efficient algorithm for accelerating
+    the convergence of oscillatory series, useful for computing the
+    polylogarithm and Hurwitz zeta functions" the Euler-Maclaurin series
+    is the fastest in most cases.
+
+    With a lot of help of the PARI/GP implementation by Prof. Henri Cohen,
+    hence the different license.
+  */
+   eps=epsilon( epsilon() * 1e-3);
+   realpart_a=re(a);
+   if(realpart_a>1.5){
+    tmp=floor(realpart_a-0.5);
+    sum1 = 0;
+    for( i = 1 ; i <= tmp ; i++){
+      sum1 += ( a - i )^( -s );
+    }
+    epsilon(eps);
+    return (hurwitzzeta(s,a-tmp)-sum1);
+   }
+   if(realpart_a<=0){
+    tmp=ceil(-realpart_a+0.5);
+    for( i = 0 ; i <= tmp-1 ; i++){
+      sum2 += ( a + i )^( -s );
+    }
+    epsilon(eps);
+    return (hurwitzzeta(s,a+tmp)+sum2);
+   }
+   precision=digits(1/epsilon());
+   realpart_a=re(s);
+   imagpart_s=im(s);
+   epsilon(1e-9);
+   result=s-1.;
+   if(abs(result)<0.1){
+     result=-1;
+   }
+   else
+     result=ln(result);
+  limit=(precision*ln(10)-re((s-.5)*result)+(1.*realpart_a)*ln(2.*pi()))/2;
+  limit=max(2,ceil(max(limit,abs(s*1.)/2)));
+  limit_function=ceil(sqrt((limit+realpart_a/2-.25)^2+(imagpart_s*1.)^2/4)/
+  		      pi());
+  if (config("user_debug") > 0) {
+     print "limit_function = " limit_function;
+     print "limit = " limit;
+     print "prec = " precision;
+  }
+  /* Full precison plus 5 digits angstzuschlag*/
+  epsilon( (10^(-precision)) * 1e-5);
+  tmp3=(a+limit_function+0.)^(-s);
+  sum3 = tmp3/2;
+  for(n=0;n<=limit_function-1;n++){
+    sum3 += (a+n)^(-s);
+  }
+  result=sum3;
+  offset=a+limit_function;
+  offset_squared=1./(offset*offset);
+  tmp1=2*s-1;
+  tmp2=s*(s-1);
+  rest_sum=bernoulli(2*limit);
+  for(k=2*limit-2;k>=2;k-=2){
+    rest_sum=bernoulli(k)+offset_squared*
+      (k*k+tmp1*k+tmp2)*rest_sum/((k+1)*(k+2));
+  }
+  rest_sum=offset*(1+offset_squared*tmp2*rest_sum/2);
+  result+=rest_sum*tmp3/(s-1);
+  epsilon(eps);
+  return result;
+}
+
+
+/*
+ * restore internal function from resource debugging
+ * report important interface functions
+ */
+config("resource_debug", resource_debug_level),;
+if (config("resource_debug") & 3) {
+    print "hurwitzzeta(s,a)";
+}

calc.h

@@ -1,7 +1,7 @@
 /*
  * calc - definitions for calculator program
  *
- * Copyright (C) 1999-2007  David I. Bell
+ * Copyright (C) 1999-2007,2014  David I. Bell
  *
  * Calc is open software; you can redistribute it and/or modify it under
  * the terms of the version 2.1 of the GNU Lesser General Public License
@@ -17,8 +17,8 @@
  * received a copy with calc; if not, write to Free Software Foundation, Inc.
  * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
  *
- * @(#) $Revision: 30.2 $
- * @(#) $Id: calc.h,v 30.2 2007/07/10 17:44:52 chongo Exp $
+ * @(#) $Revision: 30.6 $
+ * @(#) $Id: calc.h,v 30.6 2014/08/24 21:56:51 chongo Exp $
  * @(#) $Source: /usr/local/src/bin/calc/RCS/calc.h,v $
  *
  * Under source code control:	1990/02/15 01:48:31
@@ -28,8 +28,8 @@
  */
 
 
-#if !defined(__CALC_H__)
-#define __CALC_H__
+#if !defined(INCLUDE_CALC_H)
+#define INCLUDE_CALC_H
 
 #include <setjmp.h>
 #if defined(CALC_SRC)	/* if we are building from the calc source tree */
@@ -140,7 +140,8 @@ E_FUNC int ftellid(FILEID id, ZVALUE *res);
 E_FUNC int fseekid(FILEID id, ZVALUE offset, int whence);
 E_FUNC int isattyid(FILEID id);
 E_FUNC int fsearch(FILEID id, char *str, ZVALUE start, ZVALUE end, ZVALUE *res);
-E_FUNC int frsearch(FILEID id, char *str, ZVALUE first, ZVALUE last, ZVALUE *res);
+E_FUNC int frsearch(FILEID id, char *str, ZVALUE first, ZVALUE last,
+		    ZVALUE *res);
 E_FUNC void showconstants(void);
 E_FUNC void freeconstant(unsigned long);
 E_FUNC void freestringconstant(long);
@@ -151,7 +152,8 @@ E_FUNC void trimconstants(void);
  */
 E_FUNC int openstring(char *str, size_t num);
 E_FUNC int openterminal(void);
-E_FUNC int opensearchfile(char *name, char *pathlist, char *exten, int reopen_ok);
+E_FUNC int opensearchfile(char *name, char *pathlist, char *exten,
+			  int reopen_ok);
 E_FUNC char *nextline(void);
 E_FUNC int nextchar(void);
 E_FUNC void reread(void);
@@ -265,4 +267,4 @@ EXTERN char *Copyright;
 E_FUNC char *version(void);
 
 
-#endif /* !__CALC_H__ */
+#endif /* !INCLUDE_CALC_H */

calc.man

@@ -1,5 +1,5 @@
 .\"
-.\" Copyright (C) 1999-2007	Landon Curt Noll
+.\" Copyright (C) 1999-2007,2014	Landon Curt Noll
 .\"
 .\" Calc is open software; you can redistribute it and/or modify it under
 .\" the terms of the version 2.1 of the GNU Lesser General Public License
@@ -15,8 +15,8 @@
 .\" received a copy with calc; if not, write to Free Software Foundation, Inc.
 .\" 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
 .\"
-.\" @(#) $Revision: 30.3 $
-.\" @(#) $Id: calc.man,v 30.3 2007/09/08 02:53:09 chongo Exp $
+.\" @(#) $Revision: 30.6 $
+.\" @(#) $Id: calc.man,v 30.6 2014/10/06 08:39:45 chongo Exp $
 .\" @(#) $Source: /usr/local/src/bin/calc/RCS/calc.man,v $
 .\"
 .\" Under source code control:	    1991/07/23 05:48:26
@@ -1115,45 +1115,62 @@ To contribute comments, suggestions, enhancements
 and interesting
 .B calc
 resource files, and
-shell scripts please join the low volume calc mailing list.
+shell scripts please join the calc-tester
+low volume moderated calc mailing list.
 .sp
-To join the low volume calc mailing list, send EMail to:
+To the calc-tester mailing list, visit the following URL:
 .sp
 .in +0.5i
 .nf
-calc-tester-request at asthe dot com
+https://www.listbox.com/subscribe/?list_id=239342
 .fi
 .in -0.5i
 .sp
-Your subject must contain the words:
+To help determine you are a human and not just a spam bot,
+you will be required to provide the following additional information:
 .sp
 .in +0.5i
 .nf
-calc mailing list subscription
+Your Name
+
+Calc Version
+   For example, the current version is: ${VERSION}
+
+Operating System
+   If you don't know your operating system, enter: unknown
+
+The date 7 days ago
+   Consult a calendar :-)
+.fi
+.in -0.5i
+.sp
+If you need a human to help you with your mailing list subscription,
+or if you have problems with the above procedure, please
+send EMail to our special address:
+.sp
+.in +0.5i
+.nf
+calc-tester-maillist-help at asthe dot com
+
+NOTE: Remove spaces and replace 'at' with @, 'dot' with .
+
+NOTE: Yes, the EMail address uses 'asthe',
+      while the web site uses 'isthe'.
+.fi
+.in -0.5i
+.sp
+To be sure we see your EMail asking for help with your mailing list
+subscription, please use the following phase in your EMail Subject line
+your subject must contain the words:
+.sp
+.in +0.5i
+.nf
+calc tester mailing list help
 .fi
 .in -0.5i
 .sp
 You may have additional words in your subject line.
 .sp
-Your message body must contain:
-.sp
-.in +0.5i
-.nf
-subscribe calc-tester address
-end
-name your_full_name
-.fi
-.in -0.5i
-.sp
-where
-.B address
-s your EMail address and
-.B your_full_name
-is your full name.
-Feel free to follow the
-.B name
-line with additional EMail text as desired.
-.sp
 .SH "BUG REPORTS / BUG FIXES"
 \&
 .br
@@ -1162,10 +1179,12 @@ Send bug reports and bug fixes to:
 .sp
 .in +0.5i
 .nf
-calc-bugs at asthe dot com
+calc-bug-report at asthe dot com
 
-[[ NOTE: Replace 'at' with @, 'dot' is with . and remove the spaces ]]
-[[ NOTE: The EMail address uses 'asthe' and the web site URL uses 'isthe' ]]
+NOTE: Remove spaces and replace 'at' with @, 'dot' with .
+
+NOTE: Yes, the EMail address uses 'asthe',
+      while the web site uses 'isthe'.
 .fi
 .in -0.5i
 .sp
@@ -1178,6 +1197,10 @@ calc bug report
 .in -0.5i
 .sp
 You may have additional words in your subject line.
+.in
+However, you may find it more helpful to simply subscribe to the
+calc-tester mailing list (see above) and then to send your report
+to that mailing list as a wider set calc testers may be able to help you.
 .sp
 See the
 .I BUGS
@@ -1196,7 +1219,7 @@ for more information about bug reporting.
 .SH "CALC WEB SITE"
 \&
 .br
-Landon Noll maintains the the
+Landon Noll maintains the
 .B calc
 web site is located at:
 .sp

calc.spec.in

@@ -1,8 +1,7 @@
-#****h* calc/calc.spec.in
 #
 # calc.spec.in - template specfile for calc
 #
-# Copyright (C) 2003-2013  Petteri Kettunen and Landon Curt Noll
+# Copyright (C) 2003-2014  Petteri Kettunen and Landon Curt Noll
 #
 # Calc is open software; you can redistribute it and/or modify it under
 # the terms of the version 2.1 of the GNU Lesser General Public License
@@ -18,8 +17,8 @@
 # received a copy with calc; if not, write to Free Software Foundation, Inc.
 # 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
 #
-# @(#) $Revision: 30.10 $
-# @(#) $Id: calc.spec.in,v 30.10 2013/05/05 11:57:44 chongo Exp $
+# @(#) $Revision: 30.23 $
+# @(#) $Id: calc.spec.in,v 30.23 2014/09/22 05:01:58 chongo Exp $
 # @(#) $Source: /usr/local/src/bin/calc/RCS/calc.spec.in,v $
 #
 # Under source code control:	2003/02/16 20:21:39
@@ -31,13 +30,13 @@
 # BUGS
 #  - Uninstalling calc and calc-devel leaves empty dirs /usr/include/calc
 #    and /usr/share/calc and its subdirs. In case e.g. %{_includedir}/calc
-#    is defined in `%files devel' section, then rpmbuild complains that
+#    is defined in `files devel' section, then rpmbuild complains that
 #    header files are defined twice - rpmbuild bug or bug in specfile conf???
 
 Summary: Arbitrary precision calculator.
 Name: calc
 Version: <<<PROJECT_VERSION>>>
-Release: 10
+Release: 12
 License: LGPL
 Group: Applications/Engineering
 Source: http://www.isthe.com/chongo/src/calc/%{name}-%{version}.tar.bz2
@@ -46,6 +45,8 @@ Vendor: Landon Noll and Associates
 Packager: Landon Noll and Associates (http://www.isthe.com/chongo/index.html)
 Requires: ncurses >= 5.5-24, readline >= 5.1-3, less >= 358
 BuildRequires: ncurses-devel >= 5.5-24, readline-devel >= 5.1-3
+Provides: libcalc.so.%{version}()(64bit)
+Provides: libcustcalc.so.%{version}()(64bit)
 BuildRoot: %{_tmppath}/build-root
 
 %description
@@ -61,7 +62,10 @@ For the latest calc release, see the calc project home page:
 %package devel
 Summary: Development files and documentation for calc.
 Group: Applications/Engineering
-Requires: calc = %{version}-%{release}
+Requires: calc(x86-64) = %{version}-%{release}
+Requires: libcalc.so.%{version}()(64bit)
+Requires: libcustcalc.so.%{version}()(64bit)
+Requires: ncurses >= 5.5-24, readline >= 5.1-3, less >= 358
 
 %description devel
 This package contains the header files and static libraries for developing
@@ -76,42 +80,67 @@ For the latest calc release, see the project home page:
 
 %build
 echo '-=- calc.spec beginning make clobber -=-'
-make %{?_smp_mflags} T=%{_buildroot} BINDIR=%{_bindir} LIBDIR=%{_libdir} CALC_SHAREDIR=%{_datadir}/%{name} CALC_INCDIR=%{_includedir}/calc MANDIR=%{_mandir}/man1 EXT= V=@ clobber
+make -j1 T=%{_buildroot} BINDIR=%{_bindir} LIBDIR=%{_libdir} \
+     CALC_SHAREDIR=%{_datadir}/%{name} CALC_INCDIR=%{_includedir}/calc \
+     MANDIR=%{_mandir}/man1 EXT= V=@ clobber
 echo '-=- calc.spec ending make clobber -=-'
 echo '-=- calc.spec beginning make calc-static-only -=-'
-make %{?_smp_mflags} T=%{_buildroot} BINDIR=%{_bindir} LIBDIR=%{_libdir} CALC_SHAREDIR=%{_datadir}/%{name} CALC_INCDIR=%{_includedir}/calc MANDIR=%{_mandir}/man1 EXT= V=@ calc-static-only BLD_TYPE=calc-static-only
+make -j1 T=%{_buildroot} BINDIR=%{_bindir} LIBDIR=%{_libdir} \
+     CALC_SHAREDIR=%{_datadir}/%{name} CALC_INCDIR=%{_includedir}/calc \
+     MANDIR=%{_mandir}/man1 EXT= V=@ calc-static-only BLD_TYPE=calc-static-only
 echo '-=- calc.spec ending make calc-static-only -=-'
 echo '-=- calc.spec beginning make rpm-hide-static -=-'
-make %{?_smp_mflags} T=%{_buildroot} BINDIR=%{_bindir} LIBDIR=%{_libdir} CALC_SHAREDIR=%{_datadir}/%{name} CALC_INCDIR=%{_includedir}/calc MANDIR=%{_mandir}/man1 EXT= V=@ rpm-hide-static
+make -j1 T=%{_buildroot} BINDIR=%{_bindir} LIBDIR=%{_libdir} \
+     CALC_SHAREDIR=%{_datadir}/%{name} CALC_INCDIR=%{_includedir}/calc \
+     MANDIR=%{_mandir}/man1 EXT= V=@ rpm-hide-static
 echo '-=- calc.spec ending make rpm-hide-static -=-'
 echo '-=- calc.spec beginning make clobber (again) -=-'
-make %{?_smp_mflags} T=%{_buildroot} BINDIR=%{_bindir} LIBDIR=%{_libdir} CALC_SHAREDIR=%{_datadir}/%{name} CALC_INCDIR=%{_includedir}/calc MANDIR=%{_mandir}/man1 EXT= V=@ clobber
+make -j1 T=%{_buildroot} BINDIR=%{_bindir} LIBDIR=%{_libdir} \
+     CALC_SHAREDIR=%{_datadir}/%{name} CALC_INCDIR=%{_includedir}/calc \
+     MANDIR=%{_mandir}/man1 EXT= V=@ clobber
 echo '-=- calc.spec ending make clobber (again) -=-'
 echo '-=- calc.spec beginning make calc-dynamic-only -=-'
-make %{?_smp_mflags} T=%{_buildroot} BINDIR=%{_bindir} LIBDIR=%{_libdir} CALC_SHAREDIR=%{_datadir}/%{name} CALC_INCDIR=%{_includedir}/calc MANDIR=%{_mandir}/man1 EXT= V=@ calc-dynamic-only BLD_TYPE=calc-dynamic-only LD_SHARE=
+make -j1 T=%{_buildroot} BINDIR=%{_bindir} LIBDIR=%{_libdir} \
+     CALC_SHAREDIR=%{_datadir}/%{name} CALC_INCDIR=%{_includedir}/calc \
+     MANDIR=%{_mandir}/man1 EXT= V=@ calc-dynamic-only \
+     BLD_TYPE=calc-dynamic-only LD_SHARE=
 echo '-=- calc.spec ending make calc-dynamic-only -=-'
 echo '-=- calc.spec beginning make chk -=-'
-make %{?_smp_mflags} T=%{_buildroot} BINDIR=%{_bindir} LIBDIR=%{_libdir} CALC_SHAREDIR=%{_datadir}/%{name} CALC_INCDIR=%{_includedir}/calc MANDIR=%{_mandir}/man1 EXT= V=@ chk
+make -j1 T=%{_buildroot} BINDIR=%{_bindir} LIBDIR=%{_libdir} \
+     CALC_SHAREDIR=%{_datadir}/%{name} CALC_INCDIR=%{_includedir}/calc \
+     MANDIR=%{_mandir}/man1 EXT= V=@ chk
 echo '-=- calc.spec ending make chk -=-'
 echo '-=- calc.spec beginning make rpm-unhide-static -=-'
-make %{?_smp_mflags} T=%{_buildroot} BINDIR=%{_bindir} LIBDIR=%{_libdir} CALC_SHAREDIR=%{_datadir}/%{name} CALC_INCDIR=%{_includedir}/calc MANDIR=%{_mandir}/man1 EXT= V=@ rpm-unhide-static
+make -j1 T=%{_buildroot} BINDIR=%{_bindir} LIBDIR=%{_libdir} \
+     CALC_SHAREDIR=%{_datadir}/%{name} CALC_INCDIR=%{_includedir}/calc \
+     MANDIR=%{_mandir}/man1 EXT= V=@ rpm-unhide-static
 echo '-=- calc.spec ending make rpm-unhide-static -=-'
 echo '-=- calc.spec beginning make rpm-clean-static -=-'
-make %{?_smp_mflags} T=%{_buildroot} BINDIR=%{_bindir} LIBDIR=%{_libdir} CALC_SHAREDIR=%{_datadir}/%{name} CALC_INCDIR=%{_includedir}/calc MANDIR=%{_mandir}/man1 EXT= V=@ rpm-clean-static
+make -j1 BINDIR=%{_bindir} LIBDIR=%{_libdir} \
+     CALC_SHAREDIR=%{_datadir}/%{name} CALC_INCDIR=%{_includedir}/calc \
+     MANDIR=%{_mandir}/man1 EXT= V=@ rpm-clean-static
 echo '-=- calc.spec ending make rpm-clean-static -=-'
 echo '-=- calc.spec beginning make rpm-chk-static -=-'
-make %{?_smp_mflags} T=%{_buildroot} BINDIR=%{_bindir} LIBDIR=%{_libdir} CALC_SHAREDIR=%{_datadir}/%{name} CALC_INCDIR=%{_includedir}/calc MANDIR=%{_mandir}/man1 EXT= V=@ rpm-chk-static
+make -j1 T=%{_buildroot} BINDIR=%{_bindir} LIBDIR=%{_libdir} \
+     CALC_SHAREDIR=%{_datadir}/%{name} CALC_INCDIR=%{_includedir}/calc \
+     MANDIR=%{_mandir}/man1 EXT= V=@ rpm-chk-static
 echo '-=- calc.spec ending make rpm-chk-static -=-'
 
 %install
 echo '-=- calc.spec beginning make install -=-'
-rm -rf %{_buildroot}
 mkdir -p %{_buildroot}
-make T=%{_buildroot} BINDIR=%{_bindir} LIBDIR=%{_libdir} CALC_SHAREDIR=%{_datadir}/%{name} CALC_INCDIR=%{_includedir}/calc MANDIR=%{_mandir}/man1 EXT= V=@ install
+make -j1 T=%{_buildroot} BINDIR=%{_bindir} LIBDIR=%{_libdir} \
+     CALC_SHAREDIR=%{_datadir}/%{name} CALC_INCDIR=%{_includedir}/calc \
+     MANDIR=%{_mandir}/man1 EXT= V=@ install
 echo '-=- calc.spec ending make install -=-'
 
 %clean
-rm -rf %{_buildroot}
+echo '-=- calc.spec beginning make clean -=-'
+rm -rf %{_tmppath}
+rm -rf %{_builddir}
+rm -rf %{_specdir}
+rm -rf %{_buildrootdir}
+echo '-=- calc.spec beginning make clean -=-'
 
 %files
 %defattr(-, root, root)
@@ -125,11 +154,11 @@ rm -rf %{_buildroot}
 %attr(644, root, root) %{_datadir}/%{name}/custom/*.cal
 %attr(644, root, root) %{_datadir}/%{name}/help/*
 %attr(644, root, root) %{_datadir}/%{name}/*.cal
-%attr(644, root, root) %{_datadir}/%{name}/*.line
+%attr(644, root, root) %{_datadir}/%{name}/set8700.line
 %attr(644, root, root) %{_libdir}/libcalc.so
-%attr(644, root, root) %{_libdir}/libcalc.so.*
+%attr(644, root, root) %{_libdir}/libcalc.so.%{version}
 %attr(644, root, root) %{_libdir}/libcustcalc.so
-%attr(644, root, root) %{_libdir}/libcustcalc.so.*
+%attr(644, root, root) %{_libdir}/libcustcalc.so.%{version}
 
 %files devel
 %defattr(-, root, root)
@@ -140,7 +169,27 @@ rm -rf %{_buildroot}
 %attr(644, root, root) %{_libdir}/libcustcalc.a
 
 %changelog
-* Sat May 05 2013 Landon Curt Noll http://www.isthe.com/chongo
+* Mon Sep 01 2014 Landon Curt Noll http://www.isthe.com/chongo
+- Release: 12
+- Removed use of %{?_smp_mflags}.  On Ubuntu 14.04, the -j2
+  was executing make lines within a rule out of order. Always
+  use -j1 while building these RPMs to avoid this make bug.
+  This only impacts the speed of building the rpms.
+- Stopped removing %{_buildroot} at the beginning of install
+  as this, acording to a number of sources, was asking for trouble.
+- Clean removes tmp, BUILD, SPEC, and BUILDROOT dirs under topdir.
+- Fixed the building of the calc-debuginfo rpm.
+- Eliminated rpmbuild "bogus date" warnings due to inconsistent weekday,
+  by assuming the date is correct and changing the weekday via BogusDateBot.
+  Sun May 20 2006 --> Sun May 14 2006
+  Sun Sep 01 2007 --> Sun Aug 26 2007
+  Sat May 05 2013 --> Sat May 04 2013
+
+* Sun Sep 01 2013 Landon Curt Noll http://www.isthe.com/chongo
+- Release: 11
+- only list %{_datadir}/%{name}/set8700.line not *.line
+
+* Sun May 05 2013 Landon Curt Noll http://www.isthe.com/chongo
 - Release: 10
 - Major bug rpm build process bug fix
 - Source: is a URL as per rpm requirements
@@ -150,37 +199,46 @@ rm -rf %{_buildroot}
 - Fixed use of %{_buildroot}
 - Fixed permissions so that rpm build services can modify libraries
 - Added missing files to files section
+
 * Mon Mar 25 2013 Landon Curt Noll http://www.isthe.com/chongo
 - Release: 9 (was 1.1)
 - Removed deprecated PreReq for calc-devel, using Requires
-* Sun Sep 01 2007 Landon Curt Noll http://www.isthe.com/chongo
+
+* Sat Sep 01 2007 Landon Curt Noll http://www.isthe.com/chongo
 - Release: 8
 - Release of calc-2.12.2
 - Calc builds with shared libraries
+
 * Sun Jun 25 2006 Landon Curt Noll http://www.isthe.com/chongo
 - Release: 7
 - Changed Copyright to License as per new rpm v4.4 syntax
-* Sun May 20 2006 Landon Curt Noll http://www.isthe.com/chongo
+
+* Sat May 20 2006 Landon Curt Noll http://www.isthe.com/chongo
 - Release: 6
 - Release of calc-2.12.0
 - Added *.line set files to the list of packaged files
+
 * Sun Dec 11 2005 Landon Curt Noll http://www.isthe.com/chongo
 - Release: 5
 - Release of calc-2.11.11
 - Fixed description in spec file
+
 * Wed Feb 26 2003 Landon Curt Noll http://www.isthe.com/chongo
 - Release: 4
 - Release of calc-2.11.7-2
 - Fixed attributes on include and lib calc-devel files
 - Added BUGS to calc-devel as well as calc
+
 * Tue Feb 25 2003 Landon Curt Noll http://www.isthe.com/chongo
 - Release: 3
 - Release of calc-2.11.7-1
 - Require ncurses, readline and less to install.
 - Require ncurses-devel and readline-devel to build.
+
 * Tue Feb 18 2003 Landon Curt Noll http://www.isthe.com/chongo
 - Release: 2
 - Misc changes to fit local directory setup
+
 * Sun Feb 16 2003 Petteri Kettunen <petterik@users.sourceforge.net>
 - Release: 1
 - initial RPM build

calcerr.tbl

@@ -17,8 +17,8 @@
 # received a copy with calc; if not, write to Free Software Foundation, Inc.
 # 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
 #
-# @(#) $Revision: 30.1 $
-# @(#) $Id: calcerr.tbl,v 30.1 2007/03/16 11:09:46 chongo Exp $
+# @(#) $Revision: 30.4 $
+# @(#) $Id: calcerr.tbl,v 30.4 2013/09/02 01:38:08 chongo Exp $
 # @(#) $Source: /usr/local/src/bin/calc/RCS/calcerr.tbl,v $
 #
 # Under source code control:	1996/05/23 17:38:44
@@ -32,6 +32,16 @@
 #
 #	SYMBOL	meaning
 
+######################################################################
+# WARNING: The order of the lines below is critical!  If you change  #
+#	   the order, you will break code that depends on the return #
+#	   value of the errno() builtin function.		     #
+#								     #
+#	   If you need to add values to this table, ONLY add then    #
+#	   to the bottom of the file!				     #
+######################################################################
+
+
 E_1OVER0	Division by zero
 E_0OVER0	Indeterminate (0/0)
 E_ADD		Bad arguments for +
@@ -449,3 +459,19 @@ E_FGETFILE1	Non-file argument for fgetfile
 E_FGETFILE2	File argument for fgetfile not open for reading
 E_FGETFILE3	Unable to set file position in fgetfile
 E_ESTR		Non-representable type for estr
+E_STRCASECMP	Non-string argument for strcasecmp
+E_STRNCASECMP	Bad argument type for strncasecmp
+E_ISUPPER	Bad argument for isupper
+E_ISLOWER	Bad argument for islower
+E_ISALNUM	Bad argument for isalnum
+E_ISALPHA	Bad argument for isalpha
+E_ISASCII	Bad argument for isascii
+E_ISCNTRL	Bad argument for iscntrl
+E_ISDIGIT	Bad argument for isdigit
+E_ISGRAPH	Bad argument for isgraph
+E_ISPRINT	Bad argument for isprint
+E_ISPUNCT	Bad argument for ispunct
+E_ISSPACE	Bad argument for isspace
+E_ISXDIGIT	Bad argument for isxdigit
+E_STRTOUPPER    Bad argument type for strtoupper
+E_STRTOLOWER    Bad argument type for strtolower

check.awk

@@ -18,8 +18,8 @@
 # received a copy with calc; if not, write to Free Software Foundation, Inc.
 # 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
 #
-# @(#) $Revision: 30.1 $
-# @(#) $Id: check.awk,v 30.1 2007/03/16 11:09:46 chongo Exp $
+# @(#) $Revision: 30.2 $
+# @(#) $Id: check.awk,v 30.2 2013/08/11 01:08:32 chongo Exp $
 # @(#) $Source: /usr/local/src/bin/calc/RCS/check.awk,v $
 #
 # Under source code control:	1996/05/25 22:07:58
@@ -30,7 +30,7 @@
 #
 # This awk script will print 3 lines before and after any non-blank line that
 # does not begin with a number.	 This allows the 'make debug' rule to remove
-# all non-interest lines the the 'make check' regression output while providing
+# all non-interest lines the 'make check' regression output while providing
 # 3 lines of context around unexpected output.
 #
 BEGIN {

cmath.h

@@ -1,7 +1,7 @@
 /*
  * cmath - data structures for extended precision complex arithmetic
  *
- * Copyright (C) 1999-2007  David I. Bell
+ * Copyright (C) 1999-2007,2014  David I. Bell
  *
  * Calc is open software; you can redistribute it and/or modify it under
  * the terms of the version 2.1 of the GNU Lesser General Public License
@@ -17,8 +17,8 @@
  * received a copy with calc; if not, write to Free Software Foundation, Inc.
  * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
  *
- * @(#) $Revision: 30.1 $
- * @(#) $Id: cmath.h,v 30.1 2007/03/16 11:09:46 chongo Exp $
+ * @(#) $Revision: 30.3 $
+ * @(#) $Id: cmath.h,v 30.3 2014/08/24 21:56:51 chongo Exp $
  * @(#) $Source: /usr/local/src/bin/calc/RCS/cmath.h,v $
  *
  * Under source code control:	1993/07/30 19:42:45
@@ -28,8 +28,8 @@
  */
 
 
-#if !defined(__CMATH_H__)
-#define __CMATH_H__
+#if !defined(INCLUDE_CMATH_H)
+#define INCLUDE_CMATH_H
 
 
 #if defined(CALC_SRC)	/* if we are building from the calc source tree */
@@ -156,4 +156,4 @@ E_FUNC COMPLEX *swap_HALF_in_COMPLEX(COMPLEX *dest, COMPLEX *src, BOOL all);
 EXTERN COMPLEX _czero_, _cone_, _conei_;
 
 
-#endif /* !__CMATH_H__ */
+#endif /* !INCLUDE_CMATH_H */

codegen.c

@@ -1,7 +1,7 @@
 /*
  * codegen - module to generate opcodes from the input tokens
  *
- * Copyright (C) 1999-2007  David I. Bell and Ernest Bowen
+ * Copyright (C) 1999-2007,2017  David I. Bell and Ernest Bowen
  *
  * Primary author:  David I. Bell
  *
@@ -19,9 +19,9 @@
  * received a copy with calc; if not, write to Free Software Foundation, Inc.
  * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
  *
- * @(#) $Revision: 30.2 $
- * @(#) $Id: codegen.c,v 30.2 2007/07/05 13:30:38 chongo Exp $
- * @(#) $Source: /usr/local/src/bin/calc/RCS/codegen.c,v $
+ * @(#) $Revision: 30.5 $
+ * @(#) $Id: codegen.c,v 30.5 2017/05/19 16:09:14 chongo Exp $
+ * @(#) $Source: /usr/local/src/bin/calc-RHEL7/RCS/codegen.c,v $
  *
  * Under source code control:	1990/02/15 01:48:13
  * File existed as early as:	before 1990
@@ -148,7 +148,11 @@ getcommands(BOOL toplevel)
 				case 1:
 				case -1:
 					if(i == 1) {
-						strcpy(name, DEFAULTCALCHELP);
+						strncpy(name,
+							DEFAULTCALCHELP,
+							MAXCMD);
+						/* paranoia */
+						name[MAXCMD] = '\0';
 						givehelp(name);
 					}
 					break;
@@ -453,7 +457,8 @@ getsimplebody(void)
  */
 /*ARGSUSED*/
 S_FUNC void
-getbody(LABEL *contlabel, LABEL *breaklabel, LABEL *nextcaselabel, LABEL *defaultlabel)
+getbody(LABEL *contlabel, LABEL *breaklabel, LABEL *nextcaselabel,
+        LABEL *defaultlabel)
 {
 	int oldmode;
 
@@ -617,7 +622,8 @@ getonevariable(int symtype)
  *	defaultlabel		label for default case
  */
 S_FUNC void
-getstatement(LABEL *contlabel, LABEL *breaklabel, LABEL *nextcaselabel, LABEL *defaultlabel)
+getstatement(LABEL *contlabel, LABEL *breaklabel,
+	     LABEL *nextcaselabel, LABEL *defaultlabel)
 {
 	LABEL label;
 	LABEL label1, label2, label3, label4;	/* locations for jumps */
@@ -1104,7 +1110,8 @@ getobjdeclaration(int symtype)
 					newindices = (int *) malloc(maxindices *
 						sizeof(int));
 					if (newindices == NULL) {
-						scanerror(T_SEMICOLON, "Out of memory for indices malloc");
+						scanerror(T_SEMICOLON,
+					    "Out of memory for indices malloc");
 						(void) tokenmode(oldmode);
 						return;
 					}
@@ -1117,7 +1124,8 @@ getobjdeclaration(int symtype)
 						maxindices * sizeof(int));
 					if (newindices == NULL) {
 						free(indices);
-						scanerror(T_SEMICOLON, "Out of memory for indices realloc");
+						scanerror(T_SEMICOLON,
+					  "Out of memory for indices realloc");
 						(void) tokenmode(oldmode);
 						return;
 					}
@@ -1129,7 +1137,9 @@ getobjdeclaration(int symtype)
 				if (indices[i] == index) {
 					if (indices != quickindices)
 						free(indices);
-					scanerror(T_SEMICOLON, "Duplicate element name \"%s\"", tokensymbol());
+					scanerror(T_SEMICOLON,
+					    "Duplicate element name \"%s\"",
+					    tokensymbol());
 					(void) tokenmode(oldmode);
 					return;
 				}
@@ -1141,7 +1151,8 @@ getobjdeclaration(int symtype)
 			if (gettoken() != T_RIGHTBRACE) {
 				if (indices != quickindices)
 					free(indices);
-				scanerror(T_SEMICOLON, "Bad object type definition");
+				scanerror(T_SEMICOLON,
+				    "Bad object type definition");
 				(void) tokenmode(oldmode);
 				return;
 			}
@@ -1415,7 +1426,8 @@ getinitlist(void)
 	oldmode = tokenmode(TM_DEFAULT);
 
 	if (gettoken() != T_LEFTBRACE) {
-		scanerror(T_SEMICOLON, "Missing left brace for initialization list");
+		scanerror(T_SEMICOLON,
+		    "Missing left brace for initialization list");
 		(void) tokenmode(oldmode);
 		return -1;
 	}
@@ -1450,7 +1462,7 @@ getinitlist(void)
 
 		default:
 			scanerror(T_SEMICOLON,
-				  "Missing right brace for initialization list");
+			      "Missing right brace for initialization list");
 			(void) tokenmode(oldmode);
 			return -1;
 		}
@@ -2364,11 +2376,13 @@ getfilename(char *name, size_t namelen, BOOL *once)
 						/* use symbol VALUE string */
 						symstr = val.v_str->s_str;
 						if (symstr == NULL) {
-							math_error("string value pointer is NULL!!");
+							math_error(
+					"string value pointer is NULL!!");
 							/*NOTREACHED*/
 						}
 					} else {
-						math_error("a filename variable must be a string");
+						math_error(
+				    "a filename variable must be a string");
 						/*NOTREACHED*/
 					}
 				} else {
@@ -2605,7 +2619,7 @@ definesymbol(char *name, int symtype)
 		break;
 	case SYM_GLOBAL:
 		if (symtype == SYM_GLOBAL && conf->redecl_warn) {
-			warning("redeclaraion of global \"%s\"",
+			warning("redeclaration of global \"%s\"",
 				name);
 			return;
 		}
@@ -2613,7 +2627,7 @@ definesymbol(char *name, int symtype)
 
 	case SYM_LOCAL:
 		if (symtype == SYM_LOCAL && conf->redecl_warn) {
-			warning("redeclaraion of local \"%s\"",
+			warning("redeclaration of local \"%s\"",
 				name);
 			return;
 		}
@@ -2637,7 +2651,8 @@ definesymbol(char *name, int symtype)
 			break;
 		}
 		if (conf->dupvar_warn) {
-			warning("both static and parameter \"%s\" defined", name);
+			warning("both static and parameter \"%s\" defined",
+			        name);
 		}
 	}
 	if (symtype == SYM_LOCAL)

config.c

@@ -19,8 +19,8 @@
  * received a copy with calc; if not, write to Free Software Foundation, Inc.
  * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
  *
- * @(#) $Revision: 30.2 $
- * @(#) $Id: config.c,v 30.2 2007/07/05 13:30:38 chongo Exp $
+ * @(#) $Revision: 30.3 $
+ * @(#) $Id: config.c,v 30.3 2013/08/11 08:41:38 chongo Exp $
  * @(#) $Source: /usr/local/src/bin/calc/RCS/config.c,v $
  *
  * Under source code control:	1991/07/20 00:21:56
@@ -948,7 +948,8 @@ setconfig(int type, VALUE *vp)
 		} else if (vp->v_type == V_STR) {
 			temp = lookup_long(truth, vp->v_str->s_str);
 			if (temp < 0) {
-				math_error("Illegal truth value for redecl_warn");
+				math_error("Illegal truth value for "
+					   "redecl_warn");
 				/*NOTREACHED*/
 			}
 			conf->redecl_warn = (int)temp;
@@ -962,7 +963,8 @@ setconfig(int type, VALUE *vp)
 		} else if (vp->v_type == V_STR) {
 			temp = lookup_long(truth, vp->v_str->s_str);
 			if (temp < 0) {
-				math_error("Illegal truth value for dupvar_warn");
+				math_error("Illegal truth value for "
+					   "dupvar_warn");
 				/*NOTREACHED*/
 			}
 			conf->dupvar_warn = (int)temp;
@@ -1153,7 +1155,8 @@ config_value(CONFIG *cfg, int type, VALUE *vp)
 		vp->v_type = V_STR;
 		p = lookup_name(modes, cfg->outmode2);
 		if (p == NULL) {
-			math_error("invalid secondary output mode: %d", cfg->outmode2);
+			math_error("invalid secondary output mode: %d",
+				   cfg->outmode2);
 			/*NOTREACHED*/
 		}
 		vp->v_str = makenewstring(p);

config.h

@@ -1,7 +1,7 @@
 /*
  * config - configuration routines
  *
- * Copyright (C) 1999-2007  Landon Curt Noll and David I. Bell
+ * Copyright (C) 1999-2007,2014  Landon Curt Noll and David I. Bell
  *
  * Primary author:  Landon Curt Noll
  *
@@ -19,8 +19,8 @@
  * received a copy with calc; if not, write to Free Software Foundation, Inc.
  * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
  *
- * @(#) $Revision: 30.2 $
- * @(#) $Id: config.h,v 30.2 2007/09/21 01:27:27 chongo Exp $
+ * @(#) $Revision: 30.4 $
+ * @(#) $Id: config.h,v 30.4 2014/08/24 21:56:51 chongo Exp $
  * @(#) $Source: /usr/local/src/bin/calc/RCS/config.h,v $
  *
  * Under source code control:	1995/11/01 22:20:17
@@ -31,8 +31,8 @@
  */
 
 
-#if !defined(__CONFIG_H__)
-#define __CONFIG_H__
+#if !defined(INCLUDE_CONFIG_H)
+#define INCLUDE_CONFIG_H
 
 
 #if defined(CALC_SRC)	/* if we are building from the calc source tree */
@@ -226,4 +226,4 @@ E_FUNC void config_print(CONFIG*);
 E_FUNC BOOL config_cmp(CONFIG*, CONFIG*);
 
 
-#endif /* !__CONFIG_H__ */
+#endif /* !INCLUDE_CONFIG_H */

cscript/4dsphere.calc

@@ -1,4 +1,4 @@
-#!/usr/local/src/cmd/calc/calc -q -s -f
+#!/usr/local/src/bin/calc/calc -q -s -f
 /*
  * 4dsphere - determine if 6 points lie on the surface of a sphere in R^4
  *
@@ -10,7 +10,7 @@
  *	...		...
  *      x5 y5 z5 w5	point 5 in R^4
  *
- * Copyright (C) 2001  Landon Curt Noll
+ * Copyright (C) 2001,2014  Landon Curt Noll
  *
  * Calc is open software; you can redistribute it and/or modify it under
  * the terms of the version 2.1 of the GNU Lesser General Public License
@@ -26,8 +26,8 @@
  * received a copy with calc; if not, write to Free Software Foundation, Inc.
  * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
  *
- * @(#) $Revision: 30.1 $
- * @(#) $Id: 4dsphere.calc,v 30.1 2007/03/16 11:12:11 chongo Exp $
+ * @(#) $Revision: 30.3 $
+ * @(#) $Id: 4dsphere.calc,v 30.3 2014/08/24 22:04:32 chongo Exp $
  * @(#) $Source: /usr/local/src/bin/calc/cscript/RCS/4dsphere.calc,v $
  *
  * Under source code control:   2001/05/03 19:02:03

cscript/Makefile

@@ -1,7 +1,7 @@
 #
 # cscript - makefile for calc shell script files
 #
-# Copyright (C) 1999-2006  Landon Curt Noll
+# Copyright (C) 1999-2006,2014,2017  Landon Curt Noll
 #
 # Calc is open software; you can redistribute it and/or modify it under
 # the terms of the version 2.1 of the GNU Lesser General Public License
@@ -17,9 +17,9 @@
 # received a copy with calc; if not, write to Free Software Foundation, Inc.
 # 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
 #
-# @(#) $Revision: 30.3 $
-# @(#) $Id: Makefile,v 30.3 2007/09/21 01:27:27 chongo Exp $
-# @(#) $Source: /usr/local/src/bin/calc/cscript/RCS/Makefile,v $
+# @(#) $Revision: 30.10 $
+# @(#) $Id: Makefile,v 30.10 2017/05/19 16:09:14 chongo Exp $
+# @(#) $Source: /usr/local/src/bin/calc-RHEL7/cscript/RCS/Makefile,v $
 #
 # Under source code control:	1999/11/29 11:10:26
 # File existed as early as:	1999
@@ -227,7 +227,7 @@ all: ${TARGETS} .all
 
 README: README.src
 	@${RM} -f $@
-	@${SED} -e "s:#!/usr/local/src/cmd/calc/calc:#!${BINDIR}/calc:" $?>$@
+	@${SED} -e "s:#!/usr/local/src/bin/calc/calc:#!${BINDIR}/calc:" $?>$@
 
 # used by the upper level Makefile to determine if we have done all
 #
@@ -315,7 +315,8 @@ depend:
 	    if [ X"$$i" != X"/dev/null" ]; then \
 		echo "$$i: $$i.calc"; \
 		echo '	@$${RM} -f $$@'; \
-		echo '	@$${SED} -e "1s:^#!/usr/local/src/cmd/calc/calc:#!$${BINDIR}/calc:" $$?>$$@'; \
+		echo -n '	@$${SED} -e "1s:^#!/usr/local/src/bin/'; \
+		echo 'calc/calc:#!$${BINDIR}/calc:" $$?>$$@'; \
 		echo '	@$${CHMOD} +x $$@'; \
 	    fi; \
 	done >> makedep.out
@@ -323,7 +324,8 @@ depend:
 	${Q} echo forming new cscript/${MAKE_FILE}
 	${Q} ${RM} -f ${MAKE_FILE}.bak
 	${Q} ${MV} ${MAKE_FILE} ${MAKE_FILE}.bak
-	${Q} ${SED} -n '1,/^# DO NOT DELETE THIS LINE/p' ${MAKE_FILE}.bak > ${MAKE_FILE}
+	${Q} ${SED} -n '1,/^# DO NOT DELETE THIS LINE/p' \
+		    ${MAKE_FILE}.bak > ${MAKE_FILE}
 	${Q} echo "" >> ${MAKE_FILE}
 	${Q} ${CAT} makedep.out >> ${MAKE_FILE}
 	${Q} ${RM} -f makedep.out
@@ -338,9 +340,9 @@ depend:
 	    fi; \
 	    ${MV} ${MAKE_FILE}.tmp ${MAKE_FILE}; \
 	    if [ -d RCS ]; then \
-		echo '*****************************************************'; \
-		echo 'new sample ${MAKE_FILE} formed -- need to check it in'; \
-		echo '*****************************************************'; \
+		echo '******************************************************'; \
+		echo 'new cscript ${MAKE_FILE} formed -- need to check it in'; \
+		echo '******************************************************'; \
 	    fi; \
 	fi
 
@@ -369,6 +371,10 @@ clean:
 clobber: clean
 	${RM} -f .all
 	${RM} -f ${TARGETS}
+	-${Q} if [ -e .DS_Store ]; then \
+	    echo ${RM} -rf .DS_Store; \
+	    ${RM} -rf .DS_Store; \
+	fi
 
 # install everything
 #
@@ -376,8 +382,8 @@ clobber: clean
 #
 install: all
 	-${Q} if [ ! -d ${T}${BINDIR} ]; then \
-	    echo ${MKDIR} ${T}${BINDIR}; \
-	    ${MKDIR} ${T}${BINDIR}; \
+	    echo ${MKDIR} -p ${T}${BINDIR}; \
+	    ${MKDIR} -p ${T}${BINDIR}; \
 	    if [ ! -d "${T}${BINDIR}" ]; then \
 		echo ${MKDIR} -p "${T}${BINDIR}"; \
 		${MKDIR} -p "${T}${BINDIR}"; \
@@ -388,8 +394,8 @@ install: all
 	    ${TRUE}; \
 	fi
 	-${Q} if [ ! -d ${T}${SCRIPTDIR} ]; then \
-	    echo ${MKDIR} ${T}${SCRIPTDIR}; \
-	    ${MKDIR} ${T}${SCRIPTDIR}; \
+	    echo ${MKDIR} -p ${T}${SCRIPTDIR}; \
+	    ${MKDIR} -p ${T}${SCRIPTDIR}; \
 	    if [ ! -d "${T}${SCRIPTDIR}" ]; then \
 		echo ${MKDIR} -p "${T}${SCRIPTDIR}"; \
 		${MKDIR} -p "${T}${SCRIPTDIR}"; \
@@ -443,33 +449,33 @@ uninstall:
 
 4dsphere: 4dsphere.calc
 	@${RM} -f $@
-	@${SED} -e "1s:^#!/usr/local/src/cmd/calc/calc:#!${BINDIR}/calc:" $?>$@
+	@${SED} -e "1s:^#!/usr/local/src/bin/calc/calc:#!${BINDIR}/calc:" $?>$@
 	@${CHMOD} +x $@
 fproduct: fproduct.calc
 	@${RM} -f $@
-	@${SED} -e "1s:^#!/usr/local/src/cmd/calc/calc:#!${BINDIR}/calc:" $?>$@
+	@${SED} -e "1s:^#!/usr/local/src/bin/calc/calc:#!${BINDIR}/calc:" $?>$@
 	@${CHMOD} +x $@
 mersenne: mersenne.calc
 	@${RM} -f $@
-	@${SED} -e "1s:^#!/usr/local/src/cmd/calc/calc:#!${BINDIR}/calc:" $?>$@
+	@${SED} -e "1s:^#!/usr/local/src/bin/calc/calc:#!${BINDIR}/calc:" $?>$@
 	@${CHMOD} +x $@
 piforever: piforever.calc
 	@${RM} -f $@
-	@${SED} -e "1s:^#!/usr/local/src/cmd/calc/calc:#!${BINDIR}/calc:" $?>$@
+	@${SED} -e "1s:^#!/usr/local/src/bin/calc/calc:#!${BINDIR}/calc:" $?>$@
 	@${CHMOD} +x $@
 plus: plus.calc
 	@${RM} -f $@
-	@${SED} -e "1s:^#!/usr/local/src/cmd/calc/calc:#!${BINDIR}/calc:" $?>$@
+	@${SED} -e "1s:^#!/usr/local/src/bin/calc/calc:#!${BINDIR}/calc:" $?>$@
 	@${CHMOD} +x $@
 powerterm: powerterm.calc
 	@${RM} -f $@
-	@${SED} -e "1s:^#!/usr/local/src/cmd/calc/calc:#!${BINDIR}/calc:" $?>$@
+	@${SED} -e "1s:^#!/usr/local/src/bin/calc/calc:#!${BINDIR}/calc:" $?>$@
 	@${CHMOD} +x $@
 simple: simple.calc
 	@${RM} -f $@
-	@${SED} -e "1s:^#!/usr/local/src/cmd/calc/calc:#!${BINDIR}/calc:" $?>$@
+	@${SED} -e "1s:^#!/usr/local/src/bin/calc/calc:#!${BINDIR}/calc:" $?>$@
 	@${CHMOD} +x $@
 square: square.calc
 	@${RM} -f $@
-	@${SED} -e "1s:^#!/usr/local/src/cmd/calc/calc:#!${BINDIR}/calc:" $?>$@
+	@${SED} -e "1s:^#!/usr/local/src/bin/calc/calc:#!${BINDIR}/calc:" $?>$@
 	@${CHMOD} +x $@

cscript/README.src

@@ -7,7 +7,7 @@ be useful!
 
 Please note that calc shell scripts must start with the line:
 
-    #!/usr/local/src/cmd/calc/calc -q -f
+    #!/usr/local/src/bin/calc/calc -q -f
 
 The above line MUST start in column 1 of the first line.   The first line
 must also end in -f.  The -q is optional, but is recommended to disable
@@ -18,7 +18,7 @@ Comments must be /* c-like comment */ or start with a double ## symbol.
 
 This is the correct way to form a calc shell script:
 
-    #!/usr/local/src/cmd/calc/calc -q -f
+    #!/usr/local/src/bin/calc/calc -q -f
 
     /* a correct comment */
     ## another correct comment
@@ -36,27 +36,51 @@ For more informaton about calc command lines, see "help usage".
 
 This next example WRONG:
 
-    #!/usr/local/src/cmd/calc/calc -q
+    #!/usr/local/src/bin/calc/calc -q
 
     # This is not a calc calc comment because it has only a single #
     # You must to start comments with ## or /*
     # is is also wrong because the first line does not end in -f
     print "This example has invalid comments"
 
-#####
-
-If you write something that you think is useful, please send it to:
-
-    calc-contrib at asthe dot com
-
-    [[ NOTE: Replace 'at' with @, 'dot' is with . and remove the spaces ]]
-    [[ NOTE: The EMail address uses 'asthe' and the web site URL uses 'isthe' ]]
+=-=
 
 For more info, see:
 
     help script
     help cscript
 
+#####
+
+If you write something that you think is useful, please join the
+low volume calc mailing list calc-tester.  Then send your contribution
+to the calc-tester mailing list.
+
+To subscribe to the calc-tester mailing list, visit the following URL:
+
+	http://www.isthe.com/chongo/tech/comp/calc/calc-tester.html
+
+    This is a low volume moderated mailing list.
+
+    This mailing list replaces calc-tester at asthe dot com list.
+
+    If you need a human to help you with your mailing list subscription,
+    please send EMail to our special:
+
+	calc-tester-maillist-help at asthe dot com
+
+	NOTE: Remove spaces and replace 'at' with @, 'dot' with .
+
+    address.  To be sure we see your EMail asking for help with your
+    mailing list subscription, please use the following phase in your
+    EMail Subject line:
+
+	calc tester mailing list help
+
+    That phrase in your subject line will help ensure your
+    request will get past our anti-spam filters.  You may have
+    additional words in your subject line.
+
 =-=
 
 4dsphere
@@ -101,7 +125,7 @@ simple
 
     A trivial example of a calc shell script.
 
-## Copyright (C) 1999  Landon Curt Noll
+## Copyright (C) 1999,2014  Landon Curt Noll
 ##
 ## Calc is open software; you can redistribute it and/or modify it under
 ## the terms of the version 2.1 of the GNU Lesser General Public License
@@ -117,8 +141,8 @@ simple
 ## received a copy with calc; if not, write to Free Software Foundation, Inc.
 ## 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
 ##
-## @(#) $Revision: 30.1 $
-## @(#) $Id: README.src,v 30.1 2007/03/16 11:12:11 chongo Exp $
+## @(#) $Revision: 30.5 $
+## @(#) $Id: README.src,v 30.5 2014/10/12 12:23:43 chongo Exp $
 ## @(#) $Source: /usr/local/src/bin/calc/cscript/RCS/README.src,v $
 ##
 ## Under source code control:	1999/12/17 10:23:40

cscript/fproduct.calc

@@ -1,4 +1,4 @@
-#!/usr/local/src/cmd/calc/calc -q -s -f
+#!/usr/local/src/bin/calc/calc -q -s -f
 /*
  * fproduct - write the big Endian product of terms to a file
  *
@@ -8,7 +8,7 @@
  *	filename	where to write the product, use - for stdout
  *	term ...	terms to multiply
  *
- * Copyright (C) 2001  Landon Curt Noll
+ * Copyright (C) 2001,2014  Landon Curt Noll
  *
  * Calc is open software; you can redistribute it and/or modify it under
  * the terms of the version 2.1 of the GNU Lesser General Public License
@@ -24,8 +24,8 @@
  * received a copy with calc; if not, write to Free Software Foundation, Inc.
  * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
  *
- * @(#) $Revision: 30.1 $
- * @(#) $Id: fproduct.calc,v 30.1 2007/03/16 11:12:11 chongo Exp $
+ * @(#) $Revision: 30.3 $
+ * @(#) $Id: fproduct.calc,v 30.3 2014/08/24 22:04:32 chongo Exp $
  * @(#) $Source: /usr/local/src/bin/calc/cscript/RCS/fproduct.calc,v $
  *
  * Under source code control:   2001/04/07 20:13:11

cscript/mersenne.calc

@@ -1,8 +1,8 @@
-#!/usr/local/src/cmd/calc/calc -q -s -f
+#!/usr/local/src/bin/calc/calc -q -s -f
 /*
  * mersenne - print the value of a mersenne number
  *
- * Copyright (C) 1999-2007  Landon Curt Noll
+ * Copyright (C) 1999-2007,2014  Landon Curt Noll
  *
  * Calc is open software; you can redistribute it and/or modify it under
  * the terms of the version 2.1 of the GNU Lesser General Public License
@@ -18,8 +18,8 @@
  * received a copy with calc; if not, write to Free Software Foundation, Inc.
  * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
  *
- * @(#) $Revision: 30.1 $
- * @(#) $Id: mersenne.calc,v 30.1 2007/03/16 11:12:11 chongo Exp $
+ * @(#) $Revision: 30.3 $
+ * @(#) $Id: mersenne.calc,v 30.3 2014/08/24 22:04:32 chongo Exp $
  * @(#) $Source: /usr/local/src/bin/calc/cscript/RCS/mersenne.calc,v $
  *
  * Under source code control:	1999/11/30 00:09:01;

cscript/piforever.calc

@@ -1,8 +1,8 @@
-#!/usr/local/src/cmd/calc/calc -q -f
+#!/usr/local/src/bin/calc/calc -q -f
 /*
  * piforever - print digits of pi forever (or as long as your mem/cpu allow)
  *
- * Copyright (C) 1999-2007  Landon Curt Noll
+ * Copyright (C) 1999-2007,2014  Landon Curt Noll
  *
  * Calc is open software; you can redistribute it and/or modify it under
  * the terms of the version 2.1 of the GNU Lesser General Public License
@@ -18,8 +18,8 @@
  * received a copy with calc; if not, write to Free Software Foundation, Inc.
  * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
  *
- * @(#) $Revision: 30.1 $
- * @(#) $Id: piforever.calc,v 30.1 2007/03/16 11:12:11 chongo Exp $
+ * @(#) $Revision: 30.3 $
+ * @(#) $Id: piforever.calc,v 30.3 2014/08/24 22:04:32 chongo Exp $
  * @(#) $Source: /usr/local/src/bin/calc/cscript/RCS/piforever.calc,v $
  *
  * Under source code control:	1999/11/30 00:11:36

cscript/plus.calc

@@ -1,8 +1,8 @@
-#!/usr/local/src/cmd/calc/calc -q -f
+#!/usr/local/src/bin/calc/calc -q -f
 /*
  * plus - add two or more arguments together
  *
- * Copyright (C) 1999-2007  Landon Curt Noll
+ * Copyright (C) 1999-2007,2014  Landon Curt Noll
  *
  * Calc is open software; you can redistribute it and/or modify it under
  * the terms of the version 2.1 of the GNU Lesser General Public License
@@ -18,8 +18,8 @@
  * received a copy with calc; if not, write to Free Software Foundation, Inc.
  * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
  *
- * @(#) $Revision: 30.1 $
- * @(#) $Id: plus.calc,v 30.1 2007/03/16 11:12:11 chongo Exp $
+ * @(#) $Revision: 30.3 $
+ * @(#) $Id: plus.calc,v 30.3 2014/08/24 22:04:32 chongo Exp $
  * @(#) $Source: /usr/local/src/bin/calc/cscript/RCS/plus.calc,v $
  *
  * Under source code control:	1999/11/29 10:22:37

cscript/powerterm.calc

@@ -1,4 +1,4 @@
-#!/usr/local/src/cmd/calc/calc -q -s -f
+#!/usr/local/src/bin/calc/calc -q -s -f
 /*
  * powerterm - print the argument as a sum of powers of integers
  *
@@ -8,7 +8,7 @@
  *	base_limit	largest base we will consider (def: 10000)
  *      value		value to convert into sums of powers of integers
  *
- * Copyright (C) 2001  Landon Curt Noll
+ * Copyright (C) 2001,2014  Landon Curt Noll
  *
  * Calc is open software; you can redistribute it and/or modify it under
  * the powerterm of the version 2.1 of the GNU Lesser General Public License
@@ -24,8 +24,8 @@
  * received a copy with calc; if not, write to Free Software Foundation, Inc.
  * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
  *
- * @(#) $Revision: 30.1 $
- * @(#) $Id: powerterm.calc,v 30.1 2007/03/16 11:12:11 chongo Exp $
+ * @(#) $Revision: 30.3 $
+ * @(#) $Id: powerterm.calc,v 30.3 2014/08/24 22:04:32 chongo Exp $
  * @(#) $Source: /usr/local/src/bin/calc/cscript/RCS/powerterm.calc,v $
  *
  * Under source code control:   2001/04/24 23:49:11

cscript/simple.calc

@@ -1,8 +1,8 @@
-#!/usr/local/src/cmd/calc/calc -q -f
+#!/usr/local/src/bin/calc/calc -q -f
 /*
  * simple - an example of a simple calc shell script
  *
- * Copyright (C) 1999-2007  Landon Curt Noll
+ * Copyright (C) 1999-2007,2014  Landon Curt Noll
  *
  * Calc is open software; you can redistribute it and/or modify it under
  * the terms of the version 2.1 of the GNU Lesser General Public License
@@ -18,8 +18,8 @@
  * received a copy with calc; if not, write to Free Software Foundation, Inc.
  * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
  *
- * @(#) $Revision: 30.1 $
- * @(#) $Id: simple.calc,v 30.1 2007/03/16 11:12:11 chongo Exp $
+ * @(#) $Revision: 30.3 $
+ * @(#) $Id: simple.calc,v 30.3 2014/08/24 22:04:32 chongo Exp $
  * @(#) $Source: /usr/local/src/bin/calc/cscript/RCS/simple.calc,v $
  *
  * Under source code control:	1999/11/29 10:22:37

cscript/square.calc

@@ -1,8 +1,8 @@
-#!/usr/local/src/cmd/calc/calc -q -f
+#!/usr/local/src/bin/calc/calc -q -f
 /*
  * sqaure - print the squares of input values
  *
- * Copyright (C) 2000-2007  Ernest Bowen
+ * Copyright (C) 2000-2007,2014  Ernest Bowen
  *
  * Calc is open software; you can redistribute it and/or modify it under
  * the terms of the version 2.1 of the GNU Lesser General Public License
@@ -18,8 +18,8 @@
  * received a copy with calc; if not, write to Free Software Foundation, Inc.
  * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
  *
- * @(#) $Revision: 30.1 $
- * @(#) $Id: square.calc,v 30.1 2007/03/16 11:12:11 chongo Exp $
+ * @(#) $Revision: 30.3 $
+ * @(#) $Id: square.calc,v 30.3 2014/08/24 22:04:32 chongo Exp $
  * @(#) $Source: /usr/local/src/bin/calc/cscript/RCS/square.calc,v $
  *
  * Under source code control:	2000/12/15 06:52:01

custom/Makefile

@@ -2,7 +2,7 @@
 #
 # custom - makefile for calc custom routines
 #
-# Copyright (C) 1999-2006  Landon Curt Noll
+# Copyright (C) 1999-2006,2014,2017  Landon Curt Noll
 #
 # Calc is open software; you can redistribute it and/or modify it under
 # the terms of the version 2.1 of the GNU Lesser General Public License
@@ -18,9 +18,9 @@
 # received a copy with calc; if not, write to Free Software Foundation, Inc.
 # 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
 #
-# @(#) $Revision: 30.25 $
-# @(#) $Id: Makefile.head,v 30.25 2013/05/05 10:21:52 chongo Exp $
-# @(#) $Source: /usr/local/src/bin/calc/custom/RCS/Makefile.head,v $
+# @(#) $Revision: 30.42 $
+# @(#) $Id: Makefile.head,v 30.42 2017/05/19 16:09:14 chongo Exp $
+# @(#) $Source: /usr/local/src/bin/calc-RHEL7/custom/RCS/Makefile.head,v $
 #
 # Under source code control:	1997/03/09 02:28:54
 # File existed as early as:	1997
@@ -337,20 +337,6 @@ CALC_ENV= CALCPATH=./cal
 ALLOW_CUSTOM= -DCUSTOM
 #ALLOW_CUSTOM=
 
-# The install rule uses:
-#
-#	${MKDIR} ${MKDIR_ARG}
-#
-# to create directorties.  Normall this amounts to usins mkdir -p dir ...
-# Some older systems may not have mkdir -p.  If you system does not
-# make mkdir -p, then set MKDIR_ARG to empty.
-#
-# MKDIR_ARG= -p			# use mkdir -p when creating paths
-# MKDIR_ARG=			# use if system does not understand mkdir -p
-#
-MKDIR_ARG= -p
-#MKDIR_ARG=
-
 # Some out of date operating systems require / want an executable to
 # end with a certain file extension.  Some compile systems such as
 # Cygwin build calc as calc.exe.  The EXT variable is used to denote
@@ -366,18 +352,12 @@ EXT=
 
 # The default calc versions
 #
-VERSION= 2.12.4.7
-VERS= 2.12.4
-VER= 2.12
-VE= 2
+VERSION= 2.12.5.6
 
 # Names of shared libraries with versions
 #
 LIB_EXT= .so
 LIB_EXT_VERSION= ${LIB_EXT}.${VERSION}
-LIB_EXT_VERS= ${LIB_EXT}.${VERS}
-LIB_EXT_VER= ${LIB_EXT}.${VER}
-LIB_EXT_VE= ${LIB_EXT}.${VE}
 
 # standard tools
 #
@@ -473,30 +453,41 @@ LDFLAGS= ${LD_DEBUG} ${ILDFLAGS}
 #	BLD_TYPE= calc-static-only
 #
 # CC_SHARE are flags given to ${CC} to build .o files suitable for shared libs
-# DEFAULT_LIB_INSTALL_PATH is where calc progs look for calc shared libs
+# DEFAULT_LIB_INSTALL_PATH is where calc programs look for calc shared libs
 # LD_SHARE are common flags given to ${CC} to link with shared libraries
-# LIBCALC_SHLIB are flags given to ${CC} to build libcalc shared lib
+# LIBCALC_SHLIB are flags given to ${CC} to build libcalc shared libraries
 # LIBCUSTCALC_SHLIB are flags given to ${CC} to build libcustcalc shared lib
 #
-#	NOTE: The above 4 values are unused if BLD_TYPE= calc-static-only
+#	NOTE: The above 5 values are unused if BLD_TYPE= calc-static-only
 #
 # CC_STATIC are flags given to ${CC} to build .o files suitable for static libs
 # LD_STATIC are common flags given to ${CC} to link with static libraries
-# LIBCALC_STATIC are flags given to ${CC} to build libcalc static lib
+# LIBCALC_STATIC are flags given to ${CC} to build libcalc static libraries
 # LIBCUSTCALC_STATIC are flags given to ${CC} to build libcustcalc static lib
 #
 #	NOTE: The above 4 values are unused if BLD_TYPE= calc-dynamic-only
 #
 # CCOPT are flags given to ${CC} for optimization
 # CCWARN are flags given to ${CC} for warning message control
+#
+# WNO_IMPLICT, WNO_ERROR_LONG_LONG and WNO_LONG_LONG are given to ${CC}
+#	when compiling special .o files that may need special compile options
+#	NOTE: These flags simply turn off certain compiler warnings,
+#	      which is useful only when CCWERR is set to -Werror.
+#	NOTE: If your compiler does not have these -Wno files, just
+#	      set these variables to nothing as in:
+#		WNO_IMPLICT=
+#		WNO_ERROR_LONG_LONG=
+#		WNO_LONG_LONG=
+#
 # CCWERR are flags given to ${CC} to make warnings fatal errors
 #	NOTE: CCWERR is only set in development Makefiles and must only be
 #	      used with ${CC}, not ${LCC}.  If you do not want the compiler
 #	      to abort on warnings, then leave CCWERR blank.
 # CCMISC are misc flags given to ${CC}
 #
-# LCC how the the C compiler is invoked on locally executed intermediate progs
-# CC is how the the C compiler is invoked (with an optional Purify)
+# LCC is how the C compiler is invoked on locally executed intermediate programs
+# CC is how the C compiler is invoked (with an optional Purify)
 #
 # Specific target overrides or modifications to default values
 
@@ -535,7 +526,10 @@ LD_STATIC=
 LIBCALC_STATIC=
 LIBCUSTCALC_STATIC=
 #
-CCWARN= -Wall -W -Wno-comment
+CCWARN= -Wall
+WNO_IMPLICT= -Wno-implicit
+WNO_ERROR_LONG_LONG= -Wno-error=long-long
+WNO_LONG_LONG= -Wno-long-long
 CCWERR=
 CCOPT= ${DEBUG}
 CCMISC=
@@ -545,9 +539,9 @@ CC= ${PURIFY} ${LCC} ${CCWERR}
 #
 endif
 
-#################################
-# Apple MacOS X / Darwin target #
-#################################
+##############################
+# Apple OS X / Darwin target #
+##############################
 
 ifeq ($(target),Darwin)
 #
@@ -556,9 +550,23 @@ BLD_TYPE= calc-dynamic-only
 CC_SHARE= -fPIC
 DEFAULT_LIB_INSTALL_PATH= ${PWD}:${LIBDIR}:/usr/local/lib
 LD_SHARE= ${DARWIN_ARCH}
-LIBCALC_SHLIB= -single_module -undefined dynamic_lookup -dynamiclib
+#SET_INSTALL_NAME= no
+SET_INSTALL_NAME= yes
+ifeq ($(SET_INSTALL_NAME),yes)
+LIBCALC_SHLIB= -single_module -undefined dynamic_lookup -dynamiclib \
+    -install_name ${LIBDIR}/libcalc${LIB_EXT_VERSION} ${DARWIN_ARCH}
+else
+LIBCALC_SHLIB= -single_module -undefined dynamic_lookup -dynamiclib \
+    ${DARWIN_ARCH}
+endif
 ifdef ALLOW_CUSTOM
-LIBCUSTCALC_SHLIB= -single_module -undefined dynamic_lookup -dynamiclib
+ifeq ($(SET_INSTALL_NAME),yes)
+LIBCUSTCALC_SHLIB= -single_module -undefined dynamic_lookup -dynamiclib \
+    -install_name ${LIBDIR}/libcustcalc${LIB_EXT_VERSION} ${DARWIN_ARCH}
+else
+LIBCUSTCALC_SHLIB= -single_module -undefined dynamic_lookup -dynamiclib \
+    ${DARWIN_ARCH}
+endif
 else
 LIBCUSTCALC_SHLIB=
 endif
@@ -568,28 +576,37 @@ LD_STATIC= ${DARWIN_ARCH}
 LIBCALC_STATIC=
 LIBCUSTCALC_STATIC=
 #
-CCWARN= -Wall -W -Wno-comment
+CCWARN= -Wall
+WNO_IMPLICT= -Wno-implicit
+WNO_ERROR_LONG_LONG= -Wno-error=long-long
+WNO_LONG_LONG= -Wno-long-long
 CCWERR=
 CCOPT= ${DEBUG}
 CCMISC= ${DARWIN_ARCH}
 #
-LCC= MACOSX_DEPLOYMENT_TARGET=${MACOSX_DEPLOYMENT_TARGET} gcc
+LCC= clang
 CC= ${PURIFY} ${LCC} ${CCWERR}
 #
-# Darmin dynamic shared lib filenames
+# Darwin dynamic shared lib filenames
 LIB_EXT:= .dylib
 LIB_EXT_VERSION:= .${VERSION}${LIB_EXT}
-LIB_EXT_VERS:= .${VERS}${LIB_EXT}
-LIB_EXT_VER:= .${VER}${LIB_EXT}
-LIB_EXT_VE:= .${VE}${LIB_EXT}
 # LDCONFIG not required on this platform, so we redefine it to an empty string
 LDCONFIG:=
 # DARWIN_ARCH= -arch i386 -arch ppc	# Universal binary
 # DARWIN_ARCH= -arch i386		# Intel binary
 # DARWIN_ARCH= -arch ppc		# PPC binary
+# DARWIN_ARCH= -arch x86_64		# native 64-bit binary
 DARWIN_ARCH=				# native binary
-MACOSX_DEPLOYMENT_TARGET=10.8
 #
+# Starting with El Capitan OS X 10.11, root by default could not
+# mkdir under system locations, so we now use the /usr/local tree.
+#
+OPTDIR:= /usr/local
+BINDIR:= /${OPTDIR}/bin
+LIBDIR:= /${OPTDIR}/lib
+CALC_SHAREDIR:= /${OPTDIR}/share
+CALC_INCDIR:= /${OPTDIR}/include
+SCRIPTDIR:= ${BINDIR}/cscript
 endif
 
 ##################
@@ -622,7 +639,10 @@ LD_STATIC=
 LIBCALC_STATIC=
 LIBCUSTCALC_STATIC=
 #
-CCWARN= -Wall -W -Wno-comment
+CCWARN= -Wall
+WNO_IMPLICT= -Wno-implicit
+WNO_ERROR_LONG_LONG= -Wno-error=long-long
+WNO_LONG_LONG= -Wno-long-long
 CCWERR=
 CCOPT= ${DEBUG}
 CCMISC=
@@ -665,6 +685,9 @@ LIBCALC_STATIC=
 LIBCUSTCALC_STATIC=
 #
 CCWARN= -Wall
+WNO_IMPLICT= -Wno-implicit
+WNO_ERROR_LONG_LONG= -Wno-error=long-long
+WNO_LONG_LONG= -Wno-long-long
 CCWERR=
 CCOPT= ${DEBUG}
 CCMISC=
@@ -680,7 +703,6 @@ endif
 # SunOS target #
 ################
 
-# XXX - this needs to be tested
 ifeq ($(target),SunOS)
 #
 BLD_TYPE= calc-dynamic-only
@@ -701,7 +723,10 @@ LIBCALC_STATIC=
 LIBCUSTCALC_STATIC=
 LD_STATIC=
 #
-CCWARN= -Wall -W -Wno-comment
+CCWARN= -Wall
+WNO_IMPLICT= -Wno-implicit
+WNO_ERROR_LONG_LONG= -Wno-error=long-long
+WNO_LONG_LONG= -Wno-long-long
 CCWERR=
 CCOPT= ${DEBUG}
 CCMISC=
@@ -735,7 +760,10 @@ LIBCALC_STATIC=
 LIBCUSTCALC_STATIC=
 LD_STATIC=
 #
-CCWARN= -Wall -W -Wno-comment
+CCWARN= -Wall
+WNO_IMPLICT= -Wno-implicit
+WNO_ERROR_LONG_LONG= -Wno-error=long-long
+WNO_LONG_LONG= -Wno-long-long
 CCWERR=
 CCOPT= ${DEBUG}
 CCMISC=
@@ -769,7 +797,10 @@ LD_STATIC=
 LIBCALC_STATIC=
 LIBCUSTCALC_STATIC=
 #
-CCWARN= -Wall -W -Wno-comment
+CCWARN= -Wall
+WNO_IMPLICT= -Wno-implicit
+WNO_ERROR_LONG_LONG= -Wno-error=long-long
+WNO_LONG_LONG= -Wno-long-long
 CCWERR=
 CCOPT= ${DEBUG}
 CCMISC=
@@ -779,10 +810,10 @@ CC= ${PURIFY} ${LCC} ${CCWERR}
 #
 # The simple makefile forces the use of static ${CC} flags
 #
-# ICFLAGS are given to ${CC} for intermediate progs used to help compile calc
-# CFLAGS are given to ${CC} for calc progs other than intermediate progs
-# ILDFLAGS for ${CC} in linking intermediate progs used to help compile calc
-# LDFLAGS for ${CC} in linking calc progs other than intermediate progs
+# ICFLAGS are given to ${CC} for intermediate programs used to help compile calc
+# CFLAGS are given to ${CC} for calc programs other than intermediate programs
+# ILDFLAGS for ${CC} in linking intermediate programs used to help compile calc
+# LDFLAGS for ${CC} in linking calc programs other than intermediate programs
 #
 ICFLAGS= ${COMMON_CFLAGS} ${CC_STATIC}
 CFLAGS= ${ICFLAGS} ${CCOPT}
@@ -804,10 +835,10 @@ EXT=.exe
 TERMCONTROL= -DUSE_WIN32
 ifdef ALLOW_CUSTOM
 #endif	/* end of skip for non-Gnu makefiles */
-CALCPATH= .;./cal;~/.cal;${CALC_SHAREDIR};${CUSTOMCALDIR}
+CALCPATH= .;./cal;~/.cal;${T}${CALC_SHAREDIR};${T}${CUSTOMCALDIR}
 #if 0	/* start of skip for non-Gnu makefiles */
 else
-CALCPATH= .;./cal;~/.cal;${CALC_SHAREDIR}
+CALCPATH= .;./cal;~/.cal;${T}${CALC_SHAREDIR}
 endif
 CALCRC= ${CALC_SHAREDIR}/startup;~/.calcrc;./.calcinit
 #
@@ -829,7 +860,10 @@ LIBCALC_STATIC=
 LIBCUSTCALC_STATIC=
 LD_STATIC=
 #
-CCWARN= -Wall -W -Wno-comment
+CCWARN= -Wall
+WNO_IMPLICT= -Wno-implicit
+WNO_ERROR_LONG_LONG= -Wno-error=long-long
+WNO_LONG_LONG= -Wno-long-long
 CCWERR=
 CCOPT= ${DEBUG}
 CCMISC= -DNOTCYGWIN
@@ -866,7 +900,10 @@ LIBCALC_STATIC=
 LIBCUSTCALC_STATIC=
 LD_STATIC=
 #
-CCWARN= -Wall -W -Wno-comment
+CCWARN= -Wall
+WNO_IMPLICT= -Wno-implicit
+WNO_ERROR_LONG_LONG= -Wno-error=long-long
+WNO_LONG_LONG= -Wno-long-long
 CCWERR=
 CCOPT= ${DEBUG}
 CCMISC=
@@ -882,8 +919,8 @@ endif
 
 # Required flags to compile C files for calc
 #
-# ICFLAGS are given to ${CC} for intermediate progs used to help compile calc
-# CFLAGS are given to ${CC} for calc progs other than intermediate progs
+# ICFLAGS are given to ${CC} for intermediate programs used to help compile calc
+# CFLAGS are given to ${CC} for calc programs other than intermediate programs
 #
 # NOTE: This does not work for: make-XYZ-only and BLD_TYPE != make-XYZ-only
 #
@@ -896,8 +933,8 @@ CFLAGS= ${ICFLAGS} ${CCOPT}
 
 # Required flags to link files for calc
 #
-# ILDFLAGS for ${CC} in linking intermediate progs used to help compile calc
-# LDFLAGS for ${CC} in linking calc progs other than intermediate progs
+# ILDFLAGS for ${CC} in linking intermediate programs used to help compile calc
+# LDFLAGS for ${CC} in linking calc programs other than intermediate programs
 #
 ILDFLAGS= ${COMMON_LDFLAGS}
 LDFLAGS= ${LD_DEBUG} ${ILDFLAGS}
@@ -1086,8 +1123,8 @@ depend:
 	fi
 	${Q} echo forming custom/skel
 	${Q} ${RM} -rf skel
-	${Q} ${MKDIR} skel
-	${Q} ${MKDIR} skel/custom
+	${Q} ${MKDIR} -p skel
+	${Q} ${MKDIR} -p skel/custom
 	-${Q} for i in ${C_SRC} /dev/null; do \
 	    if [ X"$$i" != X"/dev/null" ]; then \
 		${SED} -n '/^#[	 ]*include[	 ]*"/p' \
@@ -1136,7 +1173,8 @@ depend:
 	${Q} echo forming new custom/${MAKE_FILE}
 	${Q} ${RM} -f ${MAKE_FILE}.bak
 	${Q} ${MV} ${MAKE_FILE} ${MAKE_FILE}.bak
-	${Q} ${SED} -n '1,/^# DO NOT DELETE THIS LINE/p' ${MAKE_FILE}.bak > ${MAKE_FILE}
+	${Q} ${SED} -n '1,/^# DO NOT DELETE THIS LINE/p' \
+		    ${MAKE_FILE}.bak > ${MAKE_FILE}
 	${Q} ${GREP} -v '^#' skel/custom/makedep.out >> ${MAKE_FILE}
 	${Q} ${RM} -rf skel
 	-${Q} if ${CMP} -s ${MAKE_FILE}.bak ${MAKE_FILE}; then \
@@ -1197,23 +1235,38 @@ clobber: clean
 	${RM} -f ${BLD_TYPE}
 	${RM} -f .all ${MAKE_FILE}.tmp ${MAKE_FILE}.bak
 	${RM} -f libcustcalc${LIB_EXT_VERSION}
-	${RM} -f libcustcalc${LIB_EXT_VERS}
-	${RM} -f libcustcalc${LIB_EXT_VER}
-	${RM} -f libcustcalc${LIB_EXT_VE}
 	${RM} -f libcustcalc${LIB_EXT}
 	${RM} -f libcustcalc.a
+	${RM} -f libcustcalc*
+	${V} echo remove files that are obsolete
 #if 0	/* start of skip for non-Gnu makefiles */
 	${RM} -f Makefile.simple
 #endif	/* end of skip for non-Gnu makefiles */
+	-${Q} if [ -e .DS_Store ]; then \
+	    echo ${RM} -rf .DS_Store; \
+	    ${RM} -rf .DS_Store; \
+	fi
 
 # install everything
 #
 # NOTE: Keep the uninstall rule in reverse order to the install rule
 #
 install: all
+	-${Q} if [ ! -d ${T}${LIBDIR} ]; then \
+	    echo ${MKDIR} -p ${T}${LIBDIR}; \
+	    ${MKDIR} -p ${T}${LIBDIR}; \
+	    if [ ! -d "${T}${LIBDIR}" ]; then \
+		echo ${MKDIR} -p "${T}${LIBDIR}"; \
+		${MKDIR} -p "${T}${LIBDIR}"; \
+	    fi; \
+	    echo ${CHMOD} 0755 ${T}${LIBDIR}; \
+	    ${CHMOD} 0755 ${T}${LIBDIR}; \
+	else \
+	    ${TRUE}; \
+	fi
 	-${Q} if [ ! -d ${T}${INCDIR} ]; then \
-	    echo ${MKDIR} ${T}${INCDIR}; \
-	    ${MKDIR} ${T}${INCDIR}; \
+	    echo ${MKDIR} -p ${T}${INCDIR}; \
+	    ${MKDIR} -p ${T}${INCDIR}; \
 	    if [ ! -d "${T}${INCDIR}" ]; then \
 		echo ${MKDIR} -p "${T}${INCDIR}"; \
 		${MKDIR} -p "${T}${INCDIR}"; \
@@ -1224,8 +1277,8 @@ install: all
 	    ${TRUE}; \
 	fi
 	-${Q} if [ ! -d ${T}${CALC_SHAREDIR} ]; then \
-	    echo ${MKDIR} ${T}${CALC_SHAREDIR}; \
-	    ${MKDIR} ${T}${CALC_SHAREDIR}; \
+	    echo ${MKDIR} -p ${T}${CALC_SHAREDIR}; \
+	    ${MKDIR} -p ${T}${CALC_SHAREDIR}; \
 	    if [ ! -d "${T}${CALC_SHAREDIR}" ]; then \
 		echo ${MKDIR} -p "${T}${CALC_SHAREDIR}"; \
 		${MKDIR} -p "${T}${CALC_SHAREDIR}"; \
@@ -1236,8 +1289,8 @@ install: all
 	    ${TRUE}; \
 	fi
 	-${Q} if [ ! -d ${T}${CALC_INCDIR} ]; then \
-	    echo ${MKDIR} ${T}${CALC_INCDIR}; \
-	    ${MKDIR} ${T}${CALC_INCDIR}; \
+	    echo ${MKDIR} -p ${T}${CALC_INCDIR}; \
+	    ${MKDIR} -p ${T}${CALC_INCDIR}; \
 	    if [ ! -d "${T}${CALC_INCDIR}" ]; then \
 		echo ${MKDIR} -p "${T}${CALC_INCDIR}"; \
 		${MKDIR} -p "${T}${CALC_INCDIR}"; \
@@ -1248,8 +1301,8 @@ install: all
 	    ${TRUE}; \
 	fi
 	-${Q} if [ ! -d ${T}${HELPDIR} ]; then \
-	    echo ${MKDIR} ${T}${HELPDIR}; \
-	    ${MKDIR} ${T}${HELPDIR}; \
+	    echo ${MKDIR} -p ${T}${HELPDIR}; \
+	    ${MKDIR} -p ${T}${HELPDIR}; \
 	    if [ ! -d "${T}${HELPDIR}" ]; then \
 		echo ${MKDIR} -p "${T}${HELPDIR}"; \
 		${MKDIR} -p "${T}${HELPDIR}"; \
@@ -1260,8 +1313,8 @@ install: all
 	    ${TRUE}; \
 	fi
 	-${Q} if [ ! -d ${T}${CUSTOMCALDIR} ]; then \
-	    echo ${MKDIR} ${T}${CUSTOMCALDIR}; \
-	    ${MKDIR} ${T}${CUSTOMCALDIR}; \
+	    echo ${MKDIR} -p ${T}${CUSTOMCALDIR}; \
+	    ${MKDIR} -p ${T}${CUSTOMCALDIR}; \
 	    if [ ! -d "${T}${CUSTOMCALDIR}" ]; then \
 		echo ${MKDIR} -p "${T}${CUSTOMCALDIR}"; \
 		${MKDIR} -p "${T}${CUSTOMCALDIR}"; \
@@ -1272,8 +1325,8 @@ install: all
 	    ${TRUE}; \
 	fi
 	-${Q} if [ ! -d ${T}${CUSTOMHELPDIR} ]; then \
-	    echo ${MKDIR} ${T}${CUSTOMHELPDIR}; \
-	    ${MKDIR} ${T}${CUSTOMHELPDIR}; \
+	    echo ${MKDIR} -p ${T}${CUSTOMHELPDIR}; \
+	    ${MKDIR} -p ${T}${CUSTOMHELPDIR}; \
 	    if [ ! -d "${T}${CUSTOMHELPDIR}" ]; then \
 		echo ${MKDIR} -p "${T}${CUSTOMHELPDIR}"; \
 		${MKDIR} -p "${T}${CUSTOMHELPDIR}"; \
@@ -1284,8 +1337,8 @@ install: all
 	    ${TRUE}; \
 	fi
 	-${Q} if [ ! -d ${T}${CUSTOMINCDIR} ]; then \
-	    echo ${MKDIR} ${T}${CUSTOMINCDIR}; \
-	    ${MKDIR} ${T}${CUSTOMINCDIR}; \
+	    echo ${MKDIR} -p ${T}${CUSTOMINCDIR}; \
+	    ${MKDIR} -p ${T}${CUSTOMINCDIR}; \
 	    if [ ! -d "${T}${CUSTOMINCDIR}" ]; then \
 		echo ${MKDIR} -p "${T}${CUSTOMINCDIR}"; \
 		${MKDIR} -p "${T}${CUSTOMINCDIR}"; \
@@ -1333,7 +1386,8 @@ install: all
 		${RM} -f ${T}${CUSTOMHELPDIR}/$$i.new; \
 		${CP} -f $$i ${T}${CUSTOMHELPDIR}/$$i.new; \
 		${CHMOD} 0444 ${T}${CUSTOMHELPDIR}/$$i.new; \
-		${MV} -f ${T}${CUSTOMHELPDIR}/$$i.new ${T}${CUSTOMHELPDIR}/$$i; \
+		${MV} -f ${T}${CUSTOMHELPDIR}/$$i.new \
+			 ${T}${CUSTOMHELPDIR}/$$i; \
 		echo "installed ${T}${CUSTOMHELPDIR}/$$i"; \
 	    fi; \
 	done

custom/Makefile.head

@@ -2,7 +2,7 @@
 #
 # custom - makefile for calc custom routines
 #
-# Copyright (C) 1999-2006  Landon Curt Noll
+# Copyright (C) 1999-2006,2014,2017  Landon Curt Noll
 #
 # Calc is open software; you can redistribute it and/or modify it under
 # the terms of the version 2.1 of the GNU Lesser General Public License
@@ -18,9 +18,9 @@
 # received a copy with calc; if not, write to Free Software Foundation, Inc.
 # 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
 #
-# @(#) $Revision: 30.25 $
-# @(#) $Id: Makefile.head,v 30.25 2013/05/05 10:21:52 chongo Exp $
-# @(#) $Source: /usr/local/src/bin/calc/custom/RCS/Makefile.head,v $
+# @(#) $Revision: 30.42 $
+# @(#) $Id: Makefile.head,v 30.42 2017/05/19 16:09:14 chongo Exp $
+# @(#) $Source: /usr/local/src/bin/calc-RHEL7/custom/RCS/Makefile.head,v $
 #
 # Under source code control:	1997/03/09 02:28:54
 # File existed as early as:	1997
@@ -337,20 +337,6 @@ CALC_ENV= CALCPATH=./cal
 ALLOW_CUSTOM= -DCUSTOM
 #ALLOW_CUSTOM=
 
-# The install rule uses:
-#
-#	${MKDIR} ${MKDIR_ARG}
-#
-# to create directorties.  Normall this amounts to usins mkdir -p dir ...
-# Some older systems may not have mkdir -p.  If you system does not
-# make mkdir -p, then set MKDIR_ARG to empty.
-#
-# MKDIR_ARG= -p			# use mkdir -p when creating paths
-# MKDIR_ARG=			# use if system does not understand mkdir -p
-#
-MKDIR_ARG= -p
-#MKDIR_ARG=
-
 # Some out of date operating systems require / want an executable to
 # end with a certain file extension.  Some compile systems such as
 # Cygwin build calc as calc.exe.  The EXT variable is used to denote
@@ -366,18 +352,12 @@ EXT=
 
 # The default calc versions
 #
-VERSION= 2.12.4.7
-VERS= 2.12.4
-VER= 2.12
-VE= 2
+VERSION= 2.12.5.6
 
 # Names of shared libraries with versions
 #
 LIB_EXT= .so
 LIB_EXT_VERSION= ${LIB_EXT}.${VERSION}
-LIB_EXT_VERS= ${LIB_EXT}.${VERS}
-LIB_EXT_VER= ${LIB_EXT}.${VER}
-LIB_EXT_VE= ${LIB_EXT}.${VE}
 
 # standard tools
 #

custom/Makefile.simple

@@ -2,7 +2,7 @@
 #
 # custom - makefile for calc custom routines
 #
-# Copyright (C) 1999-2006  Landon Curt Noll
+# Copyright (C) 1999-2006,2014,2017  Landon Curt Noll
 #
 # Calc is open software; you can redistribute it and/or modify it under
 # the terms of the version 2.1 of the GNU Lesser General Public License
@@ -18,9 +18,9 @@
 # received a copy with calc; if not, write to Free Software Foundation, Inc.
 # 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
 #
-# @(#) $Revision: 30.25 $
-# @(#) $Id: Makefile.head,v 30.25 2013/05/05 10:21:52 chongo Exp $
-# @(#) $Source: /usr/local/src/bin/calc/custom/RCS/Makefile.head,v $
+# @(#) $Revision: 30.42 $
+# @(#) $Id: Makefile.head,v 30.42 2017/05/19 16:09:14 chongo Exp $
+# @(#) $Source: /usr/local/src/bin/calc-RHEL7/custom/RCS/Makefile.head,v $
 #
 # Under source code control:	1997/03/09 02:28:54
 # File existed as early as:	1997
@@ -322,20 +322,6 @@ CALC_ENV= CALCPATH=./cal
 ALLOW_CUSTOM= -DCUSTOM
 #ALLOW_CUSTOM=
 
-# The install rule uses:
-#
-#	${MKDIR} ${MKDIR_ARG}
-#
-# to create directorties.  Normall this amounts to usins mkdir -p dir ...
-# Some older systems may not have mkdir -p.  If you system does not
-# make mkdir -p, then set MKDIR_ARG to empty.
-#
-# MKDIR_ARG= -p			# use mkdir -p when creating paths
-# MKDIR_ARG=			# use if system does not understand mkdir -p
-#
-MKDIR_ARG= -p
-#MKDIR_ARG=
-
 # Some out of date operating systems require / want an executable to
 # end with a certain file extension.  Some compile systems such as
 # Cygwin build calc as calc.exe.  The EXT variable is used to denote
@@ -351,18 +337,12 @@ EXT=
 
 # The default calc versions
 #
-VERSION= 2.12.4.7
-VERS= 2.12.4
-VER= 2.12
-VE= 2
+VERSION= 2.12.5.6
 
 # Names of shared libraries with versions
 #
 LIB_EXT= .so
 LIB_EXT_VERSION= ${LIB_EXT}.${VERSION}
-LIB_EXT_VERS= ${LIB_EXT}.${VERS}
-LIB_EXT_VER= ${LIB_EXT}.${VER}
-LIB_EXT_VE= ${LIB_EXT}.${VE}
 
 # standard tools
 #
@@ -433,30 +413,41 @@ COMMON_LDFLAGS= ${EXTRA_LDFLAGS}
 #	BLD_TYPE= calc-static-only
 #
 # CC_SHARE are flags given to ${CC} to build .o files suitable for shared libs
-# DEFAULT_LIB_INSTALL_PATH is where calc progs look for calc shared libs
+# DEFAULT_LIB_INSTALL_PATH is where calc programs look for calc shared libs
 # LD_SHARE are common flags given to ${CC} to link with shared libraries
-# LIBCALC_SHLIB are flags given to ${CC} to build libcalc shared lib
+# LIBCALC_SHLIB are flags given to ${CC} to build libcalc shared libraries
 # LIBCUSTCALC_SHLIB are flags given to ${CC} to build libcustcalc shared lib
 #
-#	NOTE: The above 4 values are unused if BLD_TYPE= calc-static-only
+#	NOTE: The above 5 values are unused if BLD_TYPE= calc-static-only
 #
 # CC_STATIC are flags given to ${CC} to build .o files suitable for static libs
 # LD_STATIC are common flags given to ${CC} to link with static libraries
-# LIBCALC_STATIC are flags given to ${CC} to build libcalc static lib
+# LIBCALC_STATIC are flags given to ${CC} to build libcalc static libraries
 # LIBCUSTCALC_STATIC are flags given to ${CC} to build libcustcalc static lib
 #
 #	NOTE: The above 4 values are unused if BLD_TYPE= calc-dynamic-only
 #
 # CCOPT are flags given to ${CC} for optimization
 # CCWARN are flags given to ${CC} for warning message control
+#
+# WNO_IMPLICT, WNO_ERROR_LONG_LONG and WNO_LONG_LONG are given to ${CC}
+#	when compiling special .o files that may need special compile options
+#	NOTE: These flags simply turn off certain compiler warnings,
+#	      which is useful only when CCWERR is set to -Werror.
+#	NOTE: If your compiler does not have these -Wno files, just
+#	      set these variables to nothing as in:
+#		WNO_IMPLICT=
+#		WNO_ERROR_LONG_LONG=
+#		WNO_LONG_LONG=
+#
 # CCWERR are flags given to ${CC} to make warnings fatal errors
 #	NOTE: CCWERR is only set in development Makefiles and must only be
 #	      used with ${CC}, not ${LCC}.  If you do not want the compiler
 #	      to abort on warnings, then leave CCWERR blank.
 # CCMISC are misc flags given to ${CC}
 #
-# LCC how the the C compiler is invoked on locally executed intermediate progs
-# CC is how the the C compiler is invoked (with an optional Purify)
+# LCC is how the C compiler is invoked on locally executed intermediate programs
+# CC is how the C compiler is invoked (with an optional Purify)
 #
 # Specific target overrides or modifications to default values
 
@@ -485,7 +476,10 @@ LD_STATIC=
 LIBCALC_STATIC=
 LIBCUSTCALC_STATIC=
 #
-CCWARN= -Wall -W -Wno-comment
+CCWARN= -Wall
+WNO_IMPLICT= -Wno-implicit
+WNO_ERROR_LONG_LONG= -Wno-error=long-long
+WNO_LONG_LONG= -Wno-long-long
 CCWERR=
 CCOPT= ${DEBUG}
 CCMISC=
@@ -495,10 +489,10 @@ CC= ${PURIFY} ${LCC} ${CCWERR}
 #
 # The simple makefile forces the use of static ${CC} flags
 #
-# ICFLAGS are given to ${CC} for intermediate progs used to help compile calc
-# CFLAGS are given to ${CC} for calc progs other than intermediate progs
-# ILDFLAGS for ${CC} in linking intermediate progs used to help compile calc
-# LDFLAGS for ${CC} in linking calc progs other than intermediate progs
+# ICFLAGS are given to ${CC} for intermediate programs used to help compile calc
+# CFLAGS are given to ${CC} for calc programs other than intermediate programs
+# ILDFLAGS for ${CC} in linking intermediate programs used to help compile calc
+# LDFLAGS for ${CC} in linking calc programs other than intermediate programs
 #
 ICFLAGS= ${COMMON_CFLAGS} ${CC_STATIC}
 CFLAGS= ${ICFLAGS} ${CCOPT}
@@ -506,7 +500,7 @@ CFLAGS= ${ICFLAGS} ${CCOPT}
 ILDFLAGS= ${COMMON_LDFLAGS} ${LD_STATIC}
 LDFLAGS= ${LD_DEBUG} ${ILDFLAGS} ${LIBCALC_STATIC} ${LIBCUSTCALC_STATIC}
 #
-CALCPATH= .;./cal;~/.cal;${CALC_SHAREDIR};${CUSTOMCALDIR}
+CALCPATH= .;./cal;~/.cal;${T}${CALC_SHAREDIR};${T}${CUSTOMCALDIR}
 
 #######################################################################
 #-=-=-=-=-=- end of target section - only make rules below -=-=-=-=-=-#
@@ -679,8 +673,8 @@ depend:
 	fi
 	${Q} echo forming custom/skel
 	${Q} ${RM} -rf skel
-	${Q} ${MKDIR} skel
-	${Q} ${MKDIR} skel/custom
+	${Q} ${MKDIR} -p skel
+	${Q} ${MKDIR} -p skel/custom
 	-${Q} for i in ${C_SRC} /dev/null; do \
 	    if [ X"$$i" != X"/dev/null" ]; then \
 		${SED} -n '/^#[	 ]*include[	 ]*"/p' \
@@ -729,7 +723,8 @@ depend:
 	${Q} echo forming new custom/${MAKE_FILE}
 	${Q} ${RM} -f ${MAKE_FILE}.bak
 	${Q} ${MV} ${MAKE_FILE} ${MAKE_FILE}.bak
-	${Q} ${SED} -n '1,/^# DO NOT DELETE THIS LINE/p' ${MAKE_FILE}.bak > ${MAKE_FILE}
+	${Q} ${SED} -n '1,/^# DO NOT DELETE THIS LINE/p' \
+		    ${MAKE_FILE}.bak > ${MAKE_FILE}
 	${Q} ${GREP} -v '^#' skel/custom/makedep.out >> ${MAKE_FILE}
 	${Q} ${RM} -rf skel
 	-${Q} if ${CMP} -s ${MAKE_FILE}.bak ${MAKE_FILE}; then \
@@ -790,20 +785,35 @@ clobber: clean
 	${RM} -f ${BLD_TYPE}
 	${RM} -f .all ${MAKE_FILE}.tmp ${MAKE_FILE}.bak
 	${RM} -f libcustcalc${LIB_EXT_VERSION}
-	${RM} -f libcustcalc${LIB_EXT_VERS}
-	${RM} -f libcustcalc${LIB_EXT_VER}
-	${RM} -f libcustcalc${LIB_EXT_VE}
 	${RM} -f libcustcalc${LIB_EXT}
 	${RM} -f libcustcalc.a
+	${RM} -f libcustcalc*
+	${V} echo remove files that are obsolete
+	-${Q} if [ -e .DS_Store ]; then \
+	    echo ${RM} -rf .DS_Store; \
+	    ${RM} -rf .DS_Store; \
+	fi
 
 # install everything
 #
 # NOTE: Keep the uninstall rule in reverse order to the install rule
 #
 install: all
+	-${Q} if [ ! -d ${T}${LIBDIR} ]; then \
+	    echo ${MKDIR} -p ${T}${LIBDIR}; \
+	    ${MKDIR} -p ${T}${LIBDIR}; \
+	    if [ ! -d "${T}${LIBDIR}" ]; then \
+		echo ${MKDIR} -p "${T}${LIBDIR}"; \
+		${MKDIR} -p "${T}${LIBDIR}"; \
+	    fi; \
+	    echo ${CHMOD} 0755 ${T}${LIBDIR}; \
+	    ${CHMOD} 0755 ${T}${LIBDIR}; \
+	else \
+	    ${TRUE}; \
+	fi
 	-${Q} if [ ! -d ${T}${INCDIR} ]; then \
-	    echo ${MKDIR} ${T}${INCDIR}; \
-	    ${MKDIR} ${T}${INCDIR}; \
+	    echo ${MKDIR} -p ${T}${INCDIR}; \
+	    ${MKDIR} -p ${T}${INCDIR}; \
 	    if [ ! -d "${T}${INCDIR}" ]; then \
 		echo ${MKDIR} -p "${T}${INCDIR}"; \
 		${MKDIR} -p "${T}${INCDIR}"; \
@@ -814,8 +824,8 @@ install: all
 	    ${TRUE}; \
 	fi
 	-${Q} if [ ! -d ${T}${CALC_SHAREDIR} ]; then \
-	    echo ${MKDIR} ${T}${CALC_SHAREDIR}; \
-	    ${MKDIR} ${T}${CALC_SHAREDIR}; \
+	    echo ${MKDIR} -p ${T}${CALC_SHAREDIR}; \
+	    ${MKDIR} -p ${T}${CALC_SHAREDIR}; \
 	    if [ ! -d "${T}${CALC_SHAREDIR}" ]; then \
 		echo ${MKDIR} -p "${T}${CALC_SHAREDIR}"; \
 		${MKDIR} -p "${T}${CALC_SHAREDIR}"; \
@@ -826,8 +836,8 @@ install: all
 	    ${TRUE}; \
 	fi
 	-${Q} if [ ! -d ${T}${CALC_INCDIR} ]; then \
-	    echo ${MKDIR} ${T}${CALC_INCDIR}; \
-	    ${MKDIR} ${T}${CALC_INCDIR}; \
+	    echo ${MKDIR} -p ${T}${CALC_INCDIR}; \
+	    ${MKDIR} -p ${T}${CALC_INCDIR}; \
 	    if [ ! -d "${T}${CALC_INCDIR}" ]; then \
 		echo ${MKDIR} -p "${T}${CALC_INCDIR}"; \
 		${MKDIR} -p "${T}${CALC_INCDIR}"; \
@@ -838,8 +848,8 @@ install: all
 	    ${TRUE}; \
 	fi
 	-${Q} if [ ! -d ${T}${HELPDIR} ]; then \
-	    echo ${MKDIR} ${T}${HELPDIR}; \
-	    ${MKDIR} ${T}${HELPDIR}; \
+	    echo ${MKDIR} -p ${T}${HELPDIR}; \
+	    ${MKDIR} -p ${T}${HELPDIR}; \
 	    if [ ! -d "${T}${HELPDIR}" ]; then \
 		echo ${MKDIR} -p "${T}${HELPDIR}"; \
 		${MKDIR} -p "${T}${HELPDIR}"; \
@@ -850,8 +860,8 @@ install: all
 	    ${TRUE}; \
 	fi
 	-${Q} if [ ! -d ${T}${CUSTOMCALDIR} ]; then \
-	    echo ${MKDIR} ${T}${CUSTOMCALDIR}; \
-	    ${MKDIR} ${T}${CUSTOMCALDIR}; \
+	    echo ${MKDIR} -p ${T}${CUSTOMCALDIR}; \
+	    ${MKDIR} -p ${T}${CUSTOMCALDIR}; \
 	    if [ ! -d "${T}${CUSTOMCALDIR}" ]; then \
 		echo ${MKDIR} -p "${T}${CUSTOMCALDIR}"; \
 		${MKDIR} -p "${T}${CUSTOMCALDIR}"; \
@@ -862,8 +872,8 @@ install: all
 	    ${TRUE}; \
 	fi
 	-${Q} if [ ! -d ${T}${CUSTOMHELPDIR} ]; then \
-	    echo ${MKDIR} ${T}${CUSTOMHELPDIR}; \
-	    ${MKDIR} ${T}${CUSTOMHELPDIR}; \
+	    echo ${MKDIR} -p ${T}${CUSTOMHELPDIR}; \
+	    ${MKDIR} -p ${T}${CUSTOMHELPDIR}; \
 	    if [ ! -d "${T}${CUSTOMHELPDIR}" ]; then \
 		echo ${MKDIR} -p "${T}${CUSTOMHELPDIR}"; \
 		${MKDIR} -p "${T}${CUSTOMHELPDIR}"; \
@@ -874,8 +884,8 @@ install: all
 	    ${TRUE}; \
 	fi
 	-${Q} if [ ! -d ${T}${CUSTOMINCDIR} ]; then \
-	    echo ${MKDIR} ${T}${CUSTOMINCDIR}; \
-	    ${MKDIR} ${T}${CUSTOMINCDIR}; \
+	    echo ${MKDIR} -p ${T}${CUSTOMINCDIR}; \
+	    ${MKDIR} -p ${T}${CUSTOMINCDIR}; \
 	    if [ ! -d "${T}${CUSTOMINCDIR}" ]; then \
 		echo ${MKDIR} -p "${T}${CUSTOMINCDIR}"; \
 		${MKDIR} -p "${T}${CUSTOMINCDIR}"; \
@@ -923,7 +933,8 @@ install: all
 		${RM} -f ${T}${CUSTOMHELPDIR}/$$i.new; \
 		${CP} -f $$i ${T}${CUSTOMHELPDIR}/$$i.new; \
 		${CHMOD} 0444 ${T}${CUSTOMHELPDIR}/$$i.new; \
-		${MV} -f ${T}${CUSTOMHELPDIR}/$$i.new ${T}${CUSTOMHELPDIR}/$$i; \
+		${MV} -f ${T}${CUSTOMHELPDIR}/$$i.new \
+			 ${T}${CUSTOMHELPDIR}/$$i; \
 		echo "installed ${T}${CUSTOMHELPDIR}/$$i"; \
 	    fi; \
 	done

custom/Makefile.tail

@@ -177,8 +177,8 @@ depend:
 	fi
 	${Q} echo forming custom/skel
 	${Q} ${RM} -rf skel
-	${Q} ${MKDIR} skel
-	${Q} ${MKDIR} skel/custom
+	${Q} ${MKDIR} -p skel
+	${Q} ${MKDIR} -p skel/custom
 	-${Q} for i in ${C_SRC} /dev/null; do \
 	    if [ X"$$i" != X"/dev/null" ]; then \
 		${SED} -n '/^#[	 ]*include[	 ]*"/p' \
@@ -227,7 +227,8 @@ depend:
 	${Q} echo forming new custom/${MAKE_FILE}
 	${Q} ${RM} -f ${MAKE_FILE}.bak
 	${Q} ${MV} ${MAKE_FILE} ${MAKE_FILE}.bak
-	${Q} ${SED} -n '1,/^# DO NOT DELETE THIS LINE/p' ${MAKE_FILE}.bak > ${MAKE_FILE}
+	${Q} ${SED} -n '1,/^# DO NOT DELETE THIS LINE/p' \
+		    ${MAKE_FILE}.bak > ${MAKE_FILE}
 	${Q} ${GREP} -v '^#' skel/custom/makedep.out >> ${MAKE_FILE}
 	${Q} ${RM} -rf skel
 	-${Q} if ${CMP} -s ${MAKE_FILE}.bak ${MAKE_FILE}; then \
@@ -288,23 +289,38 @@ clobber: clean
 	${RM} -f ${BLD_TYPE}
 	${RM} -f .all ${MAKE_FILE}.tmp ${MAKE_FILE}.bak
 	${RM} -f libcustcalc${LIB_EXT_VERSION}
-	${RM} -f libcustcalc${LIB_EXT_VERS}
-	${RM} -f libcustcalc${LIB_EXT_VER}
-	${RM} -f libcustcalc${LIB_EXT_VE}
 	${RM} -f libcustcalc${LIB_EXT}
 	${RM} -f libcustcalc.a
+	${RM} -f libcustcalc*
+	${V} echo remove files that are obsolete
 #if 0	/* start of skip for non-Gnu makefiles */
 	${RM} -f Makefile.simple
 #endif	/* end of skip for non-Gnu makefiles */
+	-${Q} if [ -e .DS_Store ]; then \
+	    echo ${RM} -rf .DS_Store; \
+	    ${RM} -rf .DS_Store; \
+	fi
 
 # install everything
 #
 # NOTE: Keep the uninstall rule in reverse order to the install rule
 #
 install: all
+	-${Q} if [ ! -d ${T}${LIBDIR} ]; then \
+	    echo ${MKDIR} -p ${T}${LIBDIR}; \
+	    ${MKDIR} -p ${T}${LIBDIR}; \
+	    if [ ! -d "${T}${LIBDIR}" ]; then \
+		echo ${MKDIR} -p "${T}${LIBDIR}"; \
+		${MKDIR} -p "${T}${LIBDIR}"; \
+	    fi; \
+	    echo ${CHMOD} 0755 ${T}${LIBDIR}; \
+	    ${CHMOD} 0755 ${T}${LIBDIR}; \
+	else \
+	    ${TRUE}; \
+	fi
 	-${Q} if [ ! -d ${T}${INCDIR} ]; then \
-	    echo ${MKDIR} ${T}${INCDIR}; \
-	    ${MKDIR} ${T}${INCDIR}; \
+	    echo ${MKDIR} -p ${T}${INCDIR}; \
+	    ${MKDIR} -p ${T}${INCDIR}; \
 	    if [ ! -d "${T}${INCDIR}" ]; then \
 		echo ${MKDIR} -p "${T}${INCDIR}"; \
 		${MKDIR} -p "${T}${INCDIR}"; \
@@ -315,8 +331,8 @@ install: all
 	    ${TRUE}; \
 	fi
 	-${Q} if [ ! -d ${T}${CALC_SHAREDIR} ]; then \
-	    echo ${MKDIR} ${T}${CALC_SHAREDIR}; \
-	    ${MKDIR} ${T}${CALC_SHAREDIR}; \
+	    echo ${MKDIR} -p ${T}${CALC_SHAREDIR}; \
+	    ${MKDIR} -p ${T}${CALC_SHAREDIR}; \
 	    if [ ! -d "${T}${CALC_SHAREDIR}" ]; then \
 		echo ${MKDIR} -p "${T}${CALC_SHAREDIR}"; \
 		${MKDIR} -p "${T}${CALC_SHAREDIR}"; \
@@ -327,8 +343,8 @@ install: all
 	    ${TRUE}; \
 	fi
 	-${Q} if [ ! -d ${T}${CALC_INCDIR} ]; then \
-	    echo ${MKDIR} ${T}${CALC_INCDIR}; \
-	    ${MKDIR} ${T}${CALC_INCDIR}; \
+	    echo ${MKDIR} -p ${T}${CALC_INCDIR}; \
+	    ${MKDIR} -p ${T}${CALC_INCDIR}; \
 	    if [ ! -d "${T}${CALC_INCDIR}" ]; then \
 		echo ${MKDIR} -p "${T}${CALC_INCDIR}"; \
 		${MKDIR} -p "${T}${CALC_INCDIR}"; \
@@ -339,8 +355,8 @@ install: all
 	    ${TRUE}; \
 	fi
 	-${Q} if [ ! -d ${T}${HELPDIR} ]; then \
-	    echo ${MKDIR} ${T}${HELPDIR}; \
-	    ${MKDIR} ${T}${HELPDIR}; \
+	    echo ${MKDIR} -p ${T}${HELPDIR}; \
+	    ${MKDIR} -p ${T}${HELPDIR}; \
 	    if [ ! -d "${T}${HELPDIR}" ]; then \
 		echo ${MKDIR} -p "${T}${HELPDIR}"; \
 		${MKDIR} -p "${T}${HELPDIR}"; \
@@ -351,8 +367,8 @@ install: all
 	    ${TRUE}; \
 	fi
 	-${Q} if [ ! -d ${T}${CUSTOMCALDIR} ]; then \
-	    echo ${MKDIR} ${T}${CUSTOMCALDIR}; \
-	    ${MKDIR} ${T}${CUSTOMCALDIR}; \
+	    echo ${MKDIR} -p ${T}${CUSTOMCALDIR}; \
+	    ${MKDIR} -p ${T}${CUSTOMCALDIR}; \
 	    if [ ! -d "${T}${CUSTOMCALDIR}" ]; then \
 		echo ${MKDIR} -p "${T}${CUSTOMCALDIR}"; \
 		${MKDIR} -p "${T}${CUSTOMCALDIR}"; \
@@ -363,8 +379,8 @@ install: all
 	    ${TRUE}; \
 	fi
 	-${Q} if [ ! -d ${T}${CUSTOMHELPDIR} ]; then \
-	    echo ${MKDIR} ${T}${CUSTOMHELPDIR}; \
-	    ${MKDIR} ${T}${CUSTOMHELPDIR}; \
+	    echo ${MKDIR} -p ${T}${CUSTOMHELPDIR}; \
+	    ${MKDIR} -p ${T}${CUSTOMHELPDIR}; \
 	    if [ ! -d "${T}${CUSTOMHELPDIR}" ]; then \
 		echo ${MKDIR} -p "${T}${CUSTOMHELPDIR}"; \
 		${MKDIR} -p "${T}${CUSTOMHELPDIR}"; \
@@ -375,8 +391,8 @@ install: all
 	    ${TRUE}; \
 	fi
 	-${Q} if [ ! -d ${T}${CUSTOMINCDIR} ]; then \
-	    echo ${MKDIR} ${T}${CUSTOMINCDIR}; \
-	    ${MKDIR} ${T}${CUSTOMINCDIR}; \
+	    echo ${MKDIR} -p ${T}${CUSTOMINCDIR}; \
+	    ${MKDIR} -p ${T}${CUSTOMINCDIR}; \
 	    if [ ! -d "${T}${CUSTOMINCDIR}" ]; then \
 		echo ${MKDIR} -p "${T}${CUSTOMINCDIR}"; \
 		${MKDIR} -p "${T}${CUSTOMINCDIR}"; \
@@ -424,7 +440,8 @@ install: all
 		${RM} -f ${T}${CUSTOMHELPDIR}/$$i.new; \
 		${CP} -f $$i ${T}${CUSTOMHELPDIR}/$$i.new; \
 		${CHMOD} 0444 ${T}${CUSTOMHELPDIR}/$$i.new; \
-		${MV} -f ${T}${CUSTOMHELPDIR}/$$i.new ${T}${CUSTOMHELPDIR}/$$i; \
+		${MV} -f ${T}${CUSTOMHELPDIR}/$$i.new \
+			 ${T}${CUSTOMHELPDIR}/$$i; \
 		echo "installed ${T}${CUSTOMHELPDIR}/$$i"; \
 	    fi; \
 	done

custom/c_sysinfo.c

@@ -17,8 +17,8 @@
  * received a copy with calc; if not, write to Free Software Foundation, Inc.
  * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
  *
- * @(#) $Revision: 30.2 $
- * @(#) $Id: c_sysinfo.c,v 30.2 2007/07/05 19:35:20 chongo Exp $
+ * @(#) $Revision: 30.4 $
+ * @(#) $Id: c_sysinfo.c,v 30.4 2013/09/01 22:16:21 chongo Exp $
  * @(#) $Source: /usr/local/src/bin/calc/custom/RCS/c_sysinfo.c,v $
  *
  * Under source code control:	1997/03/09 23:14:40
@@ -71,93 +71,180 @@ struct infoname {
 	FULL nmbr;	/* if str==NULL ==> value fo #define as a FULL */
 };
 STATIC struct infoname sys_info[] = {
-    {"S100", "slots in an subtractive 100 table", NULL, (FULL)S100},
-    {"BASE", "base for calculations", NULL, (FULL)BASE},
-    {"BASE1", "one less than base", NULL, (FULL)BASE},
-    {"BASEB", "bits in the calculation base", NULL, (FULL)BASEB},
-    {"BASEDIG", "number of digits in base", NULL, (FULL)BASEDIG},
-    {"BIG_ENDIAN", "Most Significant Byte first symbol", NULL, (FULL)BIG_ENDIAN},
-    {"BLK_CHUNKSIZE", "default allocation chunk size for blocks", NULL, (FULL)BLK_CHUNKSIZE},
-    {"BLK_DEF_MAXPRINT", "default block octets to print", NULL, (FULL)BLK_DEF_MAXPRINT},
-    {"BLUM_PREGEN", "non-default predefined Blum generators", NULL, (FULL)BLUM_PREGEN},
-    {"CALCEXT", "extension for files read in", CALCEXT, (FULL)0},
-    {"CALC_BYTE_ORDER", "Byte order (LITTLE_ENDIAN or BIG_ENDIAN)", NULL, (FULL)CALC_BYTE_ORDER},
-    {"CUSTOMHELPDIR", "location of the custom help directory", CUSTOMHELPDIR, (FULL)0},
-    {"DEFAULTCALCBINDINGS", "default key bindings file", DEFAULTCALCBINDINGS, (FULL)0},
-    {"DEFAULTCALCHELP", "help file that -h prints", DEFAULTCALCHELP, (FULL)0},
-    {"DEFAULTCALCPAGER", "default pager", DEFAULTCALCPAGER, (FULL)0},
-    {"DEFAULTCALCPATH", "default :-separated search path", DEFAULTCALCPATH, (FULL)0},
-    {"DEFAULTCALCRC", "default :-separated startup file list", DEFAULTCALCRC, (FULL)0},
-    {"DEFAULTSHELL", "default shell to use", DEFAULTSHELL, (FULL)0},
-    {"DEV_BITS", "device number size in bits", NULL, (FULL)DEV_BITS},
-    {"DISPLAY_DEFAULT", "default digits for float display", NULL, (FULL)DISPLAY_DEFAULT},
-    {"EPSILONPREC_DEFAULT", "2^-EPSILON_DEFAULT <= EPSILON_DEFAULT", NULL, (FULL)EPSILONPREC_DEFAULT},
-    {"EPSILON_DEFAULT", "allowed error for float calculations", EPSILON_DEFAULT, (FULL)0},
-    {"ERRMAX", "default errmax value", NULL, (FULL)ERRMAX},
-    {"E_USERDEF", "base of user defined errors", NULL, (FULL)E_USERDEF},
-    {"E__BASE", "calc errors start above here", NULL, (FULL)E__BASE},
-    {"E__COUNT", "number of calc errors", NULL, (FULL)E__COUNT},
-    {"E__HIGHEST", "highest calc error", NULL, (FULL)E__HIGHEST},
-    {"FALSE", "boolean false", NULL, (FULL)FALSE},
-    {"FILEPOS_BITS", "file position size in bits", NULL, (FULL)FILEPOS_BITS},
-    {"FULL_BITS", "bits in a FULL", NULL, (FULL)FULL_BITS},
-    {"HELPDIR", "location of the help directory", HELPDIR, (FULL)0},
-    {"HIST_BINDING_FILE", "Default binding file", HIST_BINDING_FILE, (FULL)0},
-    {"HIST_SIZE", "Default history size", NULL, (FULL)HIST_SIZE},
-    {"INIT_J", "initial 1st walking a55 table index", NULL, (FULL)INIT_J},
-    {"INIT_K", "initial 2nd walking a55 table index", NULL, (FULL)INIT_K},
-    {"INODE_BITS", "inode number size in bits", NULL, (FULL)INODE_BITS},
-    {"LITTLE_ENDIAN", "Least Significant Byte first symbol", NULL, (FULL)LITTLE_ENDIAN},
-    {"LONG_BITS", "bit length of a long", NULL, (FULL)LONG_BITS},
-    {"MAP_POPCNT", "number of odd primes in pr_map", NULL, (FULL)MAP_POPCNT},
-    {"MAX_CALCRC", "maximum allowed length of $CALCRC", NULL, (FULL)MAX_CALCRC},
-    {"MAXCMD", "max length of command invocation", NULL, (FULL)MAXCMD},
-    {"MAXDIM", "max number of dimensions in matrices", NULL, (FULL)MAXDIM},
-    {"MAXERROR", "max length of error message string", NULL, (FULL)MAXERROR},
-    {"MAXFILES", "max number of opened files", NULL, (FULL)MAXFILES},
-    {"MAXFULL", "largest SFULL value", NULL, (FULL)MAXFULL},
-    {"MAXHALF", "largest SHALF value", NULL, (FULL)MAXHALF},
-    {"MAXLABELS", "max number of user labels in function", NULL, (FULL)MAXLABELS},
-    {"MAXLEN", "longest storage size allowed", NULL, (FULL)MAXLEN},
-    {"MAXLONG", "largest long val", NULL, (FULL)MAXLONG},
-    {"MAXPRINT_DEFAULT", "default number of elements printed", NULL, (FULL)MAXPRINT_DEFAULT},
-    {"MAXREDC", "number of entries in REDC cache", NULL, (FULL)MAXREDC},
-    {"MAXSCANCOUNT", "default max scan errors before an abort", NULL, (FULL)MAXSCANCOUNT},
-    {"MAXSTACK", "max depth of evaluation stack", NULL, (FULL)MAXSTACK},
-    {"MAXSTRING", "max size of string constant", NULL, (FULL)MAXSTRING},
-    {"MAXUFULL", "largest FULL value", NULL, (FULL)MAXUFULL},
-    {"MAXULONG", "largest unsigned long val", NULL, (FULL)MAXULONG},
-    {"MAX_MAP_PRIME", "larest prime in pr_map", NULL, (FULL)MAX_MAP_PRIME},
-    {"MAX_MAP_VAL", "larest bit in pr_map", NULL, (FULL)MAX_MAP_VAL},
-    {"MAX_PFACT_VAL", "max x, for which pfact(x) is a long", NULL, (FULL)MAX_PFACT_VAL},
-    {"MAX_SM_PRIME", "larest 32 bit prime", NULL, (FULL)MAX_SM_PRIME},
-    {"MAX_SM_VAL", "larest 32 bit value", NULL, (FULL)MAX_SM_VAL},
-    {"MUL_ALG2", "default size for alternative multiply", NULL, (FULL)MUL_ALG2},
-    {"NXT_MAP_PRIME", "smallest odd prime not in pr_map", NULL, (FULL)NXT_MAP_PRIME},
-    {"NXT_PFACT_VAL", "next prime for higher pfact values", NULL, (FULL)NXT_PFACT_VAL},
-    {"OFF_T_BITS", "file offset size in bits", NULL, (FULL)OFF_T_BITS},
-    {"PIX_32B", "max pix() value", NULL, (FULL)PIX_32B},
-    {"POW_ALG2", "default size for using REDC for powers", NULL, (FULL)POW_ALG2},
-    {"REDC_ALG2", "default size using alternative REDC alg", NULL, (FULL)REDC_ALG2},
-    {"SBITS", "size of additive or shuffle entry in bits", NULL, (FULL)SBITS},
-    {"SBYTES", "size of additive or shuffle entry in bytes", NULL, (FULL)SBYTES},
-    {"SCNT", "length of additive 55 table in FULLs", NULL, (FULL)SCNT},
-    {"SEEDXORBITS", "low bits of a55 seed devoted to xor", NULL, (FULL)SEEDXORBITS},
-    {"SHALFS", "size of additive or shuffle entry in HALFs", NULL, (FULL)SHALFS},
-    {"SHUFCNT", "size of shuffle table in entries", NULL, (FULL)SHUFCNT},
-    {"SHUFLEN", "length of shuffle table in FULLs", NULL, (FULL)SHUFLEN},
-    {"SHUFMASK", "mask for shuffle table entry selection", NULL, (FULL)SHUFMASK},
-    {"SHUFPOW", "power of 2 size of the shuffle table", NULL, (FULL)SHUFPOW},
-    {"SLEN", "number of FULLs in a shuffle table entry", NULL, (FULL)SLEN},
-    {"SQ_ALG2", "default size for alternative squaring", NULL, (FULL)SQ_ALG2},
-    {"SYMBOLSIZE", "max symbol name size", NULL, (FULL)SYMBOLSIZE},
-    {"TEN_MAX", "10^(2^TEN_MAX): largest base10 conversion const", NULL, (FULL)TEN_MAX},
-    {"TOPFULL", "highest bit in FULL", NULL, (FULL)TOPFULL},
-    {"TOPHALF", "highest bit in a HALF", NULL, (FULL)TOPHALF},
-    {"TOPLONG", "top long bit", NULL, (FULL)TOPLONG},
-    {"TRUE", "boolean true", NULL, (FULL)TRUE},
-    {"USUAL_ELEMENTS", "usual number of elements for objects", NULL, (FULL)USUAL_ELEMENTS},
-    {"REGNUM_MAX", "highest custom register number", NULL, (FULL)CUSTOM_REG_MAX},
+    {"S100", "slots in an subtractive 100 table", NULL,
+     (FULL)S100},
+    {"BASE", "base for calculations", NULL,
+     (FULL)BASE},
+    {"BASE1", "one less than base", NULL,
+     (FULL)BASE},
+    {"BASEB", "bits in the calculation base", NULL,
+     (FULL)BASEB},
+    {"BASEDIG", "number of digits in base", NULL,
+     (FULL)BASEDIG},
+    {"BIG_ENDIAN", "Most Significant Byte first symbol", NULL,
+     (FULL)BIG_ENDIAN},
+    {"BLK_CHUNKSIZE", "default allocation chunk size for blocks", NULL,
+     (FULL)BLK_CHUNKSIZE},
+    {"BLK_DEF_MAXPRINT", "default block octets to print", NULL,
+     (FULL)BLK_DEF_MAXPRINT},
+    {"BLUM_PREGEN", "non-default predefined Blum generators", NULL,
+     (FULL)BLUM_PREGEN},
+    {"CALCEXT", "extension for files read in", CALCEXT,
+     (FULL)0},
+    {"CALC_BYTE_ORDER", "Byte order (LITTLE_ENDIAN or BIG_ENDIAN)", NULL,
+     (FULL)CALC_BYTE_ORDER},
+    {"CUSTOMHELPDIR", "location of the custom help directory", CUSTOMHELPDIR,
+     (FULL)0},
+    {"DEFAULTCALCBINDINGS", "default key bindings file", DEFAULTCALCBINDINGS,
+     (FULL)0},
+    {"DEFAULTCALCHELP", "help file that -h prints", DEFAULTCALCHELP,
+     (FULL)0},
+    {"DEFAULTCALCPAGER", "default pager", DEFAULTCALCPAGER,
+     (FULL)0},
+    {"DEFAULTCALCPATH", "default :-separated search path", DEFAULTCALCPATH,
+     (FULL)0},
+    {"DEFAULTCALCRC", "default :-separated startup file list", DEFAULTCALCRC,
+     (FULL)0},
+    {"DEFAULTSHELL", "default shell to use", DEFAULTSHELL,
+     (FULL)0},
+    {"DEV_BITS", "device number size in bits", NULL,
+     (FULL)DEV_BITS},
+    {"DISPLAY_DEFAULT", "default digits for float display", NULL,
+     (FULL)DISPLAY_DEFAULT},
+    {"EPSILONPREC_DEFAULT", "2^-EPSILON_DEFAULT <= EPSILON_DEFAULT", NULL,
+     (FULL)EPSILONPREC_DEFAULT},
+    {"EPSILON_DEFAULT", "allowed error for float calculations",
+     EPSILON_DEFAULT, (FULL)0},
+    {"ERRMAX", "default errmax value", NULL,
+     (FULL)ERRMAX},
+    {"E_USERDEF", "base of user defined errors", NULL,
+     (FULL)E_USERDEF},
+    {"E__BASE", "calc errors start above here", NULL,
+     (FULL)E__BASE},
+    {"E__COUNT", "number of calc errors", NULL,
+     (FULL)E__COUNT},
+    {"E__HIGHEST", "highest calc error", NULL,
+     (FULL)E__HIGHEST},
+    {"FALSE", "boolean false", NULL,
+     (FULL)FALSE},
+    {"FILEPOS_BITS", "file position size in bits", NULL,
+     (FULL)FILEPOS_BITS},
+    {"FULL_BITS", "bits in a FULL", NULL,
+     (FULL)FULL_BITS},
+    {"HELPDIR", "location of the help directory", HELPDIR,
+     (FULL)0},
+    {"HIST_BINDING_FILE", "Default binding file", HIST_BINDING_FILE,
+     (FULL)0},
+    {"HIST_SIZE", "Default history size", NULL,
+     (FULL)HIST_SIZE},
+    {"INIT_J", "initial 1st walking a55 table index", NULL,
+     (FULL)INIT_J},
+    {"INIT_K", "initial 2nd walking a55 table index", NULL,
+     (FULL)INIT_K},
+    {"INODE_BITS", "inode number size in bits", NULL,
+     (FULL)INODE_BITS},
+    {"LITTLE_ENDIAN", "Least Significant Byte first symbol",
+     NULL, (FULL)LITTLE_ENDIAN},
+    {"LONG_BITS", "bit length of a long", NULL,
+     (FULL)LONG_BITS},
+    {"MAP_POPCNT", "number of odd primes in pr_map", NULL,
+     (FULL)MAP_POPCNT},
+    {"MAX_CALCRC", "maximum allowed length of $CALCRC", NULL,
+     (FULL)MAX_CALCRC},
+    {"MAXCMD", "max length of command invocation", NULL,
+     (FULL)MAXCMD},
+    {"MAXDIM", "max number of dimensions in matrices", NULL,
+     (FULL)MAXDIM},
+    {"MAXERROR", "max length of error message string", NULL,
+     (FULL)MAXERROR},
+    {"MAXFILES", "max number of opened files", NULL,
+     (FULL)MAXFILES},
+    {"MAXFULL", "largest SFULL value", NULL,
+     (FULL)MAXFULL},
+    {"MAXHALF", "largest SHALF value", NULL,
+     (FULL)MAXHALF},
+    {"MAXLABELS", "max number of user labels in function", NULL,
+     (FULL)MAXLABELS},
+    {"MAXLEN", "longest storage size allowed", NULL,
+     (FULL)MAXLEN},
+    {"MAXLONG", "largest long val", NULL,
+     (FULL)MAXLONG},
+    {"MAXPRINT_DEFAULT", "default number of elements printed", NULL,
+     (FULL)MAXPRINT_DEFAULT},
+    {"MAXREDC", "number of entries in REDC cache", NULL,
+     (FULL)MAXREDC},
+    {"MAXSCANCOUNT", "default max scan errors before an abort", NULL,
+     (FULL)MAXSCANCOUNT},
+    {"MAXSTACK", "max depth of evaluation stack", NULL,
+     (FULL)MAXSTACK},
+    {"MAXSTRING", "max size of string constant", NULL,
+     (FULL)MAXSTRING},
+    {"MAXUFULL", "largest FULL value", NULL,
+     (FULL)MAXUFULL},
+    {"MAXULONG", "largest unsigned long val", NULL,
+     (FULL)MAXULONG},
+    {"MAX_MAP_PRIME", "larest prime in pr_map", NULL,
+     (FULL)MAX_MAP_PRIME},
+    {"MAX_MAP_VAL", "larest bit in pr_map", NULL,
+     (FULL)MAX_MAP_VAL},
+    {"MAX_PFACT_VAL", "max x, for which pfact(x) is a long", NULL,
+     (FULL)MAX_PFACT_VAL},
+    {"MAX_SM_PRIME", "larest 32 bit prime", NULL,
+     (FULL)MAX_SM_PRIME},
+    {"MAX_SM_VAL", "larest 32 bit value", NULL,
+     (FULL)MAX_SM_VAL},
+    {"MUL_ALG2", "default size for alternative multiply", NULL,
+     (FULL)MUL_ALG2},
+    {"NXT_MAP_PRIME", "smallest odd prime not in pr_map", NULL,
+     (FULL)NXT_MAP_PRIME},
+    {"NXT_PFACT_VAL", "next prime for higher pfact values", NULL,
+     (FULL)NXT_PFACT_VAL},
+    {"OFF_T_BITS", "file offset size in bits", NULL,
+     (FULL)OFF_T_BITS},
+    {"PIX_32B", "max pix() value", NULL,
+     (FULL)PIX_32B},
+    {"POW_ALG2", "default size for using REDC for powers", NULL,
+     (FULL)POW_ALG2},
+    {"REDC_ALG2", "default size using alternative REDC alg", NULL,
+     (FULL)REDC_ALG2},
+    {"SBITS", "size of additive or shuffle entry in bits", NULL,
+     (FULL)SBITS},
+    {"SBYTES", "size of additive or shuffle entry in bytes", NULL,
+     (FULL)SBYTES},
+    {"SCNT", "length of additive 55 table in FULLs", NULL,
+     (FULL)SCNT},
+    {"SEEDXORBITS", "low bits of a55 seed devoted to xor", NULL,
+     (FULL)SEEDXORBITS},
+    {"SHALFS", "size of additive or shuffle entry in HALFs", NULL,
+     (FULL)SHALFS},
+    {"SHUFCNT", "size of shuffle table in entries", NULL,
+     (FULL)SHUFCNT},
+    {"SHUFLEN", "length of shuffle table in FULLs", NULL,
+     (FULL)SHUFLEN},
+    {"SHUFMASK", "mask for shuffle table entry selection", NULL,
+     (FULL)SHUFMASK},
+    {"SHUFPOW", "power of 2 size of the shuffle table", NULL,
+     (FULL)SHUFPOW},
+    {"SLEN", "number of FULLs in a shuffle table entry", NULL,
+     (FULL)SLEN},
+    {"SQ_ALG2", "default size for alternative squaring", NULL,
+     (FULL)SQ_ALG2},
+    {"SYMBOLSIZE", "max symbol name size", NULL,
+     (FULL)SYMBOLSIZE},
+    {"TEN_MAX", "10^(2^TEN_MAX): largest base10 conversion const", NULL,
+     (FULL)TEN_MAX},
+    {"TOPFULL", "highest bit in FULL", NULL,
+     (FULL)TOPFULL},
+    {"TOPHALF", "highest bit in a HALF", NULL,
+     (FULL)TOPHALF},
+    {"TOPLONG", "top long bit", NULL,
+     (FULL)TOPLONG},
+    {"TRUE", "boolean true", NULL,
+     (FULL)TRUE},
+    {"USUAL_ELEMENTS", "usual number of elements for objects", NULL,
+     (FULL)USUAL_ELEMENTS},
+    {"REGNUM_MAX", "highest custom register number", NULL,
+     (FULL)CUSTOM_REG_MAX},
 
      /* must be last */
     {NULL, NULL, NULL, (FULL)0}

custom/pmodm127.cal

@@ -17,8 +17,8 @@
  * received a copy with calc; if not, write to Free Software Foundation, Inc.
  * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
  *
- * @(#) $Revision: 30.1 $
- * @(#) $Id: pmodm127.cal,v 30.1 2007/03/16 11:10:04 chongo Exp $
+ * @(#) $Revision: 30.2 $
+ * @(#) $Id: pmodm127.cal,v 30.2 2013/08/11 08:41:38 chongo Exp $
  * @(#) $Source: /usr/local/src/bin/calc/custom/RCS/pmodm127.cal,v $
  *
  * Under source code control:	2004/02/25 14:25:32
@@ -120,7 +120,8 @@ define pmodm127_test2(testcnt, seed)
 		/* compare custom function with its pmod() equivalent */
 		if (pmod(2, m127, q) != custom("pmodm127", q)) {
 			print "ERROR: pmodm127 failed for ", str(q);
-			print "ERROR: ", pmod(2,m127,q), " != ", custom("pmodm127", q);
+			print "ERROR: ", pmod(2,m127,q), " != ",
+			      custom("pmodm127", q);
 			return newerror("pmodm127 failed for " + str(q));
 		}
 	}
@@ -133,5 +134,6 @@ define pmodm127_test2(testcnt, seed)
 }
 
 if ((config("resource_debug") & 3) && !(config("resource_debug") & 8)) {
-    print "DEBUG: use config('resource_debug',", config("resource_debug")|8:") to enable more debugging";
+    print "DEBUG: use config('resource_debug',",
+          config("resource_debug")|8 : ") to enable more debugging";
 }

custom/sysinfo

@@ -11,7 +11,7 @@ TYPES
 
 DESCRIPTION
 
-    This custom function will return the value certain selected #defile
+    This custom function will return the value certain selected #define
     values.  The infoname arg must be a string that matches the given
     #define name.  For conveience, the case infoname does not matter,
     so "baseb" and "BASEB" refer to the same #define value.
@@ -69,8 +69,8 @@ SEE ALSO
 ## received a copy with calc; if not, write to Free Software Foundation, Inc.
 ## 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
 ##
-## @(#) $Revision: 30.1 $
-## @(#) $Id: sysinfo,v 30.1 2007/03/16 11:10:04 chongo Exp $
+## @(#) $Revision: 30.2 $
+## @(#) $Id: sysinfo,v 30.2 2013/08/17 01:57:37 chongo Exp $
 ## @(#) $Source: /usr/local/src/bin/calc/custom/RCS/sysinfo,v $
 ##
 ## Under source code control:	1997/03/09 23:14:40

decl.h

@@ -1,7 +1,7 @@
 /*
  * decl - variable and function declaration macros
  *
- * Copyright (C) 2007  Landon Curt Noll
+ * Copyright (C) 2007,2014  Landon Curt Noll
  *
  * Primary author:  Landon Curt Noll
  *
@@ -19,8 +19,8 @@
  * received a copy with calc; if not, write to Free Software Foundation, Inc.
  * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
  *
- * @(#) $Revision: 30.3 $
- * @(#) $Id: decl.h,v 30.3 2008/04/15 21:17:57 chongo Exp $
+ * @(#) $Revision: 30.6 $
+ * @(#) $Id: decl.h,v 30.6 2014/08/24 21:56:51 chongo Exp $
  * @(#) $Source: /usr/local/src/bin/calc/RCS/decl.h,v $
  *
  * Under source code control:	2007/02/09 05:24:25
@@ -33,8 +33,8 @@
 #include "have_const.h"
 
 
-#if !defined(__DECL_H__)
-#define __DECL_H__
+#if !defined(INCLUDE_DECL_H)
+#define INCLUDE_DECL_H
 
 
 /*
@@ -100,9 +100,10 @@
 
 /* Perform printf-style argument type checking for known compilers */
 #ifdef __GNUC__
-# define PRINTF_FORMAT(fmt_idx, arg_idx) __attribute__ ((format (printf, fmt_idx, arg_idx)))
+# define PRINTF_FORMAT(fmt_idx, arg_idx) __attribute__ \
+         ((format (printf, fmt_idx, arg_idx)))
 #else
 # define PRINTF_FORMAT(fmt_idx, arg_idx)
 #endif
 
-#endif /* !__DECL_H__ */
+#endif /* !INCLUDE_DECL_H */

endian.c

@@ -1,7 +1,7 @@
 /*
  * endian - determine the byte order of a long on your machine
  *
- * Copyright (C) 1999  Landon Curt Noll
+ * Copyright (C) 1999-2013  Landon Curt Noll
  *
  * Calc is open software; you can redistribute it and/or modify it under
  * the terms of the version 2.1 of the GNU Lesser General Public License
@@ -17,8 +17,8 @@
  * received a copy with calc; if not, write to Free Software Foundation, Inc.
  * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
  *
- * @(#) $Revision: 30.3 $
- * @(#) $Id: endian.c,v 30.3 2010/09/02 09:47:46 chongo Exp $
+ * @(#) $Revision: 30.4 $
+ * @(#) $Id: endian.c,v 30.4 2013/05/05 13:57:08 chongo Exp $
  * @(#) $Source: /usr/local/src/bin/calc/RCS/endian.c,v $
  *
  * Under source code control:	1993/11/15 04:32:58
@@ -54,7 +54,9 @@ int
 main(void)
 {
 	/* pointers into the byte order array */
+#if defined(DEBUG) || (!defined(BIG_ENDIAN) && !defined(BIG_ENDIAN))
 	int *intp = (int *)byte;
+#endif
 #if defined(DEBUG)
 	short *shortp = (short *)byte;
 	long *longp = (long *)byte;

file.c

@@ -19,8 +19,8 @@
  * received a copy with calc; if not, write to Free Software Foundation, Inc.
  * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
  *
- * @(#) $Revision: 30.2 $
- * @(#) $Id: file.c,v 30.2 2007/07/05 13:30:38 chongo Exp $
+ * @(#) $Revision: 30.3 $
+ * @(#) $Id: file.c,v 30.3 2013/08/11 08:41:38 chongo Exp $
  * @(#) $Source: /usr/local/src/bin/calc/RCS/file.c,v $
  *
  * Under source code control:	1991/07/20 00:21:56
@@ -367,7 +367,8 @@ openid(char *name, char *mode)
 
 
 /*
- * openpathid - open the specified abse filename, or relative filename along a search path
+ * openpathid - open the specified abse filename, or
+ *	        relative filename along a search path
  *
  * given:
  *	name		file name
@@ -2026,7 +2027,8 @@ showfiles(void)
  *	strptr		pointer to where the new field pointer may be found
  */
 S_FUNC void
-getscanfield(FILE *fp, BOOL skip, unsigned int width, int scannum, char *scanptr, char **strptr)
+getscanfield(FILE *fp, BOOL skip, unsigned int width, int scannum,
+	     char *scanptr, char **strptr)
 {
 	char *str;		/* current string */
 	unsigned long len;	/* current length of string */
@@ -2104,7 +2106,8 @@ getscanfield(FILE *fp, BOOL skip, unsigned int width, int scannum, char *scanptr
  *	strptr		pointer to where the new field pointer may be found
  */
 S_FUNC void
-getscanwhite(FILE *fp, BOOL skip, unsigned int width, int scannum, char **strptr)
+getscanwhite(FILE *fp, BOOL skip, unsigned int width, int scannum,
+	     char **strptr)
 {
 	char *str;		/* current string */
 	unsigned long len;	/* current length of string */

file.h

@@ -1,7 +1,7 @@
 /*
  * file - file I/O routines callable by users
  *
- * Copyright (C) 1999-2007  David I. Bell and Landon Curt Noll
+ * Copyright (C) 1999-2007,2014  David I. Bell and Landon Curt Noll
  *
  * Primary author:  David I. Bell
  *
@@ -19,8 +19,8 @@
  * received a copy with calc; if not, write to Free Software Foundation, Inc.
  * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
  *
- * @(#) $Revision: 30.1 $
- * @(#) $Id: file.h,v 30.1 2007/03/16 11:09:46 chongo Exp $
+ * @(#) $Revision: 30.5 $
+ * @(#) $Id: file.h,v 30.5 2014/08/24 21:56:51 chongo Exp $
  * @(#) $Source: /usr/local/src/bin/calc/RCS/file.h,v $
  *
  * Under source code control:	1996/05/24 05:55:58
@@ -31,8 +31,8 @@
  */
 
 
-#if !defined(__FILE_H__)
-#define __FILE_H__
+#if !defined(INCLUDE_FILE_H)
+#define INCLUDE_FILE_H
 
 
 #if defined(CALC_SRC)	/* if we are building from the calc source tree */
@@ -96,7 +96,8 @@ E_FUNC int fgetposid(FILEID id, FILEPOS *ptr);
 E_FUNC int fsetposid(FILEID id, FILEPOS *ptr);
 E_FUNC int get_open_siz(FILE *fp, ZVALUE *res);
 E_FUNC char* findfname(FILEID);
-E_FUNC FILE *f_pathopen(char *name, char *mode, char *pathlist, char **openpath);
+E_FUNC FILE *f_pathopen(char *name, char *mode, char *pathlist,
+			char **openpath);
 
 
-#endif /* !__FILE_H__ */
+#endif /* !INCLUDE_FILE_H */

func.c

@@ -19,8 +19,8 @@
  * received a copy with calc; if not, write to Free Software Foundation, Inc.
  * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
  *
- * @(#) $Revision: 30.3 $
- * @(#) $Id: func.c,v 30.3 2008/04/15 21:17:57 chongo Exp $
+ * @(#) $Revision: 30.11 $
+ * @(#) $Id: func.c,v 30.11 2014/09/30 00:55:11 chongo Exp $
  * @(#) $Source: /usr/local/src/bin/calc/RCS/func.c,v $
  *
  * Under source code control:	1990/02/15 01:48:15
@@ -1241,10 +1241,11 @@ f_primetest(int count, NUMBER **vals)
 {
 	/* parse args */
 	switch (count) {
-	case 1: return itoq((long) qprimetest(vals[0],
-			    qlink(&_qone_), qlink(&_qone_)));
-	case 2: return itoq((long) qprimetest(vals[0],
-			    vals[1], qlink(&_qone_)));
+	case 1: qlink(&_qone_);
+		qlink(&_qone_);
+		return itoq((long) qprimetest(vals[0], &_qone_, &_qone_));
+	case 2: qlink(&_qone_);
+		return itoq((long) qprimetest(vals[0], vals[1], &_qone_));
 	default: return itoq((long) qprimetest(vals[0], vals[1], vals[2]));
 	}
 }
@@ -1675,7 +1676,7 @@ f_hash(int count, VALUE **vals)
 	result.v_type = V_NUM;
 	result.v_subtype = V_NOSUBTYPE;
 
-	hash = FNV1_32_BASIS;
+	hash = QUICKHASH_BASIS;
 	while (count-- > 0)
 		hash = hashvalue(*vals++, hash);
 	result.v_num = utoq((FULL) hash);
@@ -2062,7 +2063,8 @@ f_ln(int count, VALUE **vals)
 	}
 	switch (vals[0]->v_type) {
 		case V_NUM:
-			if (!qisneg(vals[0]->v_num) && !qiszero(vals[0]->v_num)) {
+			if (!qisneg(vals[0]->v_num) &&
+			    !qiszero(vals[0]->v_num)) {
 				result.v_num = qln(vals[0]->v_num, err);
 				result.v_type = V_NUM;
 				return result;
@@ -2107,7 +2109,8 @@ f_log(int count, VALUE **vals)
 	}
 	switch (vals[0]->v_type) {
 		case V_NUM:
-			if (!qisneg(vals[0]->v_num) && !qiszero(vals[0]->v_num)) {
+			if (!qisneg(vals[0]->v_num) &&
+			    !qiszero(vals[0]->v_num)) {
 				result.v_num = qlog(vals[0]->v_num, err);
 				result.v_type = V_NUM;
 				return result;
@@ -4175,6 +4178,24 @@ f_strcmp(VALUE *v1, VALUE *v2)
 	return result;
 }
 
+S_FUNC VALUE
+f_strcasecmp(VALUE *v1, VALUE *v2)
+{
+	VALUE result;
+	FLAG flag;
+
+	/* initialize VALUE */
+	result.v_subtype = V_NOSUBTYPE;
+
+	if (v1->v_type != V_STR || v2->v_type != V_STR)
+		return error_value(E_STRCASECMP);
+
+	flag = stringcaserel(v1->v_str, v2->v_str);
+
+	result.v_type = V_NUM;
+	result.v_num = itoq((long) flag);
+	return result;
+}
 
 S_FUNC VALUE
 f_strncmp(VALUE *v1, VALUE *v2, VALUE *v3)
@@ -4207,6 +4228,69 @@ f_strncmp(VALUE *v1, VALUE *v2, VALUE *v3)
 	result.v_num = itoq((long) flag);
 	return result;
 }
+S_FUNC VALUE
+f_strncasecmp(VALUE *v1, VALUE *v2, VALUE *v3)
+{
+	long n1, n2, n;
+	FLAG flag;
+	VALUE result;
+
+	/* initialize VALUE */
+	result.v_subtype = V_NOSUBTYPE;
+
+	if (v1->v_type != V_STR || v2->v_type != V_STR ||
+		v3->v_type != V_NUM || qisneg(v3->v_num) ||
+		qisfrac(v3->v_num) || zge31b(v3->v_num->num))
+			return error_value(E_STRNCASECMP);
+	n1 = v1->v_str->s_len;
+	n2 = v2->v_str->s_len;
+	n = qtoi(v3->v_num);
+	if (n < n1)
+		v1->v_str->s_len = n;
+	if (n < n2)
+		v2->v_str->s_len = n;
+
+	flag = stringcaserel(v1->v_str, v2->v_str);
+
+	v1->v_str->s_len = n1;
+	v2->v_str->s_len = n2;
+
+	result.v_type = V_NUM;
+	result.v_num = itoq((long) flag);
+	return result;
+}
+
+S_FUNC VALUE
+f_strtoupper(VALUE *vp)
+{
+        VALUE result;
+
+        /* initialize VALUE */
+        result.v_subtype = V_NOSUBTYPE;
+
+        if (vp->v_type != V_STR)
+                return error_value(E_STRTOUPPER);
+
+        result.v_str  = stringtoupper(vp->v_str);
+        result.v_type = V_STR;
+        return result;
+}
+
+S_FUNC VALUE
+f_strtolower(VALUE *vp)
+{
+        VALUE result;
+
+        /* initialize VALUE */
+        result.v_subtype = V_NOSUBTYPE;
+
+        if (vp->v_type != V_STR)
+                return error_value(E_STRTOLOWER);
+
+        result.v_str  = stringtolower(vp->v_str);
+        result.v_type = V_STR;
+        return result;
+}
 
 
 S_FUNC VALUE
@@ -4394,6 +4478,307 @@ f_ord(VALUE *vp)
 	return result;
 }
 
+S_FUNC VALUE
+f_isupper(VALUE *vp)
+{
+	char c;
+	VALUE result;
+
+	/* initialize VALUE */
+	result.v_subtype = V_NOSUBTYPE;
+
+	switch(vp->v_type) {
+		case V_STR:
+			c = *vp->v_str->s_str;
+			break;
+		case V_OCTET:
+			c = *vp->v_octet;
+			break;
+		default:
+			return error_value(E_ISUPPER);
+	}
+
+	result.v_type = V_NUM;
+	result.v_num = itoq( (isupper( c ))?1l:0l);
+	return result;
+}
+
+S_FUNC VALUE
+f_islower(VALUE *vp)
+{
+	char c;
+	VALUE result;
+
+	/* initialize VALUE */
+	result.v_subtype = V_NOSUBTYPE;
+
+	switch(vp->v_type) {
+		case V_STR:
+			c = *vp->v_str->s_str;
+			break;
+		case V_OCTET:
+			c = *vp->v_octet;
+			break;
+		default:
+			return error_value(E_ISLOWER);
+	}
+
+	result.v_type = V_NUM;
+	result.v_num = itoq( (islower( c ))?1l:0l);
+	return result;
+}
+
+S_FUNC VALUE
+f_isalnum(VALUE *vp)
+{
+	char c;
+	VALUE result;
+
+	/* initialize VALUE */
+	result.v_subtype = V_NOSUBTYPE;
+
+	switch(vp->v_type) {
+		case V_STR:
+			c = *vp->v_str->s_str;
+			break;
+		case V_OCTET:
+			c = *vp->v_octet;
+			break;
+		default:
+			return error_value(E_ISALNUM);
+	}
+
+	result.v_type = V_NUM;
+	result.v_num = itoq( (isalnum( c ))?1l:0l);
+	return result;
+}
+
+S_FUNC VALUE
+f_isalpha(VALUE *vp)
+{
+	char c;
+	VALUE result;
+
+	/* initialize VALUE */
+	result.v_subtype = V_NOSUBTYPE;
+
+	switch(vp->v_type) {
+		case V_STR:
+			c = *vp->v_str->s_str;
+			break;
+		case V_OCTET:
+			c = *vp->v_octet;
+			break;
+		default:
+			return error_value(E_ISALPHA);
+	}
+
+	result.v_type = V_NUM;
+	result.v_num = itoq( (isalpha( c ))?1l:0l);
+	return result;
+}
+
+#if 0
+/* Not in C-standard, marked as obsolete in POSIX.1-2008 */
+S_FUNC VALUE
+f_isascii(VALUE *vp)
+{
+	char c;
+	VALUE result;
+
+	result.v_subtype = V_NOSUBTYPE;
+
+	switch(vp->v_type) {
+		case V_STR:
+			c = *vp->v_str->s_str;
+			break;
+		case V_OCTET:
+			c = *vp->v_octet;
+			break;
+		default:
+			return error_value(E_ISASCII);
+	}
+
+	result.v_type = V_NUM;
+	result.v_num = itoq( (isascii( c ))?1l:0l);
+	return result;
+}
+#endif /* 0 */
+
+S_FUNC VALUE
+f_iscntrl(VALUE *vp)
+{
+	char c;
+	VALUE result;
+
+	/* initialize VALUE */
+	result.v_subtype = V_NOSUBTYPE;
+
+	switch(vp->v_type) {
+		case V_STR:
+			c = *vp->v_str->s_str;
+			break;
+		case V_OCTET:
+			c = *vp->v_octet;
+			break;
+		default:
+			return error_value(E_ISCNTRL);
+	}
+
+	result.v_type = V_NUM;
+	result.v_num = itoq( (iscntrl( c ))?1l:0l);
+	return result;
+}
+
+S_FUNC VALUE
+f_isdigit(VALUE *vp)
+{
+	char c;
+	VALUE result;
+
+	/* initialize VALUE */
+	result.v_subtype = V_NOSUBTYPE;
+
+	switch(vp->v_type) {
+		case V_STR:
+			c = *vp->v_str->s_str;
+			break;
+		case V_OCTET:
+			c = *vp->v_octet;
+			break;
+		default:
+			return error_value(E_ISDIGIT);
+	}
+
+	result.v_type = V_NUM;
+	result.v_num = itoq( (isdigit( c ))?1l:0l);
+	return result;
+}
+
+S_FUNC VALUE
+f_isgraph(VALUE *vp)
+{
+	char c;
+	VALUE result;
+
+	/* initialize VALUE */
+	result.v_subtype = V_NOSUBTYPE;
+
+	switch(vp->v_type) {
+		case V_STR:
+			c = *vp->v_str->s_str;
+			break;
+		case V_OCTET:
+			c = *vp->v_octet;
+			break;
+		default:
+			return error_value(E_ISGRAPH);
+	}
+
+	result.v_type = V_NUM;
+	result.v_num = itoq( (isgraph( c ))?1l:0l);
+	return result;
+}
+
+S_FUNC VALUE
+f_isprint(VALUE *vp)
+{
+	char c;
+	VALUE result;
+
+	/* initialize VALUE */
+	result.v_subtype = V_NOSUBTYPE;
+
+	switch(vp->v_type) {
+		case V_STR:
+			c = *vp->v_str->s_str;
+			break;
+		case V_OCTET:
+			c = *vp->v_octet;
+			break;
+		default:
+			return error_value(E_ISPRINT);
+	}
+
+	result.v_type = V_NUM;
+	result.v_num = itoq( (isprint( c ))?1l:0l);
+	return result;
+}
+
+S_FUNC VALUE
+f_ispunct(VALUE *vp)
+{
+	char c;
+	VALUE result;
+
+	/* initialize VALUE */
+	result.v_subtype = V_NOSUBTYPE;
+
+	switch(vp->v_type) {
+		case V_STR:
+			c = *vp->v_str->s_str;
+			break;
+		case V_OCTET:
+			c = *vp->v_octet;
+			break;
+		default:
+			return error_value(E_ISPUNCT);
+	}
+
+	result.v_type = V_NUM;
+	result.v_num = itoq( (ispunct( c ))?1l:0l);
+	return result;
+}
+
+S_FUNC VALUE
+f_isspace(VALUE *vp)
+{
+	char c;
+	VALUE result;
+
+	/* initialize VALUE */
+	result.v_subtype = V_NOSUBTYPE;
+
+	switch(vp->v_type) {
+		case V_STR:
+			c = *vp->v_str->s_str;
+			break;
+		case V_OCTET:
+			c = *vp->v_octet;
+			break;
+		default:
+			return error_value(E_ISSPACE);
+	}
+
+	result.v_type = V_NUM;
+	result.v_num = itoq( (isspace( c ))?1l:0l);
+	return result;
+}
+
+S_FUNC VALUE
+f_isxdigit(VALUE *vp)
+{
+	char c;
+	VALUE result;
+
+	/* initialize VALUE */
+	result.v_subtype = V_NOSUBTYPE;
+
+	switch(vp->v_type) {
+		case V_STR:
+			c = *vp->v_str->s_str;
+			break;
+		case V_OCTET:
+			c = *vp->v_octet;
+			break;
+		default:
+			return error_value(E_ISXDIGIT);
+	}
+
+	result.v_type = V_NUM;
+	result.v_num = itoq( (isxdigit( c ))?1l:0l);
+	return result;
+}
 
 S_FUNC VALUE
 f_protect(int count, VALUE **vals)
@@ -4842,7 +5227,8 @@ f_rsearch(int count, VALUE **vals)
 			i = listrsearch(v1->v_list, v2, l_start, l_end, &indx);
 			break;
 		case V_ASSOC:
-			i = assocrsearch(v1->v_assoc, v2, l_start, l_end, &indx);
+			i = assocrsearch(v1->v_assoc, v2, l_start,
+				         l_end, &indx);
 			break;
 		case V_STR:
 			i = stringrsearch(v1->v_str, v2->v_str, l_start,
@@ -7863,6 +8249,10 @@ f_blkcpy(int count, VALUE **vals)
 	 */
 	args[0] = vals[1];
 	args[1] = vals[0];
+	null_value.v_type = V_NULL;
+	args[2] = &null_value;
+	args[3] = &null_value;
+	args[4] = &null_value;
 	switch(count) {
 	case 5:
 		args[2] = vals[4];
@@ -7873,14 +8263,10 @@ f_blkcpy(int count, VALUE **vals)
 		count = 5;
 		args[4] = vals[3];
 		args[3] = vals[2];
-		null_value.v_type = V_NULL;
-		args[2] = &null_value;
 		break;
 	case 3:
 		count = 4;
 		args[3] = vals[2];
-		null_value.v_type = V_NULL;
-		args[2] = &null_value;
 		break;
 	}
 
@@ -8130,7 +8516,8 @@ STATIC CONST struct builtin builtins[] = {
 	{"ceil", 1, 1, 0, OP_NOP, 0, f_ceil,
 	 "smallest integer greater than or equal to number"},
 	{"cfappr", 1, 3, 0, OP_NOP, f_cfappr, 0,
-	 "approximate a within accuracy b using\n\t\t\tcontinued fractions"},
+	 "approximate a within accuracy b using\n"
+	 "\t\t\tcontinued fractions"},
 	{"cfsim", 1, 2, 0, OP_NOP, f_cfsim, 0,
 	 "simplify number using continued fractions"},
 	{"char", 1, 1, 0, OP_NOP, 0, f_char,
@@ -8230,7 +8617,8 @@ STATIC CONST struct builtin builtins[] = {
 	{"fgets", 1, 1, 0, OP_NOP, 0, f_fgets,
 	 "read next line from file, newline is kept"},
 	{"fgetstr", 1, 1, 0, OP_NOP, 0, f_fgetstr,
-	 "read next null-terminated string from file, null\n\t\t\tcharacter is kept"},
+	 "read next null-terminated string from file, null\n"
+	 "\t\t\tcharacter is kept"},
 	{"files", 0, 1, 0, OP_NOP, 0, f_files,
 	 "return opened file or max number of opened files"},
 	{"floor", 1, 1, 0, OP_NOP, 0, f_floor,
@@ -8238,7 +8626,8 @@ STATIC CONST struct builtin builtins[] = {
 	{"fopen", 2, 2, 0, OP_NOP, 0, f_fopen,
 	 "open file name a in mode b"},
 	{"fpathopen", 2, 3, 0, OP_NOP, 0, f_fpathopen,
-	 "open file name a in mode b, search for a along\n\t\t\tCALCPATH or path c"},
+	 "open file name a in mode b, search for a along\n"
+	 "\t\t\tCALCPATH or path c"},
 	{"fprintf", 2, IN, 0, OP_NOP, 0, f_fprintf,
 	 "print formatted output to opened file"},
 	{"fputc", 2, 2, 0, OP_NOP, 0, f_fputc,
@@ -8262,9 +8651,11 @@ STATIC CONST struct builtin builtins[] = {
 	{"freopen", 2, 3, 0, OP_NOP, 0, f_freopen,
 	 "reopen a file stream to a named file"},
 	{"fscan", 2, IN, FA, OP_NOP, 0, f_fscan,
-	 "scan a file for assignments to one or\n\t\t\tmore variables"},
+	 "scan a file for assignments to one or\n"
+	 "\t\t\tmore variables"},
 	{"fscanf", 2, IN, FA, OP_NOP, 0, f_fscanf,
-	 "formatted scan of a file for assignment to one\n\t\t\tor more variables"},
+	 "formatted scan of a file for assignment to one\n"
+	 "\t\t\tor more variables"},
 	{"fseek", 2, 3, 0, OP_NOP, 0, f_fseek,
 	 "seek to position b (offset from c) in file a"},
 	{"fsize", 1, 1, 0, OP_NOP, 0, f_fsize,
@@ -8282,7 +8673,8 @@ STATIC CONST struct builtin builtins[] = {
 	{"getenv", 1, 1, 0, OP_NOP, 0, f_getenv,
 	 "value of environment variable (or NULL)"},
 	{"hash", 1, IN, 0, OP_NOP, 0, f_hash,
-	 "return non-negative hash value for one or\n\t\t\tmore values"},
+	 "return non-negative hash value for one or\n"
+	 "\t\t\tmore values"},
 	{"head", 2, 2, 0, OP_NOP, 0, f_head,
 	 "return list of specified number at head of a list"},
 	{"highbit", 1, 1, 0, OP_HIGHBIT, 0, 0,
@@ -8376,7 +8768,8 @@ STATIC CONST struct builtin builtins[] = {
 	{"istype", 2, 2, 0, OP_ISTYPE, 0, 0,
 	 "whether the type of a is same as the type of b"},
 	{"jacobi", 2, 2, 0, OP_NOP, qjacobi, 0,
-	 "-1 => a is not quadratic residue mod b\n\t\t\t1 => b is composite, or a is quad residue of b"},
+	 "-1 => a is not quadratic residue mod b\n"
+	 "\t\t\t1 => b is composite, or a is quad residue of b"},
 	{"join", 1, IN, 0, OP_NOP, 0, f_join,
 	 "join one or more lists into one list"},
 	{"lcm", 1, IN, 0, OP_NOP, f_lcm, 0,
@@ -8449,8 +8842,31 @@ STATIC CONST struct builtin builtins[] = {
 	 "numerator of fraction"},
 	{"ord", 1, 1, 0, OP_NOP, 0, f_ord,
 	 "integer corresponding to character value"},
+	{"isupper", 1, 1, 0, OP_NOP, 0, f_isupper,
+	 "whether character is upper case"},
+	{"islower", 1, 1, 0, OP_NOP, 0, f_islower,
+	 "whether character is lower case"},
+	{"isalnum", 1, 1, 0, OP_NOP, 0, f_isalnum,
+	 "whether character is alpha-numeric"},
+	{"isalpha", 1, 1, 0, OP_NOP, 0, f_isalpha,
+	 "whether character is alphabetic"},
+	{"iscntrl", 1, 1, 0, OP_NOP, 0, f_iscntrl,
+	 "whether character is a control character"},
+	{"isdigit", 1, 1, 0, OP_NOP, 0, f_isdigit,
+	 "whether character is a digit"},
+	{"isgraph", 1, 1, 0, OP_NOP, 0, f_isgraph,
+	 "whether character is a graphical character"},
+	{"isprint", 1, 1, 0, OP_NOP, 0, f_isprint,
+	 "whether character is printable"},
+	{"ispunct", 1, 1, 0, OP_NOP, 0, f_ispunct,
+	 "whether character is a punctuation"},
+	{"isspace", 1, 1, 0, OP_NOP, 0, f_isspace,
+	 "whether character is a space character"},
+	{"isxdigit", 1, 1, 0, OP_NOP, 0, f_isxdigit,
+	 "whether character is a hexadecimal digit"},
 	{"param", 1, 1, 0, OP_ARGVALUE, 0, 0,
-	 "value of parameter n (or parameter count if n\n\t\t\tis zero)"},
+	 "value of parameter n (or parameter count if n\n"
+	 "\t\t\tis zero)"},
 	{"perm", 2, 2, 0, OP_NOP, qperm, 0,
 	 "permutation number a!/(a-b)!"},
 	{"prevcand", 1, 5, 0, OP_NOP, f_prevcand, 0,
@@ -8470,7 +8886,8 @@ STATIC CONST struct builtin builtins[] = {
 	{"polar", 2, 3, 0, OP_NOP, 0, f_polar,
 	 "complex value of polar coordinate (a * exp(b*1i))"},
 	{"poly", 1, IN, 0, OP_NOP, 0, f_poly,
-	 "evaluates a polynomial given its coefficients\n\t\t\tor coefficient-list"},
+	 "evaluates a polynomial given its coefficients\n"
+	 "\t\t\tor coefficient-list"},
 	{"pop", 1, 1, FA, OP_NOP, 0, f_listpop,
 	 "pop value from front of list"},
 	{"popcnt", 1, 2, 0, OP_NOP, f_popcnt, 0,
@@ -8492,7 +8909,8 @@ STATIC CONST struct builtin builtins[] = {
 	{"quo", 2, 3, 0, OP_NOP, 0, f_quo,
 	 "integer quotient of a by b, rounding type c"},
 	{"quomod", 4, 5, FA, OP_NOP, 0, f_quomod,
-	 "set c and d to quotient and remainder of a\n\t\t\tdivided by b"},
+	 "set c and d to quotient and remainder of a\n"
+	 "\t\t\tdivided by b"},
 	{"rand", 0, 2, 0, OP_NOP, f_rand, 0,
 	 "additive 55 random number [0,2^64), [0,a), or [a,b)"},
 	{"randbit", 0, 1, 0, OP_NOP, f_randbit, 0,
@@ -8528,7 +8946,8 @@ STATIC CONST struct builtin builtins[] = {
 	{"round", 1, 3, 0, OP_NOP, 0, f_round,
 	 "round value a to b number of decimal places"},
 	{"rsearch", 2, 4, 0, OP_NOP, 0, f_rsearch,
-	 "reverse search matrix or list for value b\n\t\t\tstarting at index c"},
+	 "reverse search matrix or list for value b\n"
+	 "\t\t\tstarting at index c"},
 	{"runtime", 0, 0, 0, OP_NOP, f_runtime, 0,
 	 "user and kernel mode cpu time in seconds"},
 	{"saveval", 1, 1, 0, OP_SAVEVAL, 0, 0,
@@ -8536,11 +8955,14 @@ STATIC CONST struct builtin builtins[] = {
 	{"scale", 2, 2, 0, OP_SCALE, 0, 0,
 	 "scale value up or down by a power of two"},
 	{"scan", 1, IN, FA, OP_NOP, 0, f_scan,
-	 "scan standard input for assignment to one\n\t\t\tor more variables"},
+	 "scan standard input for assignment to one\n"
+	 "\t\t\tor more variables"},
 	{"scanf", 2, IN, FA, OP_NOP, 0, f_scanf,
-	 "formatted scan of standard input for assignment\n\t\t\tto variables"},
+	 "formatted scan of standard input for assignment\n"
+	 "\t\t\tto variables"},
 	{"search", 2, 4, 0, OP_NOP, 0, f_search,
-	 "search matrix or list for value b starting\n\t\t\tat index c"},
+	 "search matrix or list for value b starting\n"
+	 "\t\t\tat index c"},
 	{"sec", 1, 2, 0, OP_NOP, 0, f_sec,
 	 "sec of a within accuracy b"},
 	{"sech", 1, 2, 0, OP_NOP, 0, f_sech,
@@ -8581,10 +9003,16 @@ STATIC CONST struct builtin builtins[] = {
 	 "assign value to stoponerror flag"},
 	{"str", 1, 1, 0, OP_NOP, 0, f_str,
 	 "simple value converted to string"},
+        {"strtoupper", 1, 1, 0, OP_NOP, 0, f_strtoupper,
+         "Make string upper case"},
+        {"strtolower", 1, 1, 0, OP_NOP, 0, f_strtolower,
+         "Make string lower case"},
 	{"strcat", 1,IN, 0, OP_NOP, 0, f_strcat,
 	 "concatenate strings together"},
 	{"strcmp", 2, 2, 0, OP_NOP, 0, f_strcmp,
 	 "compare two strings"},
+	{"strcasecmp", 2, 2, 0, OP_NOP, 0, f_strcasecmp,
+	 "compare two strings case independent"},
 	{"strcpy", 2, 2, 0, OP_NOP, 0, f_strcpy,
 	 "copy string to string"},
 	{"strerror", 0, 1, 0, OP_NOP, 0, f_strerror,
@@ -8593,6 +9021,8 @@ STATIC CONST struct builtin builtins[] = {
 	 "length of string"},
 	{"strncmp", 3, 3, 0, OP_NOP, 0, f_strncmp,
 	 "compare strings a, b to c characters"},
+	{"strncasecmp", 3, 3, 0, OP_NOP, 0, f_strncasecmp,
+	 "compare strings a, b to c characters case independent"},
 	{"strncpy", 3, 3, 0, OP_NOP, 0, f_strncpy,
 	 "copy up to c characters from string to string"},
 	{"strpos", 2, 2, 0, OP_NOP, 0, f_strpos,
@@ -8775,7 +9205,8 @@ builtinfunc(long index, int argcount, VALUE *stck)
 	vpp = valargs;
 	for (i = 0; i < argcount; i++) {
 		if ((*vpp)->v_type != V_NUM) {
-			math_error("Non-real argument for builtin function %s", bp->b_name);
+			math_error("Non-real argument for builtin function %s",
+				   bp->b_name);
 			/*NOTREACHED*/
 		}
 		numargs[i] = (*vpp)->v_num;

func.h

@@ -1,7 +1,7 @@
 /*
  * func - built-in function interface definitions
  *
- * Copyright (C) 1999-2007  David I. Bell
+ * Copyright (C) 1999-2007,2014  David I. Bell
  *
  * Calc is open software; you can redistribute it and/or modify it under
  * the terms of the version 2.1 of the GNU Lesser General Public License
@@ -17,8 +17,8 @@
  * received a copy with calc; if not, write to Free Software Foundation, Inc.
  * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
  *
- * @(#) $Revision: 30.1 $
- * @(#) $Id: func.h,v 30.1 2007/03/16 11:09:46 chongo Exp $
+ * @(#) $Revision: 30.3 $
+ * @(#) $Id: func.h,v 30.3 2014/08/24 21:56:51 chongo Exp $
  * @(#) $Source: /usr/local/src/bin/calc/RCS/func.h,v $
  *
  * Under source code control:	1990/02/15 01:48:33
@@ -28,8 +28,8 @@
  */
 
 
-#if !defined(__FUNC_H__)
-#define __FUNC_H__
+#if !defined(INCLUDE_FUNC_H)
+#define INCLUDE_FUNC_H
 
 
 #if defined(CALC_SRC)	/* if we are building from the calc source tree */
@@ -110,4 +110,4 @@ E_FUNC void freenumbers(FUNC *);
 E_FUNC void freefunc(FUNC *);
 
 
-#endif /* !__FUNC_H__ */
+#endif /* !INCLUDE_FUNC_H */

hash.c

@@ -17,8 +17,8 @@
  * received a copy with calc; if not, write to Free Software Foundation, Inc.
  * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
  *
- * @(#) $Revision: 30.2 $
- * @(#) $Id: hash.c,v 30.2 2007/07/05 17:37:41 chongo Exp $
+ * @(#) $Revision: 30.3 $
+ * @(#) $Id: hash.c,v 30.3 2013/08/11 08:41:38 chongo Exp $
  * @(#) $Source: /usr/local/src/bin/calc/RCS/hash.c,v $
  *
  * Under source code control:	1995/11/23 05:13:11
@@ -848,7 +848,8 @@ hash_value(int type, void *v, HASH *state)
 			value->v_obj->o_actions->oa_index), state);
 		(state->chkpt)(state);
 
-		for (i=value->v_obj->o_actions->oa_count, vp=value->v_obj->o_table;
+		for (i=value->v_obj->o_actions->oa_count,
+		     vp=value->v_obj->o_table;
 		     i-- > 0;
 		     vp++) {
 

hash.h

@@ -1,7 +1,7 @@
 /*
  * hash - one-way hash routines
  *
- * Copyright (C) 1999-2007  Landon Curt Noll
+ * Copyright (C) 1999-2007,2014  Landon Curt Noll
  *
  * Calc is open software; you can redistribute it and/or modify it under
  * the terms of the version 2.1 of the GNU Lesser General Public License
@@ -17,8 +17,8 @@
  * received a copy with calc; if not, write to Free Software Foundation, Inc.
  * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
  *
- * @(#) $Revision: 30.2 $
- * @(#) $Id: hash.h,v 30.2 2007/07/05 17:37:41 chongo Exp $
+ * @(#) $Revision: 30.4 $
+ * @(#) $Id: hash.h,v 30.4 2014/08/24 21:56:51 chongo Exp $
  * @(#) $Source: /usr/local/src/bin/calc/RCS/hash.h,v $
  *
  * Under source code control:	1995/11/14 23:57:45
@@ -29,8 +29,8 @@
  */
 
 
-#if !defined(__HASH_H__)
-#define __HASH_H__
+#if !defined(INCLUDE_HASH_H)
+#define INCLUDE_HASH_H
 
 
 #if defined(CALC_SRC)	/* if we are building from the calc source tree */
@@ -132,4 +132,4 @@ E_FUNC HASH* hash_usb8(int, USB8*, int, HASH*);
 E_FUNC HASH* hash_value(int, void*, HASH*);
 
 
-#endif /* !__HASH_H__ */
+#endif /* !INCLUDE_HASH_H */

have_posscl.c

@@ -17,8 +17,8 @@
  * received a copy with calc; if not, write to Free Software Foundation, Inc.
  * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
  *
- * @(#) $Revision: 30.2 $
- * @(#) $Id: have_posscl.c,v 30.2 2008/04/15 21:17:57 chongo Exp $
+ * @(#) $Revision: 30.3 $
+ * @(#) $Id: have_posscl.c,v 30.3 2013/08/11 08:41:38 chongo Exp $
  * @(#) $Source: /usr/local/src/bin/calc/RCS/have_posscl.c,v $
  *
  * Under source code control:	1996/07/13 12:57:22
@@ -85,7 +85,8 @@ main(void)
 	printf("#undef HAVE_FILEPOS_SCALAR\n");
 	printf("#define HAVE_FILEPOS_SCALAR /* FILEPOS is a simple value */\n");
 #else
-	printf("#undef HAVE_FILEPOS_SCALAR /* FILEPOS is not a simple value */\n");
+	printf("#undef HAVE_FILEPOS_SCALAR "
+	       "/* FILEPOS is not a simple value */\n");
 #endif
 	/* exit(0); */
 	return 0;

have_stdvs.c

@@ -1,7 +1,7 @@
 /*
  * have_stdvs - try <stdarg.h> to see if it really works with vsprintf()
  *
- * Copyright (C) 1999  Landon Curt Noll
+ * Copyright (C) 1999,2014  Landon Curt Noll
  *
  * Calc is open software; you can redistribute it and/or modify it under
  * the terms of the version 2.1 of the GNU Lesser General Public License
@@ -17,8 +17,8 @@
  * received a copy with calc; if not, write to Free Software Foundation, Inc.
  * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
  *
- * @(#) $Revision: 30.1 $
- * @(#) $Id: have_stdvs.c,v 30.1 2007/03/16 11:09:46 chongo Exp $
+ * @(#) $Revision: 30.4 $
+ * @(#) $Id: have_stdvs.c,v 30.4 2014/08/24 21:57:31 chongo Exp $
  * @(#) $Source: /usr/local/src/bin/calc/RCS/have_stdvs.c,v $
  *
  * Under source code control:	1995/09/09 22:41:10
@@ -56,8 +56,14 @@
 # include <string.h>
 #endif
 
+#include "have_stdlib.h"
+#ifdef HAVE_STDLIB_H
+# include <stdlib.h>
+#endif
+
 #undef VSPRINTF_SIZE_T
-#if defined(FORCE_STDC) || (defined(__STDC__) && __STDC__ != 0) || defined(__cplusplus)
+#if defined(FORCE_STDC) || (defined(__STDC__) && __STDC__ != 0) || \
+    defined(__cplusplus)
 # define VSPRINTF_SIZE_T size_t
 #else
 # define VSPRINTF_SIZE_T long
@@ -86,7 +92,7 @@ try_nthis(char *fmt, VSPRINTF_SIZE_T size, ...)
 {
 	va_list ap;
 
-	va_start(ap, fmt);
+	va_start(ap, size);
 #if !defined(DONT_HAVE_VSPRINTF)
 	vsnprintf(buf, size, fmt, ap);
 #else
@@ -160,8 +166,10 @@ main(void)
 	puts("/*");
 	puts(" * SIMULATE_STDARG");
 	puts(" *");
-	puts(" * WARNING: This type of stdarg makes assumptions about the stack");
-	puts(" *	    that may not be true on your system.  You may want to");
+	puts(" * WARNING: This type of stdarg makes assumptions "
+	     "about the stack");
+	puts(" *	    that may not be true on your system.  "
+	     "You may want to");
 	puts(" *	    define STDARG (if using ANSI C) or VARARGS.");
 	puts(" */");
 	puts("typedef char *va_list;");
@@ -169,7 +177,8 @@ main(void)
 	puts("#define va_end(ap) (void)((ap) = 0)");
 	puts("#define va_arg(ap, type) \\");
 	puts("	      (((type*)((ap) = ((ap) + sizeof(type))))[-1])");
-	puts("#define SIMULATE_STDARG /* use std_arg.h to simulate <stdarg.h> */");
+	puts("#define SIMULATE_STDARG "
+	     "/* use std_arg.h to simulate <stdarg.h> */");
 #else
 	puts("#define STDARG /* use <stdarg.h> */");
 	puts("#include <stdarg.h>");

have_varvs.c

@@ -17,8 +17,8 @@
  * received a copy with calc; if not, write to Free Software Foundation, Inc.
  * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
  *
- * @(#) $Revision: 30.1 $
- * @(#) $Id: have_varvs.c,v 30.1 2007/03/16 11:09:46 chongo Exp $
+ * @(#) $Revision: 30.2 $
+ * @(#) $Id: have_varvs.c,v 30.2 2013/08/11 08:41:38 chongo Exp $
  * @(#) $Source: /usr/local/src/bin/calc/RCS/have_varvs.c,v $
  *
  * Under source code control:	1995/09/09 22:41:10
@@ -53,7 +53,8 @@
 #endif
 
 #undef VSPRINTF_SIZE_T
-#if defined(FORCE_STDC) || (defined(__STDC__) && __STDC__ != 0) || defined(__cplusplus)
+#if defined(FORCE_STDC) || (defined(__STDC__) && __STDC__ != 0) || \
+    defined(__cplusplus)
 # define VSPRINTF_SIZE_T size_t
 #else
 # define VSPRINTF_SIZE_T long

help.c

@@ -1,7 +1,7 @@
 /*
  * help - display help for calc
  *
- * Copyright (C) 1999-2007  Landon Curt Noll
+ * Copyright (C) 1999-2007,2014  Landon Curt Noll
  *
  * Calc is open software; you can redistribute it and/or modify it under
  * the terms of the version 2.1 of the GNU Lesser General Public License
@@ -17,8 +17,8 @@
  * received a copy with calc; if not, write to Free Software Foundation, Inc.
  * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
  *
- * @(#) $Revision: 30.2 $
- * @(#) $Id: help.c,v 30.2 2007/09/21 01:27:27 chongo Exp $
+ * @(#) $Revision: 30.4 $
+ * @(#) $Id: help.c,v 30.4 2014/08/24 21:57:31 chongo Exp $
  * @(#) $Source: /usr/local/src/bin/calc/RCS/help.c,v $
  *
  * Under source code control:	1997/09/14 10:58:30
@@ -274,7 +274,7 @@ givehelp(char *type)
 
 #if defined(CUSTOM)
 	/*
-	 * open the the helpfile (looking in CUSTOMHELPDIR last)
+	 * open the helpfile (looking in CUSTOMHELPDIR last)
 	 */
 	} else {
 

help/Makefile

@@ -2,7 +2,7 @@
 #
 # help - makefile for calc help files
 #
-# Copyright (C) 1999-2006  Landon Curt Noll
+# Copyright (C) 1999-2006,2017  Landon Curt Noll
 #
 # Calc is open software; you can redistribute it and/or modify it under
 # the terms of the version 2.1 of the GNU Lesser General Public License
@@ -18,9 +18,9 @@
 # received a copy with calc; if not, write to Free Software Foundation, Inc.
 # 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
 #
-# @(#) $Revision: 30.5 $
-# @(#) $Id: Makefile,v 30.5 2007/09/21 01:27:27 chongo Exp $
-# @(#) $Source: /usr/local/src/bin/calc/help/RCS/Makefile,v $
+# @(#) $Revision: 30.11 $
+# @(#) $Id: Makefile,v 30.11 2017/05/19 16:09:14 chongo Exp $
+# @(#) $Source: /usr/local/src/bin/calc-RHEL7/help/RCS/Makefile,v $
 #
 # Under source code control:	1991/07/23 06:47:57
 # File existed as early as:	1991
@@ -204,7 +204,7 @@ EXT=
 STD_HELP_FILES_1= intro overview help
 STD_HELP_FILES_2= assoc
 
-BLT_HELP_FILES_3= builtin
+BLT_HELP_FILES_3= builtin man
 
 STD_HELP_FILES_4= command config custom define environment expression
 
@@ -272,43 +272,46 @@ BLT_HELP_FILES= ${BLT_HELP_FILES_3} ${BLT_HELP_FILES_5} \
 		${BLT_HELP_FILES_11} ${BLT_HELP_FILES_13} \
 		${BLT_HELP_FILES_14}
 
+# The detailed list of help files to install
+#
 # This list is prodiced by the detaillist rule when no WARNINGS are detected.
 #
 # Please use:
 #
-#	make detaillist
+#	make detail_help_list
 #
 # to keep this list in nice sorted order and to check that these
 # non-special help files are under RCS control.
 #
-DETAIL_HELP= abs access acos acosh acot acoth acsc acsch address agd append \
-	appr arg argv arrow asec asech asin asinh assign atan atan2 atanh \
-	avg base base2 bernoulli bit blk blkcpy blkfree blocks bround btrunc \
-	calc_tty calclevel calcpath catalan ceil cfappr cfsim char cmdbuf \
-	cmp comb conj cos cosh cot coth count cp csc csch ctime delete den \
-	dereference det digit digits display dp epsilon errcount errmax \
+DETAIL_HELP= abs access acos acosh acot acoth acsc acsch address agd \
+	append appr arg argv arrow asec asech asin asinh assign atan atan2 \
+	atanh avg base base2 bernoulli bit blk blkcpy blkfree blocks bround \
+	btrunc calc_tty calclevel calcpath catalan ceil cfappr cfsim char \
+	cmdbuf cmp comb conj cos cosh cot coth count cp csc csch ctime delete \
+	den dereference det digit digits display dp epsilon errcount errmax \
 	errno error estr euler eval exp fact factor fclose fcnt feof ferror \
-	fflush fgetc fgetfield fgetfile fgetline fgets fgetstr fib files \
-	floor fopen forall fpathopen fprintf fputc fputs fputstr frac free \
+	fflush fgetc fgetfield fgetfile fgetline fgets fgetstr fib files floor \
+	fopen forall fpathopen fprintf fputc fputs fputstr frac free \
 	freebernoulli freeeuler freeglobals freeredc freestatics frem freopen \
 	fscan fscanf fseek fsize ftell gcd gcdrem gd getenv hash head highbit \
 	hmean hnrmod hypot ilog ilog10 ilog2 im indices inputlevel insert int \
-	inverse iroot isassoc isatty isblk isconfig isdefined iserror iseven \
-	isfile ishash isident isint islist ismat ismult isnull isnum isobj \
-	isobjtype isodd isprime isptr isqrt isrand israndom isreal isrel \
-	issimple issq isstr istype jacobi join lcm lcmfact lfactor ln log \
-	lowbit ltol makelist matdim matfill matmax matmin matsum mattrace \
-	mattrans max memsize meq min minv mmin mne mod modify name near \
-	newerror nextcand nextprime norm null num oldvalue ord param perm \
-	pfact pi pix places pmod polar poly pop popcnt pound power prevcand \
-	prevprime printf prompt protect ptest push putenv quo quomod rand \
-	randbit random randombit randperm rcin rcmul rcout rcpow rcsq re \
-	remove reverse rewind rm root round rsearch runtime saveval scale \
-	scan scanf search sec sech seed segment select sgn sha1 sin \
-	sinh size sizeof sleep sort sqrt srand srandom ssq stoponerror \
-	str strcat strcmp strcpy strerror strlen strncmp strncpy strpos \
-	strprintf strscan strscanf substr sum swap system systime tail tan \
-	tanh test time trunc usertime version xor
+	inverse iroot isalnum isalpha isassoc isatty isblk iscntrl isconfig \
+	isdefined isdigit iserror iseven isfile isgraph ishash isident isint \
+	islist islower ismat ismult isnull isnum isobj isobjtype isodd isprime \
+	isprint isptr ispunct isqrt isrand israndom isreal isrel issimple \
+	isspace issq isstr istype isupper isxdigit jacobi join lcm lcmfact \
+	lfactor ln log lowbit ltol makelist matdim matfill matmax matmin \
+	matsum mattrace mattrans max memsize meq min minv mmin mne mod modify \
+	name near newerror nextcand nextprime norm null num oldvalue ord param \
+	perm pfact pi pix places pmod polar poly pop popcnt pound power \
+	prevcand prevprime printf prompt protect ptest push putenv quo quomod \
+	rand randbit random randombit randperm rcin rcmul rcout rcpow rcsq re \
+	remove reverse rewind rm root round rsearch runtime saveval scale scan \
+	scanf search sec sech seed segment select sgn sha1 sin sinh size \
+	sizeof sleep sort sqrt srand srandom ssq stoponerror str strcasecmp \
+	strcat strcmp strcpy strerror strlen strncasecmp strncmp strncpy \
+	strpos strprintf strscan strscanf strtolower strtoupper substr sum \
+	swap system systime tail tan tanh test time trunc usertime version xor
 
 # This list is of files that are clones of DETAIL_HELP files.  They are
 # built from DETAIL_HELP files.
@@ -469,6 +472,18 @@ copy: blkcpy
 	    ${TRUE}; \
 	fi
 
+man: help
+	${RM} -f $@
+	${CP} help $@
+	${CHMOD} 0444 $@
+	-@if [ -z "${Q}" ]; then \
+	    echo ''; \
+	    echo '=-=-= skipping the ${CAT} of help/$@ =-=-='; \
+	    echo ''; \
+	else \
+	    ${TRUE}; \
+	fi
+
 COPYING: ../COPYING
 	${RM} -f $@
 	${CP} ../COPYING $@
@@ -633,11 +648,26 @@ bsdi: all
 	${RM} -f obj
 	${CP} obj.file obj
 
-# These next rule help me form the ${DETAIL_HELP} makefile variables above.
+# These next rule help form the ${DETAIL_HELP} makefile variables above.
 #
-detaillist:
-	${Q} -(echo "xxxxx"; \
-	   for i in ${DETAIL_HELP} /dev/null; do \
+# NOTE: The ${DETAIL_HELP} list does not include special help files.
+#
+detail_help_list:
+	${Q} -(find .  -mindepth 1 -maxdepth 1 -type f \
+	         ! -name Makefile ! -name '.Makefile.swp' \
+		 ! -name archive ! -name assoc ! -name builtin.end \
+		 ! -name builtin.top ! -name command ! -name config \
+		 ! -name contrib ! -name credit ! -name custom ! -name define \
+		 ! -name environment ! -name errorcodes.hdr \
+		 ! -name errorcodes.sed ! -name expression ! -name file \
+		 ! -name funclist.c ! -name funclist.sed ! -name help \
+		 ! -name history ! -name interrupt ! -name intro ! -name list \
+		 ! -name mat ! -name obj.file ! -name operator \
+		 ! -name overview ! -name script ! -name statement \
+		 ! -name todo ! -name types ! -name unexpected \
+		 ! -name variable ! -name wishlist \
+		 -print | \
+	   while read i; do \
 		if [ X"$$i" != X"/dev/null" ]; then \
 		    if [ ! -f RCS/$$i,v ]; then \
 			echo "WARNING: $$i not under RCS control" 1>&2; \
@@ -645,8 +675,10 @@ detaillist:
 			echo $$i; \
 		    fi; \
 		fi; \
-	done | LANG=C ${SORT}) | ${FMT} -70 | \
-	    ${SED} -e '1s/xxxxx/DETAIL_HELP=/' -e '2,$$s/^/	/' \
+	done; \
+	echo '--first_line--') | \
+	    ${SED} -e 's:^\./::' | LANG=C ${SORT} | ${FMT} -70 | \
+	    ${SED} -e '1s/--first_line--/DETAIL_HELP=/' -e '2,$$s/^/	/' \
 		   -e 's/$$/ \\/' -e '$$s/ \\$$//'
 
 ##
@@ -677,6 +709,10 @@ clean:
 clobber: clean
 	${RM} -f ${BLT_HELP_FILES} full .all calc
 	${RM} -f ${SINGULAR_FILES} ${DETAIL_CLONE}
+	-${Q} if [ -e .DS_Store ]; then \
+	    echo ${RM} -rf .DS_Store; \
+	    ${RM} -rf .DS_Store; \
+	fi
 
 # install everything
 #
@@ -684,8 +720,8 @@ clobber: clean
 #
 install: all
 	-${Q} if [ ! -d ${T}${CALC_SHAREDIR} ]; then \
-	    echo ${MKDIR} ${T}${CALC_SHAREDIR}; \
-	    ${MKDIR} ${T}${CALC_SHAREDIR}; \
+	    echo ${MKDIR} -p ${T}${CALC_SHAREDIR}; \
+	    ${MKDIR} -p ${T}${CALC_SHAREDIR}; \
 	    if [ ! -d "${T}${CALC_SHAREDIR}" ]; then \
 		echo ${MKDIR} -p "${T}${CALC_SHAREDIR}"; \
 		${MKDIR} -p "${T}${CALC_SHAREDIR}"; \
@@ -696,8 +732,8 @@ install: all
 	    ${TRUE}; \
 	fi
 	-${Q} if [ ! -d ${T}${HELPDIR} ]; then \
-	    echo ${MKDIR} ${T}${HELPDIR}; \
-	    ${MKDIR} ${T}${HELPDIR}; \
+	    echo ${MKDIR} -p ${T}${HELPDIR}; \
+	    ${MKDIR} -p ${T}${HELPDIR}; \
 	    if [ ! -d "${T}${HELPDIR}" ]; then \
 		echo ${MKDIR} -p "${T}${HELPDIR}"; \
 		${MKDIR} -p "${T}${HELPDIR}"; \

help/archive

@@ -1,4 +1,4 @@
-Where to get the the latest versions of calc
+Where to get the latest versions of calc
 
     Landon Noll maintains the official calc home page at:
 
@@ -31,8 +31,8 @@ chongo <was here> /\../\
 ## received a copy with calc; if not, write to Free Software Foundation, Inc.
 ## 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
 ##
-## @(#) $Revision: 30.1 $
-## @(#) $Id: archive,v 30.1 2007/03/16 11:10:42 chongo Exp $
+## @(#) $Revision: 30.2 $
+## @(#) $Id: archive,v 30.2 2013/08/11 01:08:32 chongo Exp $
 ## @(#) $Source: /usr/local/src/bin/calc/help/RCS/archive,v $
 ##
 ## Under source code control:	1996/06/13 02:51:48

help/blkcpy

@@ -86,7 +86,7 @@ DESCRIPTION
 	; blkcpy(L, M);
 
     For copying from a block B (named or unnamed), the total number of octets
-    available for copying is taken to the the datalen for that block,
+    available for copying is taken to the datalen for that block,
     so that num can be at most size(B) - ssi.
 
     For copying to a block B (named or unnamed), reallocation will be
@@ -206,8 +206,8 @@ SEE ALSO
 ## received a copy with calc; if not, write to Free Software Foundation, Inc.
 ## 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
 ##
-## @(#) $Revision: 30.1 $
-## @(#) $Id: blkcpy,v 30.1 2007/03/16 11:10:42 chongo Exp $
+## @(#) $Revision: 30.2 $
+## @(#) $Id: blkcpy,v 30.2 2013/08/11 01:08:32 chongo Exp $
 ## @(#) $Source: /usr/local/src/bin/calc/help/RCS/blkcpy,v $
 ##
 ## Under source code control:	1997/04/05 14:08:50

help/contrib

@@ -36,45 +36,73 @@ as a diff -c /dev/null your-file patch, or as a uuencoded and
 gziped (or compressed) tar file).
 
 To contribute code, scripts, resource files and/or to help please
-join the low volume calc mailing list by sending EMail to:
+join the low volume calc mailing list calc-tester.  Then send
+your contribution to the calc-tester mailing list.
 
-    calc-contrib at asthe dot com
+To subscribe to the calc-tester mailing list, visit the following URL:
 
-    [[ NOTE: Replace 'at' with @, 'dot' is with . and remove the spaces ]]
-    [[ NOTE: The EMail address uses 'asthe' and the web site URL uses 'isthe' ]]
+	https://www.listbox.com/subscribe/?list_id=239342
 
-Your subject must contain the words:
+    To help determine you are a human and not just a spam bot,
+    you will be required to provide the following additional info:
 
-    calc mailing list subscription
+	Your Name
+	Calc Version
+	Operating System
+	The date 7 days ago
 
-You may have additional words in your subject line.
+    This is a low volume moderated mailing list.
 
-Feel free to follow the name line with additional EMail text as desired.
+    This mailing list replaces calc-tester at asthe dot com list.
 
-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!
+    If you need a human to help you with your mailing list subscription,
+    please send EMail to our special:
+
+	calc-tester-maillist-help at asthe dot com
+
+	NOTE: Remove spaces and replace 'at' with @, 'dot' with .
+
+    address.  To be sure we see your EMail asking for help with your
+    mailing list subscription, please use the following phase in your
+    EMail Subject line:
+
+	calc tester mailing list help
+
+    That phrase in your subject line will help ensure your
+    request will get past our anti-spam filters.  You may have
+    additional words in your subject line.
 
 =-=
 
 Calc bug reports and calc bug fixes should be sent to:
 
-    calc-bugs at asthe dot com
+	calc-bug-report at asthe dot com
 
-    [[ NOTE: Replace 'at' with @, 'dot' is with . and remove the spaces ]]
-    [[ NOTE: The EMail address uses 'asthe' and the web site URL uses 'isthe' ]]
+	NOTE: Remove spaces and replace 'at' with @, 'dot' with .
 
-Your subject must contain the words:
+    This replaces the old calc-bugs at asthe dot com address.
 
-       calc bug report
+    To be sure we see your EMail reporting a calc bug, please use the
+    following phase in your EMail Subject line:
 
-You may have additional words in your subject line.
+	calc bug report
 
-See the BUGS file or try the help command:
+    That phrase in your subject line will help ensure your
+    request will get past our anti-spam filters.  You may have
+    additional words in your subject line.
 
-    help bugs
+    However, you may find it more helpful to simply subscribe
+    to the calc-tester mailing list (see above) and then to
+    send your report to that mailing list as a wider set calc
+    testers may be able to help you.
 
-for details on bug reporting.
+=-=
+
+The calc web site is located at:
+
+    http://www.isthe.com/chongo/tech/comp/calc/
+
+NOTE: The EMail address uses 'asthe', while the web site uses 'isthe'.
 
 =-=
 
@@ -83,7 +111,7 @@ http://www.isthe.com/chongo/
 
 chongo (share and enjoy) /\../\
 
-## Copyright (C) 1999  Landon Curt Noll
+## Copyright (C) 1999,2014  Landon Curt Noll
 ##
 ## Calc is open software; you can redistribute it and/or modify it under
 ## the terms of the version 2.1 of the GNU Lesser General Public License
@@ -99,8 +127,8 @@ chongo (share and enjoy) /\../\
 ## received a copy with calc; if not, write to Free Software Foundation, Inc.
 ## 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
 ##
-## @(#) $Revision: 30.1 $
-## @(#) $Id: contrib,v 30.1 2007/03/16 11:10:42 chongo Exp $
+## @(#) $Revision: 30.2 $
+## @(#) $Id: contrib,v 30.2 2014/10/06 08:39:45 chongo Exp $
 ## @(#) $Source: /usr/local/src/bin/calc/help/RCS/contrib,v $
 ##
 ## Under source code control:	1997/03/09 16:33:22

help/factor

@@ -30,8 +30,8 @@ DESCRIPTION
     If 1 < n < nextprime(limit)^2, then f(n, limit) == 1 <==> n is prime.
     For example, if 1 < n < 121, n is prime if and only if f(n,7) == 1.
 
-    If limit >= 2^32, factor(n, limit) causes an error and factor(n,
-    limit, err) returns the value of err.
+    If limit >= 2^32, factor(n, limit) causes an error and
+    factor(n, limit, err) returns the value of err.
 
 EXAMPLE
     ; print factor(35,4), factor(35,5), factor(35), factor(-35)
@@ -66,8 +66,8 @@ SEE ALSO
 ## received a copy with calc; if not, write to Free Software Foundation, Inc.
 ## 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
 ##
-## @(#) $Revision: 30.1 $
-## @(#) $Id: factor,v 30.1 2007/03/16 11:10:42 chongo Exp $
+## @(#) $Revision: 30.2 $
+## @(#) $Id: factor,v 30.2 2013/08/18 20:15:13 chongo Exp $
 ## @(#) $Source: /usr/local/src/bin/calc/help/RCS/factor,v $
 ##
 ## Under source code control:	1995/12/18 12:34:57

help/fpathopen

@@ -1,5 +1,5 @@
 NAME
-    fpathopen - open an absolute filename, or a relative filename along a search path
+    fpathopen - open an absolute or relative filename along a search path
 
 SYNOPSIS
     fpathopen(filename, mode [,searchpath])
@@ -165,7 +165,8 @@ EXAMPLE
 	    FILE 7 "/home/chongo/tmp/output" (writing, pos 0)
 
     ; badfile = fpathopen("no_such_file", "r")
-    ; if (!isfile(badfile)) print "error #" errno(badfile) : ":" : strerror(badfile);
+    ; if (!isfile(badfile)) print "error #" errno(badfile) : \
+      ":" : strerror(badfile);
     error #2: No such file or directory
 
 LIMITS
@@ -195,8 +196,8 @@ SEE ALSO
 ## received a copy with calc; if not, write to Free Software Foundation, Inc.
 ## 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
 ##
-## @(#) $Revision: 30.1 $
-## @(#) $Id: fpathopen,v 30.1 2007/03/16 11:10:42 chongo Exp $
+## @(#) $Revision: 30.2 $
+## @(#) $Id: fpathopen,v 30.2 2013/08/11 08:41:38 chongo Exp $
 ## @(#) $Source: /usr/local/src/bin/calc/help/RCS/fpathopen,v $
 ##
 ## Under source code control:	2006/05/07 23:56:04

help/hash

@@ -13,11 +13,21 @@ DESCRIPTION
     Returns a hash value for one or more values of arbitrary types.
 
     The calc hash value is based on the core Fowler/Noll/Vo hash
-    known as FNV-1.  The return value, however, cannot be used
+    known as FNV-0.  The return value, however, cannot be used
     as an FNV hash value because calc's internal function also
     takes into account more abstract concepts such as data types.
 
-    See:
+    WARNING: Use of FNV-0 is NOT recommended for general purposes.
+    	     Calc uses FNV-0 for internal objects such as associative
+	     arrays as well as other internal processes.  Calc
+	     maintains the use of FNV-0 for backwards compatibility.
+
+	     You should use the FNV-1a hash instead!!!!!
+
+    Use the fnv() builtin function for general quick, non-cryptographic
+    hash purposes.
+
+    See also:
 
 	 http://www.isthe.com/chongo/tech/comp/fnv/
 
@@ -35,9 +45,9 @@ LINK LIBRARY
     none
 
 SEE ALSO
-    ishash, sha1
+    ishash, fnv, sha1
 
-## Copyright (C) 1999-2007  Landon Curt Noll
+## Copyright (C) 1999-2007,2014  Landon Curt Noll
 ##
 ## Calc is open software; you can redistribute it and/or modify it under
 ## the terms of the version 2.1 of the GNU Lesser General Public License
@@ -53,8 +63,8 @@ SEE ALSO
 ## received a copy with calc; if not, write to Free Software Foundation, Inc.
 ## 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
 ##
-## @(#) $Revision: 30.2 $
-## @(#) $Id: hash,v 30.2 2007/07/05 17:37:41 chongo Exp $
+## @(#) $Revision: 30.3 $
+## @(#) $Id: hash,v 30.3 2014/10/06 08:45:56 chongo Exp $
 ## @(#) $Source: /usr/local/src/bin/calc/help/RCS/hash,v $
 ##
 ## Under source code control:	1996/03/12 23:10:01

help/help

@@ -115,7 +115,15 @@ as the same effect as:
 
     custom("help", "anything");
 
-## Copyright (C) 1999-2007  Landon Curt Noll
+The man command is an alias for the help command.  For example:
+
+    man sin
+
+Any help file that the help command is able to display may be
+displayed by the man command.  The man command may only display
+calc help files.
+
+## Copyright (C) 1999-2007,2017  Landon Curt Noll
 ##
 ## Calc is open software; you can redistribute it and/or modify it under
 ## the terms of the version 2.1 of the GNU Lesser General Public License
@@ -131,9 +139,9 @@ as the same effect as:
 ## received a copy with calc; if not, write to Free Software Foundation, Inc.
 ## 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
 ##
-## @(#) $Revision: 30.1 $
-## @(#) $Id: help,v 30.1 2007/03/16 11:10:42 chongo Exp $
-## @(#) $Source: /usr/local/src/bin/calc/help/RCS/help,v $
+## @(#) $Revision: 30.2 $
+## @(#) $Id: help,v 30.2 2017/05/19 16:09:14 chongo Exp $
+## @(#) $Source: /usr/local/src/bin/calc-RHEL7/help/RCS/help,v $
 ##
 ## Under source code control:	1991/07/21 04:37:20
 ## File existed as early as:	1991

help/intro

@@ -54,7 +54,7 @@ What is calc?
 
     Functions can be used in expressions.  There are a great number of
     pre-defined functions.  For example, the following will calculate
-    the factorial of the value of 'old':
+    the factorial of the value of 'whey':
 
 	    fact(whey)
 
@@ -185,8 +185,8 @@ What is calc?
 ## received a copy with calc; if not, write to Free Software Foundation, Inc.
 ## 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
 ##
-## @(#) $Revision: 30.1 $
-## @(#) $Id: intro,v 30.1 2007/03/16 11:10:42 chongo Exp $
+## @(#) $Revision: 30.2 $
+## @(#) $Id: intro,v 30.2 2016/02/06 08:25:30 chongo Exp $
 ## @(#) $Source: /usr/local/src/bin/calc/help/RCS/intro,v $
 ##
 ## Under source code control:	1991/07/21 04:37:21

help/isalnum

@@ -0,0 +1,54 @@
+NAME
+    isalnum - whether character is alpha-numeric
+
+SYNOPSIS
+    isalnum(c)
+
+TYPES
+    c           string
+
+    return      int
+
+DESCRIPTION
+    Determine if c is alpha-numeric.  This function will return 1 if c is
+    alpha-numeric, 0 otherwise.
+
+EXAMPLE
+    ; print isalnum("A"), isalnum("a"), isalnum("2"), isalnum("\t")
+    1 1 1 0
+
+LIMITS
+    none
+
+LINK LIBRARY
+    none
+
+SEE ALSO
+    islower, isupper, isalpha, iscntrl, isdigit, isgraph, isprint,
+    ispunct, isspace, isxdigit
+
+## Copyright (C) 2013  Christoph Zurnieden
+##
+## Calc is open software; you can redistribute it and/or modify it under
+## the terms of the version 2.1 of the GNU Lesser General Public License
+## as published by the Free Software Foundation.
+##
+## Calc is distributed in the hope that it will be useful, but WITHOUT
+## ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+## or FITNESS FOR A PARTICULAR PURPOSE.	 See the GNU Lesser General
+## Public License for more details.
+##
+## A copy of version 2.1 of the GNU Lesser General Public License is
+## distributed with calc under the filename COPYING-LGPL.  You should have
+## received a copy with calc; if not, write to Free Software Foundation, Inc.
+## 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
+##
+## @(#) $Revision: 30.3 $
+## @(#) $Id: isalnum,v 30.3 2013/09/01 20:00:20 chongo Exp $
+## @(#) $Source: /usr/local/src/bin/calc/help/RCS/isalnum,v $
+##
+## Under source code control:	2013/09/01 18:10:53
+## File existed as early as:	2013
+##
+## chongo <was here> /\oo/\	http://www.isthe.com/chongo/
+## Share and enjoy!  :-)	http://www.isthe.com/chongo/tech/comp/calc/

help/isalpha

@@ -0,0 +1,54 @@
+NAME
+    isalpha - whether character is alphabetic
+
+SYNOPSIS
+    isalpha(c)
+
+TYPES
+    c           string
+
+    return      int
+
+DESCRIPTION
+    Determine if c is alphabetic.  This function will return 1 if c is
+    alphabetic, 0 otherwise.
+
+EXAMPLE
+    ; print isalpha("A"), isalpha("a"), isalpha("1"), isalpha("\t")
+    1 1 0 0
+
+LIMITS
+    none
+
+LINK LIBRARY
+    none
+
+SEE ALSO
+    islower, isupper, isalnum, iscntrl, isdigit, isgraph, isprint,
+    ispunct, isspace, isxdigit
+
+## Copyright (C) 2013  Christoph Zurnieden
+##
+## Calc is open software; you can redistribute it and/or modify it under
+## the terms of the version 2.1 of the GNU Lesser General Public License
+## as published by the Free Software Foundation.
+##
+## Calc is distributed in the hope that it will be useful, but WITHOUT
+## ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+## or FITNESS FOR A PARTICULAR PURPOSE.	 See the GNU Lesser General
+## Public License for more details.
+##
+## A copy of version 2.1 of the GNU Lesser General Public License is
+## distributed with calc under the filename COPYING-LGPL.  You should have
+## received a copy with calc; if not, write to Free Software Foundation, Inc.
+## 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
+##
+## @(#) $Revision: 30.3 $
+## @(#) $Id: isalpha,v 30.3 2013/09/01 20:00:20 chongo Exp $
+## @(#) $Source: /usr/local/src/bin/calc/help/RCS/isalpha,v $
+##
+## Under source code control:	2013/09/01 18:10:53
+## File existed as early as:	2013
+##
+## chongo <was here> /\oo/\	http://www.isthe.com/chongo/
+## Share and enjoy!  :-)	http://www.isthe.com/chongo/tech/comp/calc/

help/iscntrl

@@ -0,0 +1,54 @@
+NAME
+    iscntrl - whether character is a control character
+
+SYNOPSIS
+    iscntrl(c)
+
+TYPES
+    c           string
+
+    return      int
+
+DESCRIPTION
+    Determine if c is a control character.  This function will return 1 if c is
+    alphabetic, 0 otherwise.
+
+EXAMPLE
+    ; print iscntrl("A"), iscntrl("a"), iscntrl("1"), iscntrl("\t")
+    0 0 0 1
+
+LIMITS
+    none
+
+LINK LIBRARY
+    none
+
+SEE ALSO
+    islower, isupper, isalnum, isalpha, isdigit, isgraph, isprint,
+    ispunct, isspace, isxdigit
+
+## Copyright (C) 2013  Christoph Zurnieden
+##
+## Calc is open software; you can redistribute it and/or modify it under
+## the terms of the version 2.1 of the GNU Lesser General Public License
+## as published by the Free Software Foundation.
+##
+## Calc is distributed in the hope that it will be useful, but WITHOUT
+## ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+## or FITNESS FOR A PARTICULAR PURPOSE.	 See the GNU Lesser General
+## Public License for more details.
+##
+## A copy of version 2.1 of the GNU Lesser General Public License is
+## distributed with calc under the filename COPYING-LGPL.  You should have
+## received a copy with calc; if not, write to Free Software Foundation, Inc.
+## 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
+##
+## @(#) $Revision: 30.3 $
+## @(#) $Id: iscntrl,v 30.3 2013/09/01 20:00:20 chongo Exp $
+## @(#) $Source: /usr/local/src/bin/calc/help/RCS/iscntrl,v $
+##
+## Under source code control:	2013/09/01 18:10:53
+## File existed as early as:	2013
+##
+## chongo <was here> /\oo/\	http://www.isthe.com/chongo/
+## Share and enjoy!  :-)	http://www.isthe.com/chongo/tech/comp/calc/

help/isdigit

@@ -0,0 +1,54 @@
+NAME
+    isdigit - whether character is a digit character
+
+SYNOPSIS
+    isdigit(c)
+
+TYPES
+    c           string
+
+    return      int
+
+DESCRIPTION
+    Determine if c is a digit character.  This function will return 1 if c is
+    a digit character, 0 otherwise.
+
+EXAMPLE
+    ; print isdigit("A"), isdigit("a"), isdigit("1"), isdigit("\t")
+    0 0 1 0
+
+LIMITS
+    none
+
+LINK LIBRARY
+    none
+
+SEE ALSO
+    islower, isupper, isalnum, isalpha, iscntrl, isgraph, isprint,
+    ispunct, isspace, isxdigit
+
+## Copyright (C) 2013  Christoph Zurnieden
+##
+## Calc is open software; you can redistribute it and/or modify it under
+## the terms of the version 2.1 of the GNU Lesser General Public License
+## as published by the Free Software Foundation.
+##
+## Calc is distributed in the hope that it will be useful, but WITHOUT
+## ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+## or FITNESS FOR A PARTICULAR PURPOSE.	 See the GNU Lesser General
+## Public License for more details.
+##
+## A copy of version 2.1 of the GNU Lesser General Public License is
+## distributed with calc under the filename COPYING-LGPL.  You should have
+## received a copy with calc; if not, write to Free Software Foundation, Inc.
+## 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
+##
+## @(#) $Revision: 30.3 $
+## @(#) $Id: isdigit,v 30.3 2013/09/01 20:00:20 chongo Exp $
+## @(#) $Source: /usr/local/src/bin/calc/help/RCS/isdigit,v $
+##
+## Under source code control:	2013/09/01 18:10:53
+## File existed as early as:	2013
+##
+## chongo <was here> /\oo/\	http://www.isthe.com/chongo/
+## Share and enjoy!  :-)	http://www.isthe.com/chongo/tech/comp/calc/

help/isgraph

@@ -0,0 +1,54 @@
+NAME
+    isgraph - whether character is a graphical character
+
+SYNOPSIS
+    isgraph(c)
+
+TYPES
+    c           string
+
+    return      int
+
+DESCRIPTION
+    Determine if c is a graphical character.  This function will return 1 if c
+    is graphical, 0 otherwise.
+
+EXAMPLE
+    ; print isgraph("A"), isgraph("a"), isgraph("1"), isgraph(" ")
+    1 1 1 0
+
+LIMITS
+    none
+
+LINK LIBRARY
+    none
+
+SEE ALSO
+    islower, isupper, isalnum, isalpha, iscntrl, isdigit, isprint,
+    ispunct, isspace, isxdigit
+
+## Copyright (C) 2013  Christoph Zurnieden
+##
+## Calc is open software; you can redistribute it and/or modify it under
+## the terms of the version 2.1 of the GNU Lesser General Public License
+## as published by the Free Software Foundation.
+##
+## Calc is distributed in the hope that it will be useful, but WITHOUT
+## ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+## or FITNESS FOR A PARTICULAR PURPOSE.	 See the GNU Lesser General
+## Public License for more details.
+##
+## A copy of version 2.1 of the GNU Lesser General Public License is
+## distributed with calc under the filename COPYING-LGPL.  You should have
+## received a copy with calc; if not, write to Free Software Foundation, Inc.
+## 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
+##
+## @(#) $Revision: 30.3 $
+## @(#) $Id: isgraph,v 30.3 2013/09/01 20:00:20 chongo Exp $
+## @(#) $Source: /usr/local/src/bin/calc/help/RCS/isgraph,v $
+##
+## Under source code control:	2013/09/01 18:10:53
+## File existed as early as:	2013
+##
+## chongo <was here> /\oo/\	http://www.isthe.com/chongo/
+## Share and enjoy!  :-)	http://www.isthe.com/chongo/tech/comp/calc/

help/islower

@@ -0,0 +1,54 @@
+NAME
+    islower - whether character is lower case
+
+SYNOPSIS
+    islower(c)
+
+TYPES
+    c           string
+
+    return      int
+
+DESCRIPTION
+    Determine if c is lower case.  This function will return 1 if c is
+    lower case, 0 otherwise.
+
+EXAMPLE
+    ; print islower("A"), islower("a"), islower("1")
+    0 1 0
+
+LIMITS
+    none
+
+LINK LIBRARY
+    none
+
+SEE ALSO
+    isupper, isalnum, isalpha, iscntrl, isdigit, isgraph, isprint,
+    ispunct, isspace, isxdigit
+
+## Copyright (C) 2013  Christoph Zurnieden
+##
+## Calc is open software; you can redistribute it and/or modify it under
+## the terms of the version 2.1 of the GNU Lesser General Public License
+## as published by the Free Software Foundation.
+##
+## Calc is distributed in the hope that it will be useful, but WITHOUT
+## ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+## or FITNESS FOR A PARTICULAR PURPOSE.	 See the GNU Lesser General
+## Public License for more details.
+##
+## A copy of version 2.1 of the GNU Lesser General Public License is
+## distributed with calc under the filename COPYING-LGPL.  You should have
+## received a copy with calc; if not, write to Free Software Foundation, Inc.
+## 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
+##
+## @(#) $Revision: 30.3 $
+## @(#) $Id: islower,v 30.3 2013/09/01 20:00:20 chongo Exp $
+## @(#) $Source: /usr/local/src/bin/calc/help/RCS/islower,v $
+##
+## Under source code control:	2013/09/01 18:10:53
+## File existed as early as:	2013
+##
+## chongo <was here> /\oo/\	http://www.isthe.com/chongo/
+## Share and enjoy!  :-)	http://www.isthe.com/chongo/tech/comp/calc/

help/isprint

@@ -0,0 +1,54 @@
+NAME
+    isprint - whether character is a printable
+
+SYNOPSIS
+    isprint(c)
+
+TYPES
+    c           string
+
+    return      int
+
+DESCRIPTION
+    Determine if c is printable.  This function will return 1 if c is
+    printable, 0 otherwise.
+
+EXAMPLE
+    ; print isprint("A"), ispint("a"), isprint(" "), isprint("\t")
+    1 1 1 0
+
+LIMITS
+    none
+
+LINK LIBRARY
+    none
+
+SEE ALSO
+    islower, isupper, isalnum, isalpha, iscntrl, isdigit, isgraph,
+    ispunct, isspace, isxdigit
+
+## Copyright (C) 2013  Christoph Zurnieden
+##
+## Calc is open software; you can redistribute it and/or modify it under
+## the terms of the version 2.1 of the GNU Lesser General Public License
+## as published by the Free Software Foundation.
+##
+## Calc is distributed in the hope that it will be useful, but WITHOUT
+## ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+## or FITNESS FOR A PARTICULAR PURPOSE.	 See the GNU Lesser General
+## Public License for more details.
+##
+## A copy of version 2.1 of the GNU Lesser General Public License is
+## distributed with calc under the filename COPYING-LGPL.  You should have
+## received a copy with calc; if not, write to Free Software Foundation, Inc.
+## 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
+##
+## @(#) $Revision: 30.3 $
+## @(#) $Id: isprint,v 30.3 2013/09/01 20:00:20 chongo Exp $
+## @(#) $Source: /usr/local/src/bin/calc/help/RCS/isprint,v $
+##
+## Under source code control:	2013/09/01 18:10:53
+## File existed as early as:	2013
+##
+## chongo <was here> /\oo/\	http://www.isthe.com/chongo/
+## Share and enjoy!  :-)	http://www.isthe.com/chongo/tech/comp/calc/

help/ispunct

@@ -0,0 +1,54 @@
+NAME
+    ispunct - whether character is a punctuation
+
+SYNOPSIS
+    ispunct(c)
+
+TYPES
+    c           string
+
+    return      int
+
+DESCRIPTION
+    Determine if c is a punctuation.  This function will return 1 if c is
+    a punctuation, 0 otherwise.
+
+EXAMPLE
+    ; print ispunct("A"), ispunct("a"), ispunct(" "), ispunct("?")
+    0 0 0 1
+
+LIMITS
+    none
+
+LINK LIBRARY
+    none
+
+SEE ALSO
+    islower, isupper, isalnum, isalpha, iscntrl, isdigit, isgraph, isprint
+    isspace, isxdigit
+
+## Copyright (C) 2013  Christoph Zurnieden
+##
+## Calc is open software; you can redistribute it and/or modify it under
+## the terms of the version 2.1 of the GNU Lesser General Public License
+## as published by the Free Software Foundation.
+##
+## Calc is distributed in the hope that it will be useful, but WITHOUT
+## ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+## or FITNESS FOR A PARTICULAR PURPOSE.	 See the GNU Lesser General
+## Public License for more details.
+##
+## A copy of version 2.1 of the GNU Lesser General Public License is
+## distributed with calc under the filename COPYING-LGPL.  You should have
+## received a copy with calc; if not, write to Free Software Foundation, Inc.
+## 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
+##
+## @(#) $Revision: 30.3 $
+## @(#) $Id: ispunct,v 30.3 2013/09/01 20:00:20 chongo Exp $
+## @(#) $Source: /usr/local/src/bin/calc/help/RCS/ispunct,v $
+##
+## Under source code control:	2013/09/01 18:10:53
+## File existed as early as:	2013
+##
+## chongo <was here> /\oo/\	http://www.isthe.com/chongo/
+## Share and enjoy!  :-)	http://www.isthe.com/chongo/tech/comp/calc/

help/isspace

@@ -0,0 +1,54 @@
+NAME
+    isspace - whether character is a space character
+
+SYNOPSIS
+    isspace(c)
+
+TYPES
+    c           string
+
+    return      int
+
+DESCRIPTION
+    Determine if c is a space character.  This function will return 1 if c is
+    a space character, 0 otherwise.
+
+EXAMPLE
+    ; print isspace("A"), isspace("Kirk"), isspace(" "), isspace("?")
+    0 0 1 0
+
+LIMITS
+    none
+
+LINK LIBRARY
+    none
+
+SEE ALSO
+    islower, isupper, isalnum, isalpha, iscntrl, isdigit, isgraph, isprint
+    ispunct, isxdigit
+
+## Copyright (C) 2013  Christoph Zurnieden
+##
+## Calc is open software; you can redistribute it and/or modify it under
+## the terms of the version 2.1 of the GNU Lesser General Public License
+## as published by the Free Software Foundation.
+##
+## Calc is distributed in the hope that it will be useful, but WITHOUT
+## ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+## or FITNESS FOR A PARTICULAR PURPOSE.	 See the GNU Lesser General
+## Public License for more details.
+##
+## A copy of version 2.1 of the GNU Lesser General Public License is
+## distributed with calc under the filename COPYING-LGPL.  You should have
+## received a copy with calc; if not, write to Free Software Foundation, Inc.
+## 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
+##
+## @(#) $Revision: 30.3 $
+## @(#) $Id: isspace,v 30.3 2013/09/01 20:00:20 chongo Exp $
+## @(#) $Source: /usr/local/src/bin/calc/help/RCS/isspace,v $
+##
+## Under source code control:	2013/09/01 18:10:53
+## File existed as early as:	2013
+##
+## chongo <was here> /\oo/\	http://www.isthe.com/chongo/
+## Share and enjoy!  :-)	http://www.isthe.com/chongo/tech/comp/calc/

help/isupper

@@ -0,0 +1,54 @@
+NAME
+    isupper - whether character is upper case
+
+SYNOPSIS
+    isupper(c)
+
+TYPES
+    c           string
+
+    return      int
+
+DESCRIPTION
+    Determine if c is upper case.  This function will return 1 if c is
+    upper case, 0 otherwise.
+
+EXAMPLE
+    ; print isupper("A"), isupper("a"), isupper("1")
+    1 0 0
+
+LIMITS
+    none
+
+LINK LIBRARY
+    none
+
+SEE ALSO
+    islower, isalnum, isalpha, iscntrl, isdigit, isgraph, isprint,
+    ispunct, isspace, isxdigit
+
+## Copyright (C) 2013  Christoph Zurnieden
+##
+## Calc is open software; you can redistribute it and/or modify it under
+## the terms of the version 2.1 of the GNU Lesser General Public License
+## as published by the Free Software Foundation.
+##
+## Calc is distributed in the hope that it will be useful, but WITHOUT
+## ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+## or FITNESS FOR A PARTICULAR PURPOSE.	 See the GNU Lesser General
+## Public License for more details.
+##
+## A copy of version 2.1 of the GNU Lesser General Public License is
+## distributed with calc under the filename COPYING-LGPL.  You should have
+## received a copy with calc; if not, write to Free Software Foundation, Inc.
+## 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
+##
+## @(#) $Revision: 30.3 $
+## @(#) $Id: isupper,v 30.3 2013/09/01 20:00:20 chongo Exp $
+## @(#) $Source: /usr/local/src/bin/calc/help/RCS/isupper,v $
+##
+## Under source code control:	2013/09/01 18:10:53
+## File existed as early as:	2013
+##
+## chongo <was here> /\oo/\	http://www.isthe.com/chongo/
+## Share and enjoy!  :-)	http://www.isthe.com/chongo/tech/comp/calc/

help/isxdigit

@@ -0,0 +1,54 @@
+NAME
+    isxdigit - whether character a hexadecimal digit
+
+SYNOPSIS
+    isxdigit(c)
+
+TYPES
+    c           string
+
+    return      int
+
+DESCRIPTION
+    Determine if c is a hexadecimal digit.  This function will return 1 if c is
+    a hexadecimal digit, 0 otherwise.
+
+EXAMPLE
+    ; print isxdigit("A"), isxdigit("f"), isxdigit("2"), isxdigit("x")
+    1 1 1 0
+
+LIMITS
+    none
+
+LINK LIBRARY
+    none
+
+SEE ALSO
+    islower, isupper, isalnum, isalpha, iscntrl, isdigit, isgraph, isprint
+    ispunct, isspace
+
+## Copyright (C) 2013  Christoph Zurnieden
+##
+## Calc is open software; you can redistribute it and/or modify it under
+## the terms of the version 2.1 of the GNU Lesser General Public License
+## as published by the Free Software Foundation.
+##
+## Calc is distributed in the hope that it will be useful, but WITHOUT
+## ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+## or FITNESS FOR A PARTICULAR PURPOSE.	 See the GNU Lesser General
+## Public License for more details.
+##
+## A copy of version 2.1 of the GNU Lesser General Public License is
+## distributed with calc under the filename COPYING-LGPL.  You should have
+## received a copy with calc; if not, write to Free Software Foundation, Inc.
+## 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
+##
+## @(#) $Revision: 30.3 $
+## @(#) $Id: isxdigit,v 30.3 2013/09/01 20:00:20 chongo Exp $
+## @(#) $Source: /usr/local/src/bin/calc/help/RCS/isxdigit,v $
+##
+## Under source code control:	2013/09/01 18:10:53
+## File existed as early as:	2013
+##
+## chongo <was here> /\oo/\	http://www.isthe.com/chongo/
+## Share and enjoy!  :-)	http://www.isthe.com/chongo/tech/comp/calc/