in ASin() and Gamma() there is used SetZeroNan() to suppress a warning about an uninitialized variable (for gcc)

git-svn-id: svn://ttmath.org/publicrep/ttmath/trunk@1229 e52654a7-88a9-db11-a3e9-0013d4bc506e

CHANGELOG

@@ -1,3 +1,274 @@
+Version 0.9.4_prerelease  (....):
+    * fixed:   cannot compile on MS Windows when compiling with GCC (Mingw) for 64 bit platform:
+               incorrect size of ttmath::uint and ::sint were used
+               they were 'long' but 'long' is a 32bit type on Windows
+    * fixed:   a crash in Big::Add() (buffer overflow)
+               there was an offset calculated from Int type by using Abs() method and a carry was not checked
+               (if there is a carry we should not make addition -- the argument is too small)
+               this had no impact on calculated values because there was a crash (bus error) immediately
+               following program could crash (64bit):
+               typedef ttmath::Big<1, 8> MyBig;
+               ttmath::Parser<MyBig> parser;
+               parser.Parse("2^(2^63) + 1");
+    * fixed:   similar problems were in methods Big::BitAnd() Big::BitOr() and Big::BitXor() (bitwise operations)
+               and they could return incorrect values
+    * fixed:   in x86_64 asm code (*.asm for Win64) there was in some places esp register used,
+               there should be rsp used instead
+               this affects MS Windows users when they use the asm version (ttmathuint_x86_64_msvc.asm)
+
+
+Version 0.9.3  (2012.11.28):
+    * fixed:   in Big::FromDouble(double value) (only 32 bit version)
+               buffer overflow in referencing to UInt<2>
+               this was used when 'value' was in so called "unnormalized" state
+               (E=0 and F is nonzero)
+               it produced incorrect mantissa (on about 8th decimal digit up)
+    * added:   Parser::InitCGamma()
+               initializing coefficients used when calculating the gamma (or factorial) function
+               this speed up the next calculations
+               you don't have to call this method explicitly
+               these coefficients will be calculated when needed
+    * added:   option 'group_digits' to Conv struct
+               you can set how many digits should be grouped
+    * changed: small optimizations in UInt::ToString() and Big::FromString()
+
+
+Version 0.9.2 (2010.09.23):
+    * fixed:   Big::Add() sometimes incorrectly rounded the last bit from its mantissa
+    * fixed:   Big::BigAnd() Big::BigOr() Big::BigXor() should have set NaN
+               when the argument was negative (they only returned 2)
+    * fixed:   recurrence calling in Big::FromString(const std::string &, uint, const wchar_t **, bool *)
+               it should have the signature: Big::FromString(const std::string &, uint, const char **, bool *)               
+    * fixed:   Big::ToString method
+               in some cases when in the output string the exponent should be equal zero
+               the method changes the exponent to one so the last digit from the mantissa
+               was lost
+    * fixed:   Big::ToDouble(double &) set always +INF (infinity)
+               when the value was too large (even for negative values)
+               (it should set -INF in such a case)
+    * added:   some missing operators
+               UInt::operator~()    /* bitwise neg  */
+               UInt::operator&()    /* bitwise and  */
+               UInt::operator&=()
+               UInt::operator|()    /* bitwise or   */
+               UInt::operator|=()
+               UInt::operator^()    /* bitwise xor  */
+               UInt::operator^=()
+               Big::operator&()
+               Big::operator&=()
+               Big::operator|()
+               Big::operator|=()
+               Big::operator^()
+               Big::operator^=()
+               for Big<> we do not define bitwise neg
+               Big::operator++()
+               Big::operator++(int)
+               Big::operator--()
+               Big::operator--(int)
+    * added:   macro TTMATH_DONT_USE_WCHAR
+               if defined then the library does not use wide characters
+               (wchar_t, std::wstring, ...) this is a workaround for some compilers
+    * added:   bool UInt::IsOnlyTheHighestBitSet()
+               bool UInt::IsOnlyTheLowestBitSet()
+               returning true if only the highest/lowest bit is set
+    * added:   uint Int::MulInt(sint ss2)       
+    * added:   void UInt::Swap(UInt<value_size> & ss2)       
+               void Big::Swap(UInt<value_size> & ss2)
+               method for swapping this for an argument
+    * added:   macro TTMATH_BIG_DEFAULT_CLEAR
+               when defined the default constructor from Big<> clears its mantissa and exponent
+               Big<1, 2> var;
+               var.mantissa and var.exponent will be set to zero
+               (but var has the NaN flag set too - it is not zero value, this is mainly for debug purposes)
+    * added:   only on 32bit platforms:
+               uint UInt::FromUInt(uint64_t n)
+               uint Int::FromInt(int64_t n)
+               void Big::FromUInt(uint64_t n)
+               void Big::FromInt(int64_t n)        
+               and appropriate constructors and operators
+    * added:   TTMATH_FORCEASM macro
+               asm version of the library is available by default only for:
+               x86 and amd64 platforms and for Microsoft Visual and GCC compilers,
+               but you can force using asm version (the same asm as for Microsoft Visual)
+               by defining TTMATH_FORCEASM macro
+               you have to be sure that your compiler accept such an asm format
+    * added:   some missing methods for converting
+               for UInt<>, Int<> and Big<> classes:
+               uint ToUInt()
+               sint ToInt()
+               ToUInt(uint32_t &)
+               ToInt(uint32_t &)
+               ToInt(int32_t &)
+               ToUInt(uint64_t &)
+               ToInt(uint64_t &)
+               ToInt(int64_t &)
+               FromUInt(uint32_t &)
+               FromInt(uint32_t &)
+               FromInt(int32_t &)
+               FromUInt(uint64_t &)
+               FromInt(uint64_t &)
+               FromInt(int64_t &)
+               and appropriate constructors and operators
+    * added:   double Big::ToDouble() /there was only Big::ToDouble(double &) /
+               uint Big::ToFloat(float &)
+               float Big::ToFloat()
+    * changed: now asm version is available only on x86 and amd64
+               (and only for GCC and MS VC compilers)
+    * removed: macro TTMATH_RELEASE (now the 'release' version is default)
+               for debug version define TTMATH_DEBUG macro
+               TTMATH_DEBUG is also automatically defined when DEBUG or _DEBUG is set
+    * removed: macro TTMATH_REFERENCE_ASSERT from all methods in public interface
+
+
+Version 0.9.1 (2010.02.07):
+    * fixed:   the parser didn't use characters for changing the base (# and &)
+               those characters were skipped
+               (this bug was introduced in 0.9.0)
+    * fixed:   added in the parser: operator's associativity
+               operator ^ (powering) is right-associative:
+               sample: 2^3^4 is equal 2^(3^4) and it is: 2.41e+24
+               previously was: 2^3^4 = (2^3)^4 = 4096
+    * fixed:   in Big::ToString_CreateNewMantissaAndExponent() changed the formula:
+               new_exp_ = [log base (2^exponent)] + 1
+               now the part '+ 1' is only made when the logarithm is positive and with fraction
+               if the value is negative we can only skip the fraction, previously
+               we lost some last digits from the new mantissa
+               Consider this binary value (32 bit mantissa):
+               (bin)1.0000000000000000000000000000011
+               previously ToString() gave 1, now we have: 1.000000001
+    * changed: in Big::ToString() the base rounding is made only if the result value
+               would not be an integer, e.g. if the value is 1.999999999999 then
+               the base rounding will not be done - because as the result would be 2
+    * added:   IEEE 754 half-to-even rounding (bankers' rounding) to the following
+               floating point algorithms: Big::Add, Big::Sub, Big::Mul, Big::Div
+    * added:   static sint UInt<value_size>::FindLowestBitInWord(uint x)
+               this method is looking for the lowest set bit in a word
+    * added:   UInt::FindLowestBit(uint & table_id, uint & index)
+               this method is looking for the lowest set bit
+
+               
+Version 0.9.0 (2009.11.25):
+    * added:   support for wide characters (wchar_t, std::wstring)
+    * added:   Big::IsInteger()
+               returns true if the value is integer (without fraction)
+               (NaN flag is not checked)
+    * added:   global Gamma() function
+    * added:   gamma() function to the parser
+    * added:   CGamma<ValueType> class
+               is used with Gamma() and Factorial() in multithreaded environment
+    * added:   multithread support for Big<> class
+               you should compile with TTMATH_MULTITHREADS
+               and use TTMATH_MULTITHREADS_HELPER macro somewhere in your *.cpp file
+    * added:   x86_64 asm code for Microsoft Visual compiler
+               file: ttmathuint_x86_64_msvc.asm
+               (this file should be compiled first because MS VC doesn't support inline assembler in x86_64 mode) 
+    * added:   flag TTMATH_BIG_ZERO to Big<> class
+               if this flag is set then there is a value zero
+               Big::IsZero() is faster now
+    * added:   Big::ClearInfoBit(unsigned char)
+               Big::SetInfoBit(unsigned char)
+               Big::IsInfoBit(unsigned char)
+               some methods for manipulating the info flags
+    * added:   macro: TTMATH_BITS(min_bits)
+		       which returns the number of machine words 
+		       capable to hold min_bits bits
+    * added:   bool Parser::Calculated()
+               this method returns true is something was calculated
+               (at least one mathematical operator was used or a function or variable)
+    * added:   to the parser: operator percentage
+               e.g. 1000-50%=1000-(1000*0,5)=500
+    * added:   struct: Conv
+               consists of some parameters used
+               in ToString() and FromString()
+    * added:   Big::ToString() can group digits
+               e.g. 1234567 -> 1`234`567
+    * added:   Parser::SetGroup(int g)
+               Parser::SetComma(int c, int c2 = 0)
+               Parser::SetParamSep(int s)
+    * added:   std::string  UInt::ToString(uint b = 10)
+               std::wstring UInt::ToWString(uint b = 10)
+               std::string  Int::ToString(uint b = 10)
+               std::wstring Int::ToWString(uint b = 10)
+               uint         Big::ToString(std::string & result, const Conv & conv)
+               uint         Big::ToString(std::wstring & result, const Conv & conv)
+               std::string  Big::ToString(const Conv & conv)
+               std::string  Big::ToString()
+               std::wstring Big::ToWString(const Conv & conv)
+               std::wstring Big::ToWString()
+    * added:   uint FromString(const char * source, const Conv & conv, const char **, bool *)
+               uint FromString(const wchar_t * source, const Conv & conv, const wchar_t **, bool *)
+               uint FromString(const std::string & string, const Conv & conv, const wchar_t **, bool *)
+               uint FromString(const std::wstring & string, const Conv & conv, const wchar_t **, bool *)
+    * added:   UInt::Sqrt() - a new algorithm for calculating the square root
+    * added:   to the parser: function frac() - returns a value without the integer part
+               (only fraction remains)
+    * added:   Int::DivInt(sint divisor, sint * remainder)
+    * added:   const char * UInt::LibTypeStr()
+               const char * Big::LibTypeStr()
+               LibTypeCode UInt::LibType()
+               LibTypeCode Big::LibType()
+               returning a string/enum represents the currect type of the library
+               we have following types:
+                   asm_vc_32   - with asm code designed for Microsoft Visual C++ (32 bits)
+                   asm_gcc_32  - with asm code designed for GCC (32 bits)
+                   asm_vc_64   - with asm for VC (64 bit)
+                   asm_gcc_64  - with asm for GCC (64 bit)
+                   no_asm_32   - pure C++ version (32 bit) - without any asm code
+                   no_asm_64   - pure C++ version (64 bit) - without any asm code
+    * added:   UInt::operator>>(int)
+               UInt::operator>>=(int)
+               UInt::operator<<(int)
+               UInt::operator<<=(int)
+    * changed: Factorial() is using the Gamma() function now
+    * changed: Big::Div(ss2)
+               Big::Mod(ss2)
+               they return 2 when ss2 is zero
+               previously returned 1
+    * changed: algorithms in Big::Sqrt() and ttmath::Root(x ; n)
+               they were not too much accurate for some integers
+               e.g. Root(16;4) returned a value very closed to 2 (not exactly 2)
+    * changed: added specializations to Big::ToString() when the base is equal 4, 8 or 16
+               the previous version was not accurate on some last digits (after the comma operator)
+               consider this binary value (32 bit mantissa):
+               base 2: 1.1111 1111 1111 1111 1111 1111 1110 101
+               previous ToString() gave:
+                  base 4:  1.33333333333332
+                  base 8:  1.777777777
+                  base 16: 1.FFFFFF
+               now we have:
+                  base 4:  1.3333333333333222
+                  base 8:  1.77777777724
+                  base 16: 1.FFFFFFEA
+    * changed: in Big::ToString() some additional rounding (base_round) is now made only 
+               when the value is not an integer
+    * changed: another compilers than MS VC or GCC by default use no asm version (TTMATH_NOASM)
+    * removed: Parser<>::SetFactorialMax() method
+               the factorial() is such a fast now that we don't need the method longer
+    * removed: ErrorCode::err_too_big_factorial
+    * removed: macros: TTMATH_COMMA_CHARACTER_1 and TTMATH_COMMA_CHARACTER_2
+               the comma characters we have in Conv struct now
+               
+
+Version 0.8.6 (2009.10.25):
+    * fixed:   UInt::SetBitInWord(uint & value, uint bit) set 1 if the bit was
+               equal 1 (should be set 2)
+               this affected only no-asm parts - when macro TTMATH_NOASM was defined
+    * fixed:   UInt<value_size>::MulInt(uint ss2)
+               there was a buffer overflow when value_size was equal 1
+    * fixed:   UInt::AddVector() and UInt::SubVector() didn't want to compile
+               when macro TTMATH_NOASM was defined
+    * fixed:   Big::operator>> didn't correctly recognize values in scientific mode (with 'e' character)
+    * fixed:   Int::FromString(const tt_string & s, uint b = 10)
+               didn't use 'b' (always was '10')
+    * fixed:   buffer overflow in Big::ToInt(Int<int_size> & result)
+    * fixed:   powering algorithm in:
+               UInt::Pow(UInt<value_size> pow)
+               Big::Pow(UInt<pow_size> pow)
+               Big::PowUInt(Big<exp, man> pow)
+               when 'pow' was sufficient large the algorithm returned carry
+               but the result could have been calculated correctly
+
+
 Version 0.8.5 (2009.06.16):
     * fixed:   Big::Mod(x) didn't correctly return a carry
                and the result was sometimes very big (even greater than x)

CMakeLists.txt

@@ -0,0 +1,6 @@
+# CMake configuration for ttmath
+cmake_minimum_required (VERSION 3.0)
+project(ttmath)
+enable_testing()
+include_directories(${ttmath_SOURCE_DIR})
+add_subdirectory(samples)

COPYRIGHT

@@ -1,4 +1,4 @@
-Copyright (c) 2006-2009, Tomasz Sowa
+Copyright (c) 2006-2017, Tomasz Sowa
 All rights reserved.
 
 Redistribution and use in source and binary forms, with or without

README

@@ -1,26 +1,23 @@
-TTMath - a bignum library for C++
+A bignum library for C++
 
 TTMath is a small library which allows one to perform arithmetic operations
-with big unsigned integer, big signed integer and big floating point
-numbers. It provides standard mathematical operations like adding,
-subtracting, multiplying, dividing etc. With the library also goes
-a mathematical parser which helps you solving input formulas read directly
-from a user.
+with big unsigned integer, big signed integer and big floating point numbers.
+It provides standard mathematical operations like adding, subtracting,
+multiplying, dividing. With the library also goes a mathematical parser to
+help you solving mathematical expressions.
 
-TTMath is developed under the BSD licence which means that it is free
-for both personal and commercial use.
-
-The library has some technical limitations:
- * there are only two platforms that are supported: x86 and x86_64,
- * you can use this library only with the C++ programming language.
+TTMath is developed under the BSD licence which means that it is free for
+both personal and commercial use.
 
 The main goal of the library is to allow one to use big values in the same
-way as the standard types like int, float, etc. It does not need to be
-compiled first because the whole library is written as the C++ templates.
-This means only C++ developers can use this library and one thing they have
-to do is to use 'include' directive of the preprocessor. How big the
-values can be is set directly in the source code by the programmer.
+way as the standard types like int or float. It does not need to be compiled
+first because the whole library is written as the C++ templates. This means
+only C++ developers can use this library and one thing they have to do is
+to use 'include' directive of the preprocessor. How big the values can be
+is set at compile time.
 
 Author: Tomasz Sowa <t.sowa@ttmath.org>
-Project pages: http://www.ttmath.org
-http://sourceforge.net/projects/ttmath
+WWW:    http://www.ttmath.org
+
+Contributors:
+Christian Kaiser <chk@online.de>

constgen/Makefile

@@ -1,6 +1,6 @@
 o      = main.o
-CC     = g++
-CFLAGS = -s -O2 -DTTMATH_CONSTANTSGENERATOR
+CC     = clang++
+CFLAGS = -O2 -DTTMATH_CONSTANTSGENERATOR
 name   = gen
 
 
@@ -13,7 +13,7 @@ name   = gen
 all: $(name)
 
 $(name): $(o)
-	$(CC) -o $(name) $(CFLAGS) $(o)
+	$(CC) -o $(name) -s $(CFLAGS) $(o)
 
 
 

constgen/main.cpp

@@ -1,7 +1,7 @@
 /*
  * This file is a part of TTMath Bignum Library
- * and is distributed under the (new) BSD licence.
- * Author: Tomasz Sowa <t.sowa@slimaczek.pl>
+ * and is distributed under the 3-Clause BSD Licence.
+ * Author: Tomasz Sowa <t.sowa@ttmath.org>
  */
 
 /* 

makerelease.sh

@@ -0,0 +1,137 @@
+#!/usr/bin/env bash
+
+a=""
+b=""
+c=""
+p=""
+d=""
+doxygen=""
+
+# reading until not empty
+
+while [ -z $a ]
+do
+	echo -n "Major: " ; read a
+done
+
+
+while [ -z $b ]
+do
+	echo -n "Minor: " ; read b;
+done
+
+
+while [ -z $c ]
+do
+	echo -n "Revision: " ; read c;
+done
+
+
+while [ -z $p ]
+do
+	echo -n "Prerelease? (y/n): " ; read p;
+done
+
+
+while [ -z $d ]
+do
+	echo -n "Add date? (y/n): " ; read d;
+done
+
+
+while [ -z $doxygen ]
+do
+	echo -n "Clean make and add to the package doxygen doc? (y/n): " ; read doxygen;
+done
+
+
+mkdir -p releases
+package_dir_name="ttmath-$a.$b.$c"
+datestr=""
+
+
+if [ $p = "y" -o $p = "Y" ]
+then
+	package_dir_name=${package_dir_name}.prerelease
+fi
+
+package_dir_name=${package_dir_name}-src
+
+if [ $d = "y" -o $d = "Y" ]
+then
+	datestr=`/bin/date "+%G.%m.%d"`;
+	package_dir_name=${package_dir_name}-$datestr
+fi
+
+
+
+
+dir="releases/"${package_dir_name}
+package=${package_dir_name}.tar.gz
+
+
+if [ -d $dir ]
+then
+	echo "Directory $dir exists! (exiting)";
+	exit 1;
+fi
+
+if [ -f "releases/"${package} ] ; then
+	echo "File releases/${package} exists! (exiting)"
+	exit 1;
+fi
+
+
+mkdir $dir
+
+
+
+if [ $doxygen = "y" -o $doxygen = "Y" ]
+then
+
+	echo "------------------------------------------------------"
+	echo "creating doxygen doc"
+	echo "------------------------------------------------------"
+
+	rm -rf doc/doxygen
+	doxygen doxygen.cfg
+
+	cp -r doc $dir/
+fi
+
+
+echo "------------------------------------------------------"
+echo "make clean in samples"
+echo "------------------------------------------------------"
+
+make -C samples clean 
+make -C constgen clean
+
+
+echo "------------------------------------------------------"
+echo "making source package"
+echo "------------------------------------------------------"
+
+mkdir $dir/ttmath
+mkdir $dir/samples
+mkdir $dir/res
+
+cp ttmath/*  $dir/ttmath/
+
+cp samples/* $dir/samples/
+# cmake is not ready yet (cmake will generate Makefile which overwrites our own one)
+rm $dir/samples/CMakeLists.txt
+
+cp COPYRIGHT         $dir/
+cp README            $dir/
+cp CHANGELOG         $dir/
+
+cp res/ttmath_logo.svg	$dir/res/
+
+cd releases
+tar -czf $package ${package_dir_name}
+
+echo "the package has been created to:" releases/${package}
+
+exit 0
+

res/ttmath_logo.svg

@@ -0,0 +1,96 @@
+<?xml version="1.0" encoding="UTF-8" standalone="no"?>
+<!-- Created with Inkscape (http://www.inkscape.org/) -->
+
+<svg
+   xmlns:dc="http://purl.org/dc/elements/1.1/"
+   xmlns:cc="http://creativecommons.org/ns#"
+   xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
+   xmlns:svg="http://www.w3.org/2000/svg"
+   xmlns="http://www.w3.org/2000/svg"
+   xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd"
+   xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape"
+   width="38.084499"
+   height="38.083"
+   id="svg2"
+   version="1.1"
+   inkscape:version="0.48.4 r9939"
+   sodipodi:docname="ttmath_logo.svg">
+  <defs
+     id="defs4" />
+  <sodipodi:namedview
+     id="base"
+     pagecolor="#ffffff"
+     bordercolor="#666666"
+     borderopacity="1.0"
+     inkscape:pageopacity="0.0"
+     inkscape:pageshadow="2"
+     inkscape:zoom="11.2"
+     inkscape:cx="24.511255"
+     inkscape:cy="17.621314"
+     inkscape:document-units="px"
+     inkscape:current-layer="layer1"
+     showgrid="false"
+     inkscape:window-width="1440"
+     inkscape:window-height="830"
+     inkscape:window-x="0"
+     inkscape:window-y="25"
+     inkscape:window-maximized="1"
+     fit-margin-top="0"
+     fit-margin-left="0"
+     fit-margin-right="0"
+     fit-margin-bottom="0" />
+  <metadata
+     id="metadata7">
+    <rdf:RDF>
+      <cc:Work
+         rdf:about="">
+        <dc:format>image/svg+xml</dc:format>
+        <dc:type
+           rdf:resource="http://purl.org/dc/dcmitype/StillImage" />
+        <dc:title />
+      </cc:Work>
+    </rdf:RDF>
+  </metadata>
+  <g
+     inkscape:label="Layer 1"
+     inkscape:groupmode="layer"
+     id="layer1"
+     transform="translate(-350.08291,-424.26987)">
+    <path
+       style="font-size:medium;font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;text-indent:0;text-align:start;text-decoration:none;line-height:normal;letter-spacing:normal;word-spacing:normal;text-transform:none;direction:ltr;block-progression:tb;writing-mode:lr-tb;text-anchor:start;baseline-shift:baseline;color:#000000;fill:#231f20;fill-opacity:1;stroke:none;stroke-width:1.625;marker:none;visibility:visible;display:inline;overflow:visible;enable-background:accumulate;font-family:Sans;-inkscape-font-specification:Sans"
+       d="m 369.125,424.28125 c -10.50699,0 -19.03125,8.52426 -19.03125,19.03125 0,10.50603 8.52426,19.03125 19.03125,19.03125 10.50746,0 19.03125,-8.52522 19.03125,-19.03125 0,-10.50699 -8.52379,-19.03125 -19.03125,-19.03125 z"
+       id="path3773" />
+    <path
+       style="font-size:medium;font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;text-indent:0;text-align:start;text-decoration:none;line-height:normal;letter-spacing:normal;word-spacing:normal;text-transform:none;direction:ltr;block-progression:tb;writing-mode:lr-tb;text-anchor:start;baseline-shift:baseline;color:#000000;fill:#ffaa35;fill-opacity:1;stroke:none;stroke-width:1.62500000000000000;marker:none;visibility:visible;display:inline;overflow:visible;enable-background:accumulate;font-family:Sans;-inkscape-font-specification:Sans"
+       d="m 369.125,425.90625 c 9.62928,0 17.40625,7.77748 17.40625,17.40625 0,9.62773 -7.77697,17.40625 -17.40625,17.40625 -9.62877,0 -17.40625,-7.77852 -17.40625,-17.40625 0,-9.62877 7.77748,-17.40625 17.40625,-17.40625 z"
+       id="path2999-2" />
+    <path
+       style="font-size:medium;font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;text-indent:0;text-align:start;text-decoration:none;line-height:normal;letter-spacing:normal;word-spacing:normal;text-transform:none;direction:ltr;block-progression:tb;writing-mode:lr-tb;text-anchor:start;baseline-shift:baseline;color:#000000;fill:#231f20;fill-opacity:1;stroke:none;stroke-width:1.5;marker:none;visibility:visible;display:inline;overflow:visible;enable-background:accumulate;font-family:Sans;-inkscape-font-specification:Sans"
+       d="m 368.1875,432.65625 -0.1875,0.5 -6.34375,16.71875 -1.40625,-4 -0.5,-1.40625 -0.84375,1.25 -2.0625,3.125 1.25,0.8125 1.25,-1.875 1.5625,4.53125 0.6875,1.9375 0.71875,-1.9375 6.90625,-18.15625 8.65625,0 0,-1.5 -9.15625,0 -0.53125,0 z"
+       id="path3003-3"
+       inkscape:connector-curvature="0" />
+    <path
+       style="font-size:medium;font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;text-indent:0;text-align:start;text-decoration:none;line-height:normal;letter-spacing:normal;word-spacing:normal;text-transform:none;direction:ltr;block-progression:tb;writing-mode:lr-tb;text-anchor:start;baseline-shift:baseline;color:#000000;fill:#231f20;fill-opacity:1;stroke:none;stroke-width:1.5;marker:none;visibility:visible;display:inline;overflow:visible;enable-background:accumulate;font-family:Sans;-inkscape-font-specification:Sans"
+       d="m 359.84375,434.34375 0,3 -3,0 0,1.5 3,0 0,3.21875 1.5,0 0,-3.21875 3,0 0,-1.5 -3,0 0,-3 -1.5,0 z"
+       id="path3005-3"
+       inkscape:connector-curvature="0" />
+    <path
+       style="font-size:medium;font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;text-indent:0;text-align:start;text-decoration:none;line-height:normal;letter-spacing:normal;word-spacing:normal;text-transform:none;direction:ltr;block-progression:tb;writing-mode:lr-tb;text-anchor:start;baseline-shift:baseline;color:#000000;fill:#231f20;fill-opacity:1;stroke:none;stroke-width:0.9375;marker:none;visibility:visible;display:inline;overflow:visible;enable-background:accumulate;font-family:Sans;-inkscape-font-specification:Sans"
+       d="m 371.0625,443.5625 0,0.46875 -0.125,4.6875 0,0.15625 0.125,0.15625 c 0,0 0.37411,0.46147 0.9375,0.9375 0.2817,0.23802 0.59509,0.49145 0.96875,0.6875 0.37366,0.19605 0.8013,0.32003 1.25,0.34375 0.93515,0.0491 1.77613,-0.36791 2.40625,-0.78125 0.63012,-0.41334 1.0625,-0.84375 1.0625,-0.84375 l 0.125,-0.15625 0,-0.1875 -0.0937,-5 0,-0.46875 -0.46875,0 -5.71875,0 -0.46875,0 z"
+       id="path3009-7"
+       inkscape:connector-curvature="0"
+       sodipodi:nodetypes="cccccsscsccccccccc" />
+    <path
+       style="font-size:medium;font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;text-indent:0;text-align:start;text-decoration:none;line-height:normal;letter-spacing:normal;word-spacing:normal;text-transform:none;direction:ltr;block-progression:tb;writing-mode:lr-tb;text-anchor:start;baseline-shift:baseline;color:#000000;fill:#231f20;fill-opacity:1;stroke:none;stroke-width:0.9375;marker:none;visibility:visible;display:inline;overflow:visible;enable-background:accumulate;font-family:Sans;-inkscape-font-specification:Sans"
+       d="m 377.65625,450.375 -0.0312,0.0312 -0.21875,0.125 c -2.30252,1.65271 -4.15395,1.52806 -6.1125,-0.1562 -1.78014,2.66032 -3.73192,7.25254 -4.29375,8.18745 -0.56183,0.93491 -0.17739,2.37424 -0.28125,3.375 l 0.5625,-0.0312 8.03125,-0.71875 0.125,-0.0312 0.0937,-0.0625 6.96875,-4.5625 0.1875,-0.125 0.0937,-0.0625 -0.0625,-0.3125 c 0.0108,-0.11043 0.0366,-0.43638 0,-0.96875 -0.0466,-1.03041 -0.48309,-1.60724 -0.93745,-2.24995 -0.5467,-0.58055 -1.33798,-1.08483 -2.15625,-1.5 -0.91162,-0.46254 -1.66155,-0.7891 -1.71875,-0.8125 l -0.0625,-0.0312 -0.1875,-0.0937 z"
+       id="path3013-6"
+       inkscape:connector-curvature="0"
+       sodipodi:nodetypes="cccczcccccccccccscccc" />
+    <path
+       style="font-size:medium;font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;text-indent:0;text-align:start;text-decoration:none;line-height:normal;letter-spacing:normal;word-spacing:normal;text-transform:none;direction:ltr;block-progression:tb;writing-mode:lr-tb;text-anchor:start;baseline-shift:baseline;color:#000000;fill:#231f20;fill-opacity:1;stroke:none;stroke-width:0.625;marker:none;visibility:visible;display:inline;overflow:visible;enable-background:accumulate;font-family:Sans;-inkscape-font-specification:Sans"
+       d="m 374.0625,438.21875 c -1.68685,0.1362 -5.3125,0.8125 -5.3125,0.8125 l -0.125,0.0312 -0.0625,0.125 -1.375,2.09375 -0.0312,0.0625 0,0.0937 0.1875,6.15625 0.0312,0.15625 0.125,0.0937 0.3125,0.21875 0.5,0.3125 -0.0312,-0.59375 -0.0937,-5.90625 c 0.081,-0.0694 0.35604,-0.32331 0.6875,-0.4375 l 0.0312,0 c 0.0323,-0.01 0.0883,0.01 0.125,0 l 1.46875,-0.0937 c 0.1093,0.60186 0.39332,1.20774 0.4812,1.8125 1.27397,-0.0782 2.40491,-0.17552 3.5031,-0.16892 1.08715,0.0103 2.17115,0.13001 3.2344,0.12517 l 0.0312,-0.6125 c 0,0 0.0422,-0.16533 0.0625,-0.28125 l 0.40625,-0.9375 c 0.004,-0.005 -0.004,-0.0267 0,-0.0312 l 0.0312,0 c 0.17093,-0.1709 0.53319,-0.25585 0.8125,-0.375 0.26159,-0.11159 0.42068,-0.28594 0.75,-0.375 l 0.0312,0 0.125,-0.0312 2.28125,-0.1875 0.0312,0 0.28125,0 0,-0.0312 0,-0.28125 0,-0.71875 0,-0.25 -0.25,-0.0625 -3.75,-0.71875 c -0.0114,-10e-4 -0.02,0.001 -0.0312,0 -1.37083,-0.13192 -2.78705,-0.1247 -4.40625,0 l -0.0312,0 -0.0312,0 z"
+       id="path3017-6"
+       inkscape:connector-curvature="0"
+       sodipodi:nodetypes="ccccccccccccccccccccccccccsccccccccccccccccc" />
+  </g>
+</svg>

samples/CMakeLists.txt

@@ -0,0 +1,16 @@
+# CMake configuration for ttmath/samples
+
+# Building with Visual C++ x86_64 needs to compile the asm utilities first
+if (MSVC AND "x${CMAKE_VS_PLATFORM_NAME}" STREQUAL "xx64")
+  set(TTMATH_MSVC64_ASM ttmathuint_x86_64_msvc.asm) 
+  enable_language(ASM_MASM)
+  message(STATUS "Enabled MASM to compile '${TTMATH_MSVC64_ASM}'")
+  set(TTMATH_SRC_ASM ${ttmath_SOURCE_DIR}/ttmath/${TTMATH_MSVC64_ASM})
+endif()
+
+set(SAMPLES big big2 int uint parser)
+foreach(sample ${SAMPLES})
+	add_executable(${sample} ${sample}.cpp ${TTMATH_SRC_ASM})
+	# Allow to run all utilities at once as a test
+	add_test(${sample} ${sample})
+endforeach()

samples/Makefile

@@ -1,5 +1,6 @@
 CC     = g++
-CFLAGS = -Wall -pedantic -s -O2 -I.. 
+#CC     = clang++
+CFLAGS = -Wall -pedantic -O2 -I.. -DTTMATH_DONT_USE_WCHAR
 
 
 .SUFFIXES: .cpp .o
@@ -8,25 +9,29 @@ CFLAGS = -Wall -pedantic -s -O2 -I..
 	$(CC) -c $(CFLAGS) $<
 
 
-all: uint int big parser
+all: uint int big big2 parser
 
 
 uint: uint.o
-	$(CC) -o uint $(CFLAGS) uint.o
+	$(CC) -o uint -s $(CFLAGS) uint.o
 
 int: int.o
-	$(CC) -o int $(CFLAGS) int.o
+	$(CC) -o int -s $(CFLAGS) int.o
 
 big: big.o
-	$(CC) -o big $(CFLAGS) big.o
+	$(CC) -o big -s $(CFLAGS) big.o
+
+big2: big2.o
+	$(CC) -o big2 -s $(CFLAGS) big2.o
 
 parser: parser.o
-	$(CC) -o parser $(CFLAGS) parser.o
+	$(CC) -o parser -s $(CFLAGS) parser.o
 
 
 uint.o:		uint.cpp
 int.o:		int.cpp
 big.o:		big.cpp
+big2.o:		big2.cpp
 parser.o:	parser.cpp
 
 
@@ -36,6 +41,7 @@ clean:
 	rm -f uint
 	rm -f int
 	rm -f big
+	rm -f big2
 	rm -f parser
-# on MS Windows can automatically be added suffixes .exe to the names of output programs
+# on MS Windows suffixes .exe will be automatically added
 	rm -f *.exe

samples/big.cpp

@@ -5,8 +5,10 @@
 // this type has 2 words for its mantissa and 1 word for its exponent
 // (on a 32bit platform one word means a word of 32 bits,
 // and on a 64bit platform one word means a word of 64 bits)
+
+// Big<exponent, mantissa>
 typedef ttmath::Big<1,2> MyBig;
- 
+
        
 void SimpleCalculating(const MyBig & a, const MyBig & b)
 {
@@ -56,6 +58,7 @@ MyBig atemp;
 	
 }
 
+
 int main()
 {
 MyBig a,b;
@@ -85,13 +88,13 @@ a = 123456.543456
 b = 98767878.124322
 a + b = 98891334.667778
 a - b = -98644421.580866
-a * b = 12193540837712.2708
-a / b = 0.0012499665458095765
+a * b = 12193540837712.27076
+a / b = 0.00124996654580957646
 Calculating with a carry
-a = 1.624801256070839555e+646457012
-b = 456.31999999999999
-a + b = 1.624801256070839555e+646457012
-a - b = 1.624801256070839555e+646457012
+a = 1.6248012560666408782e+646457012
+b = 456.319999999999993
+a + b = 1.6248012560666408782e+646457012
+a - b = 1.6248012560666408782e+646457012
 a * b = (carry)
-a / b = 3.56066193914542334e+646457009
+a / b = 3.560661939136222174e+646457009
 */

samples/big2.cpp

@@ -0,0 +1,113 @@
+#include <ttmath/ttmath.h>
+#include <iostream>
+
+
+// this is a similar example to big.cpp
+// but now we're using TTMATH_BITS() macro
+// this macro returns how many words we need to store
+// the given number of bits
+
+// TTMATH_BITS(64) 
+//   on a 32bit platform the macro returns 2 (2*32=64)
+//   on a 64bit platform the macro returns 1
+
+// TTMATH_BITS(128)
+//   on a 32bit platform the macro returns 4 (4*32=128)
+//   on a 64bit platform the macro returns 2 (2*64=128)
+
+// Big<exponent, mantissa>
+typedef ttmath::Big<TTMATH_BITS(64), TTMATH_BITS(128)> MyBig;
+
+// consequently on a 32bit platform we define: Big<2, 4>
+// and on a 64bit platform: Big<1, 2>
+// and the calculations will be the same on both platforms       
+       
+       
+void SimpleCalculating(const MyBig & a, const MyBig & b)
+{
+	std::cout << "Simple calculating" << std::endl;
+	std::cout << "a = " << a << std::endl;
+	std::cout << "b = " << b << std::endl;
+	std::cout << "a + b = " << a+b << std::endl;
+	std::cout << "a - b = " << a-b << std::endl;
+	std::cout << "a * b = " << a*b << std::endl;
+	std::cout << "a / b = " << a/b << std::endl;
+}
+
+
+void CalculatingWithCarry(const MyBig & a, const MyBig & b)
+{
+MyBig atemp;
+
+	std::cout << "Calculating with a carry" << std::endl;
+	std::cout << "a = " << a << std::endl;
+	std::cout << "b = " << b << std::endl;
+
+	atemp = a;
+	if( !atemp.Add(b) )
+		std::cout << "a + b = " << atemp << std::endl;
+	else
+		std::cout << "a + b = (carry)" << std::endl;
+		// it have no sense to print 'atemp' (it's undefined)
+
+	atemp = a;
+	if( !atemp.Sub(b) )
+		std::cout << "a - b = " << atemp << std::endl;
+	else
+		std::cout << "a - b = (carry)" << std::endl;
+		
+	atemp = a;
+	if( !atemp.Mul(b) )
+		std::cout << "a * b = " << atemp << std::endl;
+	else
+		std::cout << "a * b = (carry)" << std::endl;
+		
+		
+	atemp = a;
+	if( !atemp.Div(b) )
+		std::cout << "a / b = " << atemp << std::endl;
+	else
+		std::cout << "a / b = (carry or division by zero) " << std::endl;
+	
+}
+
+
+int main()
+{
+MyBig a,b;
+	
+	// conversion from 'const char *'
+	a = "123456.543456";
+	b = "98767878.124322";
+	
+	SimpleCalculating(a,b);
+	
+	// 'a' will have the max value which can be held in this type
+	a.SetMax();
+	
+	// conversion from double
+	b = 456.32;
+	
+	// Look at the value 'a' and the product from a+b and a-b
+	// Don't worry this is the nature of floating point numbers
+	CalculatingWithCarry(a,b);
+}
+
+/*
+the result (the same on a 32 or 64bit platform):
+
+Simple calculating
+a = 123456.543456
+b = 98767878.124322
+a + b = 98891334.667778
+a - b = -98644421.580866
+a * b = 12193540837712.270763536832
+a / b = 0.0012499665458095764605964485261668609133
+Calculating with a carry
+a = 2.34953455457111777368832820909595050034e+2776511644261678604
+b = 456.3199999999999931787897367030382156
+a + b = 2.34953455457111777368832820909595050034e+2776511644261678604
+a - b = 2.34953455457111777368832820909595050034e+2776511644261678604
+a * b = (carry)
+a / b = 5.1488748127873374141170361292780486452e+2776511644261678601
+*/

samples/parser.cpp

@@ -29,6 +29,11 @@ const char equation[] = " (34 + 24) * 123 - 34.32 ^ 6 * sin(2.56) - atan(10)";
 
 /*
 the result (on 32 bit platform):
-
--897705014.52573107
+-897705014.525731067
+*/
+
+
+/*
+the result (on 64 bit platform):
+-897705014.5257310676097719585259773124
 */

tests/Makefile

@@ -1,6 +1,6 @@
-CC     = g++
+CC     = clang++
 o      = main.o uinttest.o
-CFLAGS = -Wall -O2 -s
+CFLAGS = -Wall -O2
 ttmath = ..
 name   = tests
 
@@ -16,7 +16,7 @@ all: $(name)
 
 
 $(name): $(o)
-	$(CC) -o $(name) $(CFLAGS) -I$(ttmath) $(o)
+	$(CC) -o $(name) -s $(CFLAGS) -I$(ttmath) $(o)
 
 
 main.o: main.cpp uinttest.h

tests/main.cpp

@@ -1,7 +1,7 @@
 /*
  * This file is a part of TTMath Bignum Library
- * and is distributed under the (new) BSD licence.
- * Author: Tomasz Sowa <t.sowa@slimaczek.pl>
+ * and is distributed under the 3-Clause BSD Licence.
+ * Author: Tomasz Sowa <t.sowa@ttmath.org>
  */
 
 /* 

tests/uinttest.cpp

@@ -1,7 +1,7 @@
 /*
  * This file is a part of TTMath Bignum Library
- * and is distributed under the (new) BSD licence.
- * Author: Tomasz Sowa <t.sowa@slimaczek.pl>
+ * and is distributed under the 3-Clause BSD Licence.
+ * Author: Tomasz Sowa <t.sowa@ttmath.org>
  */
 
 /* 

tests/uinttest.h

@@ -1,7 +1,7 @@
 /*
  * This file is a part of TTMath Bignum Library
- * and is distributed under the (new) BSD licence.
- * Author: Tomasz Sowa <t.sowa@slimaczek.pl>
+ * and is distributed under the 3-Clause BSD Licence.
+ * Author: Tomasz Sowa <t.sowa@ttmath.org>
  */
 
 /* 

tests2/Makefile

@@ -0,0 +1,133 @@
+#CC     = g++6
+#CFLAGS = -Wall -pedantic -O3 -s -I.. -Wl,-rpath=/usr/local/lib/gcc6
+#CFLAGS = -Wall -pedantic -O3 -DTTMATH_NOASM -s -I.. -Wl,-rpath=/usr/local/lib/gcc6
+
+CC     = clang++
+CFLAGS = -Wall -pedantic -O3 -s -I..
+#CFLAGS = -Wall -pedantic -O3 -DTTMATH_NOASM -s -I..
+
+
+
+.SUFFIXES: .cpp .o
+
+.cpp.o:
+	$(CC) -c $(CFLAGS) $<
+
+
+
+
+all: big_64_64 big_64_128 big_64_192 big_64_256 big_64_512 big_64_1024 big_64_2048 big_64_4096 big_128_512 big_256_1024 big_512_2048 big_128_4096
+
+
+
+
+
+
+big_64_64: main.cpp
+	$(CC) -o big_64_64 -s $(CFLAGS) -DTTMATH_TEST_BIG_EXPONENT=64 -DTTMATH_TEST_BIG_MANTISSA=64 main.cpp
+
+big_64_128: main.cpp
+	$(CC) -o big_64_128 -s $(CFLAGS) -DTTMATH_TEST_BIG_EXPONENT=64 -DTTMATH_TEST_BIG_MANTISSA=128 main.cpp
+
+big_64_192: main.cpp
+	$(CC) -o big_64_192 -s $(CFLAGS) -DTTMATH_TEST_BIG_EXPONENT=64 -DTTMATH_TEST_BIG_MANTISSA=192 main.cpp
+
+big_64_256: main.cpp
+	$(CC) -o big_64_256 -s $(CFLAGS) -DTTMATH_TEST_BIG_EXPONENT=64 -DTTMATH_TEST_BIG_MANTISSA=256 main.cpp
+
+big_64_512: main.cpp
+	$(CC) -o big_64_512 -s $(CFLAGS) -DTTMATH_TEST_BIG_EXPONENT=64 -DTTMATH_TEST_BIG_MANTISSA=512 main.cpp
+
+big_64_1024: main.cpp
+	$(CC) -o big_64_1024 -s $(CFLAGS) -DTTMATH_TEST_BIG_EXPONENT=64 -DTTMATH_TEST_BIG_MANTISSA=1024 main.cpp
+
+big_64_2048: main.cpp
+	$(CC) -o big_64_2048 -s $(CFLAGS) -DTTMATH_TEST_BIG_EXPONENT=64 -DTTMATH_TEST_BIG_MANTISSA=2048 main.cpp
+
+big_64_4096: main.cpp
+	$(CC) -o big_64_4096 -s $(CFLAGS) -DTTMATH_TEST_BIG_EXPONENT=64 -DTTMATH_TEST_BIG_MANTISSA=4096 main.cpp
+
+big_128_512: main.cpp
+	$(CC) -o big_128_512 -s $(CFLAGS) -DTTMATH_TEST_BIG_EXPONENT=128 -DTTMATH_TEST_BIG_MANTISSA=512 main.cpp
+
+big_256_1024: main.cpp
+	$(CC) -o big_256_1024 -s $(CFLAGS) -DTTMATH_TEST_BIG_EXPONENT=256 -DTTMATH_TEST_BIG_MANTISSA=1024 main.cpp
+
+big_512_2048: main.cpp
+	$(CC) -o big_512_2048 -s $(CFLAGS) -DTTMATH_TEST_BIG_EXPONENT=512 -DTTMATH_TEST_BIG_MANTISSA=2048 main.cpp
+
+big_128_4096: main.cpp
+	$(CC) -o big_128_4096 -s $(CFLAGS) -DTTMATH_TEST_BIG_EXPONENT=128 -DTTMATH_TEST_BIG_MANTISSA=4096 main.cpp
+
+
+
+
+test: all
+	@echo "****************************************************************************"
+	@echo "making tests for exponent=64 and mantissa=64"
+	@echo "****************************************************************************"
+	cat tests.txt | xargs -S 4096 -I foo ./big_64_64 "foo" | tee big_64_64.out
+
+	@echo "****************************************************************************"
+	@echo "making tests for exponent=64 and mantissa=128"
+	@echo "****************************************************************************"
+	cat tests.txt | xargs -S 4096 -I foo ./big_64_128 "foo" | tee big_64_128.out
+
+	@echo "****************************************************************************"
+	@echo "making tests for exponent=64 and mantissa=192"
+	@echo "****************************************************************************"
+	cat tests.txt | xargs -S 4096 -I foo ./big_64_192 "foo" | tee big_64_192.out
+
+	@echo "****************************************************************************"
+	@echo "making tests for exponent=64 and mantissa=256"
+	@echo "****************************************************************************"
+	cat tests.txt | xargs -S 4096 -I foo ./big_64_256 "foo" | tee big_64_256.out
+
+	@echo "****************************************************************************"
+	@echo "making tests for exponent=64 and mantissa=512"
+	@echo "****************************************************************************"
+	cat tests.txt | xargs -S 4096 -I foo ./big_64_512 "foo" | tee big_64_512.out
+
+	@echo "****************************************************************************"
+	@echo "making tests for exponent=64 and mantissa=1024"
+	@echo "****************************************************************************"
+	cat tests.txt | xargs -S 4096 -I foo ./big_64_1024 "foo" | tee big_64_1024.out
+
+	@echo "****************************************************************************"
+	@echo "making tests for exponent=64 and mantissa=2048"
+	@echo "****************************************************************************"
+	cat tests.txt | xargs -S 4096 -I foo ./big_64_2048 "foo" | tee big_64_2048.out
+
+	@echo "****************************************************************************"
+	@echo "making tests for exponent=64 and mantissa=4096"
+	@echo "****************************************************************************"
+	cat tests.txt | xargs -S 4096 -I foo ./big_64_4096 "foo" | tee big_64_4096.out
+
+	@echo "****************************************************************************"
+	@echo "making tests for exponent=128 and mantissa=512"
+	@echo "****************************************************************************"
+	cat tests.txt | xargs -S 4096 -I foo ./big_128_512 "foo" | tee big_128_512.out
+
+	@echo "****************************************************************************"
+	@echo "making tests for exponent=128 and mantissa=4096"
+	@echo "****************************************************************************"
+	cat tests.txt | xargs -S 4096 -I foo ./big_128_4096 "foo" | tee big_128_4096.out
+
+	@echo "****************************************************************************"
+	@echo "making tests for exponent=256 and mantissa=1024"
+	@echo "****************************************************************************"
+	cat tests.txt | xargs -S 4096 -I foo ./big_256_1024 "foo" | tee big_256_1024.out
+
+	@echo "****************************************************************************"
+	@echo "making tests for exponent=512 and mantissa=2048"
+	@echo "****************************************************************************"
+	cat tests.txt | xargs -S 4096 -I foo ./big_512_2048 "foo" | tee big_512_2048.out
+
+	./check_files.sh
+
+
+clean:
+	rm -f *.out
+	rm -f big_64_64 big_64_128 big_64_192 big_64_256 big_64_512 big_64_1024 big_64_2048 big_64_4096 big_128_512 big_256_1024 big_512_2048 big_128_4096
+# on MS Windows suffixes .exe will be automatically added
+	rm -f *.exe

tests2/check_files.sh

@@ -0,0 +1,31 @@
+#!/bin/sh
+
+was_error=0
+
+for expected in *.expected ; do
+
+	out=`basename $expected .expected`.out
+	
+	if [ -f $out ] ; then
+		diff -u $out $expected
+		
+		if [ `diff -u $out $expected | wc -l` -ne 0 ] ; then
+			was_error=1
+		fi
+	else
+		echo "there is no file: $out"
+		was_error=1
+	fi
+
+done
+
+
+if [ $was_error -eq 0 ] ; then
+	echo "****************************************************************************"
+	echo "             congratulations: all tests passed successfully"
+	echo "****************************************************************************"
+else
+	echo "****************************************************************************"
+	echo "             error: not all tests passed successfully"
+	echo "****************************************************************************"
+fi

tests2/main.cpp

@@ -0,0 +1,39 @@
+#include <ttmath/ttmath.h>
+#include <iostream>
+
+int main(int argc, char ** argv)
+{
+	typedef ttmath::Big<TTMATH_BITS(TTMATH_TEST_BIG_EXPONENT), TTMATH_BITS(TTMATH_TEST_BIG_MANTISSA)> MyBig;
+	ttmath::Parser<MyBig> parser;
+	std::string all_input;
+
+	for(int i=1 ; i<argc ; ++i)
+	{
+		if( i > 1 )
+			all_input += ' ';
+			
+		all_input += argv[i];
+	}
+
+	std::cout << all_input << " = ";
+	ttmath::ErrorCode err = parser.Parse(all_input);
+
+	if( err == ttmath::err_ok )
+	{
+		for(size_t i=0 ; i < parser.stack.size() ; ++i)
+		{
+			if( i > 0 )
+				std::cout << " ; ";
+			
+			std::cout << parser.stack[i].value;
+		}
+		
+		std::cout << std::endl;
+	}
+	else
+	{
+		std::cout << "error: " << static_cast<int>(err) << std::endl;
+	}
+	
+	return 0;
+}

ttmath/ttmathdec.h

@@ -0,0 +1,419 @@
+/*
+ * This file is a part of TTMath Bignum Library
+ * and is distributed under the 3-Clause BSD Licence.
+ * Author: Tomasz Sowa <t.sowa@ttmath.org>
+ */
+
+/* 
+ * Copyright (c) 2012, Tomasz Sowa
+ * All rights reserved.
+ * 
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are met:
+ * 
+ *  * Redistributions of source code must retain the above copyright notice,
+ *    this list of conditions and the following disclaimer.
+ *    
+ *  * Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in the
+ *    documentation and/or other materials provided with the distribution.
+ *    
+ *  * Neither the name Tomasz Sowa nor the names of contributors to this
+ *    project may be used to endorse or promote products derived
+ *    from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+ * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+ * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
+ * THE POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#ifndef headerfilettmathdec
+#define headerfilettmathdec
+
+#include "ttmathtypes.h"
+#include "ttmaththreads.h"
+#include "ttmathuint.h"
+
+
+
+namespace ttmath
+{
+
+template<uint value_size, uint dec_digits>
+class Dec
+{
+public:
+
+	UInt<value_size> value;
+	unsigned char info;
+
+
+	/*!
+		Sign
+		the mask of a bit from 'info' which means that there is a sign
+		(when the bit is set)
+	*/
+	#define TTMATH_DEC_SIGN 128
+
+
+	/*!
+		Not a number
+		if this bit is set that there is not a valid number
+	*/
+	#define TTMATH_DEC_NAN  64
+
+
+
+
+	Dec()
+	{
+		info = TTMATH_DEC_NAN;
+	}
+
+
+	Dec(const char * s)
+	{
+		info = TTMATH_DEC_NAN;
+		FromString(s);
+	}
+
+
+	Dec<value_size, dec_digits> & operator=(const char * s)
+	{
+		FromString(s);
+
+	return *this;
+	}
+
+
+	uint FromString(const char * s, const char ** after_source = 0, bool * value_read = 0)
+	{
+		return FromStringBase(s, after_source, value_read);
+	}
+
+
+	void ToString(std::string & result) const
+	{
+		ToStringBase(result);
+	}
+
+
+	/*!
+		this method clears a specific bit in the 'info' variable
+
+		bit is one of: 
+	*/
+	void ClearInfoBit(unsigned char bit)
+	{
+		info = info & (~bit);
+	}
+
+
+	/*!
+		this method sets a specific bit in the 'info' variable
+
+		bit is one of: 
+
+	*/
+	void SetInfoBit(unsigned char bit)
+	{
+		info = info | bit;
+	}
+
+
+	/*!
+		this method returns true if a specific bit in the 'info' variable is set
+
+		bit is one of: 
+	*/
+	bool IsInfoBit(unsigned char bit) const
+	{
+		return (info & bit) != 0;
+	}
+
+
+	bool IsNan() const 
+	{
+		return IsInfoBit(TTMATH_DEC_NAN);
+	}
+
+
+	bool IsSign() const 
+	{
+		return IsInfoBit(TTMATH_DEC_SIGN);
+	}
+
+
+	/*!
+		this method sets the sign
+
+			e.g.
+			-1 -> -1
+			2  -> -2
+
+		we do not check whether there is a zero or not, if you're using this method
+		you must be sure that the value is (or will be afterwards) different from zero
+	*/
+	void SetSign()
+	{
+		SetInfoBit(TTMATH_DEC_SIGN);
+	}
+
+
+	void SetNaN()
+	{
+		SetInfoBit(TTMATH_DEC_NAN);
+	}
+
+
+	void Abs()
+	{
+		ClearInfoBit(TTMATH_DEC_SIGN);
+	}
+
+
+
+	uint Add(const Dec<value_size, dec_digits> & arg)
+	{
+	uint c = 0;
+
+		if( IsSign() == arg.IsSign() )
+		{
+			c += value.Add(arg.value);		
+		}
+		else
+		{
+			bool is_sign;
+
+			if( value > arg.value )
+			{
+				is_sign = IsSign();
+				value.Sub(arg.value);
+			}
+			else
+			{
+				is_sign = arg.IsSign();
+				UInt<value_size> temp(this->value);
+				value = arg.value;
+				value.Sub(temp);
+			}
+
+			is_sign ? SetSign() : Abs();
+		}
+
+		if( c )
+			SetNaN();
+
+	return (c==0)? 0 : 1;
+	}
+
+/*
+	uint Sub(const Dec<value_size, dec_digits> & arg)
+	{
+	}
+*/
+
+private:
+
+
+
+
+
+
+#ifndef TTMATH_MULTITHREADS
+
+	/*!
+	*/
+	void SetMultipler(UInt<value_size> & result)
+	{
+		// this guardian is initialized before the program runs (static POD type)
+		static int guardian = 0;
+		static UInt<value_size> multipler;
+	
+		if( guardian == 0 )
+		{
+			multipler = 10;
+			multipler.Pow(dec_digits);
+			guardian = 1;
+		}
+
+		result = multipler;
+	}
+
+#else
+
+	/*!
+	*/
+	void SetMultipler(UInt<value_size> & result)
+	{
+		// this guardian is initialized before the program runs (static POD type)
+		volatile static sig_atomic_t guardian = 0;
+		static UInt<value_size> * pmultipler;
+	
+		// double-checked locking
+		if( guardian == 0 )
+		{
+			ThreadLock thread_lock;
+
+			// locking
+			if( thread_lock.Lock() )
+			{
+				static UInt<value_size> multipler;
+
+				if( guardian == 0 )
+				{
+					pmultipler = &multipler;
+					multipler = 10;
+					multipler.Pow(dec_digits);
+					guardian = 1;
+				}
+			}
+			else
+			{
+				// there was a problem with locking, we store the result directly in 'result' object
+				result = 10;
+				result.Pow(dec_digits);
+				
+			return;
+			}
+
+			// automatically unlocking
+		}
+
+		result = *pmultipler;
+	}
+
+#endif
+
+
+
+	/*!
+		an auxiliary method for converting from a string
+	*/
+	template<class char_type>
+	uint FromStringBase(const char_type * s, const char_type ** after_source = 0, bool * value_read = 0)
+	{
+		UInt<value_size> multipler;
+		const char_type * after;
+		uint c = 0;
+		info = 0;
+
+		Misc::SkipWhiteCharacters(s);
+
+		if( *s == '-' )
+		{
+			s += 1;
+			SetSign();
+		}
+		else
+		if( *s == '+' )
+		{
+			s += 1;
+		}
+
+		c += value.FromString(s, 10, &after, value_read);
+
+		if( after_source )
+			*after_source = after;
+
+		SetMultipler(multipler);
+		c += value.Mul(multipler);
+
+		if( *after == '.' )
+			c += FromStringBaseAfterComma(after+1, after_source);
+
+		if( c )
+			SetInfoBit(TTMATH_DEC_NAN);
+
+	return (c==0)? 0 : 1;
+	}
+
+
+	template<class char_type>
+	uint FromStringBaseAfterComma(const char_type * s, const char_type ** after_source = 0, bool * value_read = 0)
+	{
+		UInt<value_size> temp;
+		UInt<value_size> multipler;
+		sint z;
+		uint c = 0;
+		size_t i = dec_digits;
+
+		SetMultipler(multipler);
+
+		for( ; i>0 && (z=Misc::CharToDigit(*s, 10)) != -1 ; --i, ++s )
+		{
+			multipler.DivInt(10);
+			temp.SetZero();
+
+			if( value_read )
+				*value_read = true;
+
+			if( c == 0 )
+			{
+				temp.table[0] = z;
+				c += temp.Mul(multipler);
+				c += value.Add(temp);
+			}
+		}
+
+		if( i == 0 && (z=Misc::CharToDigit(*s, 10)) != -1 && z >= 5 )
+			c += value.AddOne();
+
+		if( after_source )
+		{
+			while( (z=Misc::CharToDigit(*s, 10)) != -1 )
+				s += 1;
+
+			*after_source = s;
+		}
+
+	return c;
+	}
+
+
+
+	template<class string_type>
+	void ToStringBase(string_type & result) const
+	{
+		if( IsNan() )
+		{
+			result = "NaN";
+			return;
+		}
+
+		value.ToStringBase(result, 10, IsSign());
+
+		if( dec_digits > 0 )
+		{
+			size_t size = result.size();
+
+			if( IsSign() && size > 0 )
+				size -= 1;
+
+			if( dec_digits >= size )
+			{
+				size_t zeroes = dec_digits - size + 1;
+				size_t start  = IsSign() ? 1 : 0;
+				result.insert(start, zeroes, '0');
+			}
+
+			result.insert(result.end() - dec_digits, '.');
+		}
+	}
+
+
+
+};
+
+
+} // namespace
+
+#endif

ttmath/ttmathmisc.h

@@ -0,0 +1,250 @@
+/*
+ * This file is a part of TTMath Bignum Library
+ * and is distributed under the 3-Clause BSD Licence.
+ * Author: Tomasz Sowa <t.sowa@ttmath.org>
+ */
+
+/* 
+ * Copyright (c) 2006-2010, Tomasz Sowa
+ * All rights reserved.
+ * 
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are met:
+ * 
+ *  * Redistributions of source code must retain the above copyright notice,
+ *    this list of conditions and the following disclaimer.
+ *    
+ *  * Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in the
+ *    documentation and/or other materials provided with the distribution.
+ *    
+ *  * Neither the name Tomasz Sowa nor the names of contributors to this
+ *    project may be used to endorse or promote products derived
+ *    from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+ * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+ * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
+ * THE POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#ifndef headerfilettmathmisc
+#define headerfilettmathmisc
+
+
+/*!
+	\file ttmathmisc.h
+    \brief some helpful functions
+*/
+
+
+#include <string>
+
+
+namespace ttmath
+{
+
+/*!
+	some helpful functions
+*/
+class Misc
+{
+public:
+
+
+/*
+ *
+ *	AssignString(result, str)
+ *	result = str
+ *
+ */
+
+/*!
+	result = str
+*/
+static void AssignString(std::string & result, const char * str)
+{
+	result = str;
+}
+
+
+#ifndef TTMATH_DONT_USE_WCHAR
+
+/*!
+	result = str
+*/
+static void AssignString(std::wstring & result, const char * str)
+{
+	result.clear();
+
+	for( ; *str ; ++str )
+		result += *str;
+}
+
+
+/*!
+	result = str
+*/
+static void AssignString(std::wstring & result, const std::string & str)
+{
+	return AssignString(result, str.c_str());
+}
+
+
+/*!
+	result = str
+*/
+static void AssignString(std::string & result, const wchar_t * str)
+{
+	result.clear();
+
+	for( ; *str ; ++str )
+		result += static_cast<char>(*str);
+}
+
+
+/*!
+	result = str
+*/
+static void AssignString(std::string & result, const std::wstring & str)
+{
+	return AssignString(result, str.c_str());
+}
+
+#endif
+
+
+/*
+ *
+ *	AddString(result, str)
+ *	result += str
+ *
+ */
+
+
+/*!
+	result += str
+*/
+static void AddString(std::string & result, const char * str)
+{
+	result += str;
+}
+
+
+#ifndef TTMATH_DONT_USE_WCHAR
+
+/*!
+	result += str
+*/
+static void AddString(std::wstring & result, const char * str)
+{
+	for( ; *str ; ++str )
+		result += *str;
+}
+
+#endif
+
+
+/*
+	this method omits any white characters from the string
+	char_type is char or wchar_t
+*/
+template<class char_type>
+static void SkipWhiteCharacters(const char_type * & c)
+{
+	// 13 is at the end in a DOS text file (\r\n)
+	while( (*c==' ' ) || (*c=='\t') || (*c==13 ) || (*c=='\n') )
+		++c;
+}
+
+
+
+
+/*!
+	this static method converts one character into its value
+
+	for example:
+	-  1 -> 1
+	-  8 -> 8
+	-  A -> 10
+	-  f -> 15
+
+	this method don't check whether c is correct or not
+*/
+static uint CharToDigit(uint c)
+{
+	if(c>='0' && c<='9')
+		return c-'0';
+
+	if(c>='a' && c<='z')
+		return c-'a'+10;
+
+return c-'A'+10;
+}
+
+
+/*!
+	this method changes a character 'c' into its value
+	(if there can't be a correct value it returns -1)
+
+	for example:
+	-  c=2, base=10 -> function returns 2
+	-  c=A, base=10 -> function returns -1
+	-  c=A, base=16 -> function returns 10
+*/
+static sint CharToDigit(uint c, uint base)
+{
+	if( c>='0' && c<='9' )
+		c=c-'0';
+	else
+	if( c>='a' && c<='z' )
+		c=c-'a'+10;
+	else
+	if( c>='A' && c<='Z' )
+		c=c-'A'+10;
+	else
+		return -1;
+
+
+	if( c >= base )
+		return -1;
+
+
+return sint(c);
+}
+
+
+
+/*!
+	this method converts a digit into a char
+	digit should be from <0,F>
+	(we don't have to get a base)
+	
+	for example:
+	-  1  -> 1
+	-  8  -> 8
+	-  10 -> A
+	-  15 -> F
+*/
+static uint DigitToChar(uint digit)
+{
+	if( digit < 10 )
+		return digit + '0';
+
+return digit - 10 + 'A';
+}
+
+
+}; // struct Misc
+
+}
+
+
+#endif

ttmath/ttmathobjects.h

@@ -1,11 +1,11 @@
 /*
  * This file is a part of TTMath Mathematical Library
- * and is distributed under the (new) BSD licence.
- * Author: Tomasz Sowa <t.sowa@slimaczek.pl>
+ * and is distributed under the 3-Clause BSD Licence.
+ * Author: Tomasz Sowa <t.sowa@ttmath.org>
  */
 
 /* 
- * Copyright (c) 2006-2009, Tomasz Sowa
+ * Copyright (c) 2006-2017, Tomasz Sowa
  * All rights reserved.
  * 
  * Redistribution and use in source and binary forms, with or without
@@ -44,12 +44,13 @@
     \brief Mathematic functions.
 */
 
-#include "ttmathtypes.h"
-
 #include <string>
+#include <vector>
 #include <list>
 #include <map>
 
+#include "ttmathtypes.h"
+#include "ttmathmisc.h"
 
 
 namespace ttmath
@@ -73,13 +74,14 @@ public:
 	struct Item
 	{
 		// name of a variable of a function
+		// internally we store variables and funcions as std::string (not std::wstring even when wide characters are used)
 		std::string value;
 
 		// number of parameters required by the function
 		// (if there's a variable this 'param' is ignored)
 		int param;
 
-		Item() {}
+		Item() { param = 0; }
 		Item(const std::string & v, int p) : value(v), param(p) {}
 	};
 
@@ -112,7 +114,8 @@ public:
 	/*!
 		this method returns true if the name can be as a name of an object
 	*/
-	static bool IsNameCorrect(const std::string & name)
+	template<class string_type>
+	static bool IsNameCorrect(const string_type & name)
 	{
 		if( name.empty() )
 			return false;
@@ -120,7 +123,7 @@ public:
 		if( !CorrectCharacter(name[0], false) )
 			return false;
 
-		std::string::const_iterator i=name.begin();
+		typename string_type::const_iterator i = name.begin();
 
 		for(++i ; i!=name.end() ; ++i)
 			if( !CorrectCharacter(*i, true) )
@@ -145,6 +148,27 @@ public:
 	}
 
 
+
+#ifndef TTMATH_DONT_USE_WCHAR
+
+	/*!
+		this method returns true if such an object is defined (name exists)
+	*/
+	bool IsDefined(const std::wstring & name)
+	{
+		// we should check whether the name (in wide characters) are correct
+		// before calling AssignString() function
+		if( !IsNameCorrect(name) )
+			return false;
+
+		Misc::AssignString(str_tmp1, name);
+
+	return IsDefined(str_tmp1);
+	}
+
+#endif
+
+
 	/*!
 		this method adds one object (variable of function) into the table
 	*/
@@ -165,6 +189,27 @@ public:
 	}
 
 
+#ifndef TTMATH_DONT_USE_WCHAR
+
+	/*!
+		this method adds one object (variable of function) into the table
+	*/
+	ErrorCode Add(const std::wstring & name, const std::wstring & value, int param = 0)
+	{
+		// we should check whether the name (in wide characters) are correct
+		// before calling AssignString() function
+		if( !IsNameCorrect(name) )
+			return err_incorrect_name;
+
+		Misc::AssignString(str_tmp1, name);
+		Misc::AssignString(str_tmp2, value);
+		
+	return Add(str_tmp1, str_tmp2, param);
+	}
+
+#endif
+
+
 	/*!
 		this method returns 'true' if the table is empty
 	*/
@@ -222,6 +267,28 @@ public:
 	}
 
 
+#ifndef TTMATH_DONT_USE_WCHAR
+
+
+	/*!
+		this method changes the value and the number of parameters for a specific object
+	*/
+	ErrorCode EditValue(const std::wstring & name, const std::wstring & value, int param = 0)
+	{
+		// we should check whether the name (in wide characters) are correct
+		// before calling AssignString() function
+		if( !IsNameCorrect(name) )
+			return err_incorrect_name;
+
+		Misc::AssignString(str_tmp1, name);
+		Misc::AssignString(str_tmp2, value);
+		
+	return EditValue(str_tmp1, str_tmp2, param);
+	}
+
+#endif
+
+
 	/*!
 		this method changes the name of a specific object
 	*/
@@ -253,6 +320,29 @@ public:
 	}
 
 
+
+#ifndef TTMATH_DONT_USE_WCHAR
+
+
+	/*!
+		this method changes the name of a specific object
+	*/
+	ErrorCode EditName(const std::wstring & old_name, const std::wstring & new_name)
+	{
+		// we should check whether the name (in wide characters) are correct
+		// before calling AssignString() function
+		if( !IsNameCorrect(old_name) || !IsNameCorrect(new_name) )
+			return err_incorrect_name;
+
+		Misc::AssignString(str_tmp1, old_name);
+		Misc::AssignString(str_tmp2, new_name);
+
+	return EditName(str_tmp1, str_tmp2);
+	}
+
+#endif
+
+
 	/*!
 		this method deletes an object
 	*/
@@ -272,6 +362,27 @@ public:
 	}
 
 
+#ifndef TTMATH_DONT_USE_WCHAR
+
+
+	/*!
+		this method deletes an object
+	*/
+	ErrorCode Delete(const std::wstring & name)
+	{
+		// we should check whether the name (in wide characters) are correct
+		// before calling AssignString() function
+		if( !IsNameCorrect(name) )
+			return err_incorrect_name;
+
+		Misc::AssignString(str_tmp1, name);
+
+	return Delete(str_tmp1);
+	}	
+		
+#endif
+
+
 	/*!
 		this method gets the value of a specific object
 	*/
@@ -294,6 +405,28 @@ public:
 	}
 
 
+#ifndef TTMATH_DONT_USE_WCHAR
+
+	/*!
+		this method gets the value of a specific object
+	*/
+	ErrorCode GetValue(const std::wstring & name, std::wstring & value)
+	{
+		// we should check whether the name (in wide characters) are correct
+		// before calling AssignString() function
+		if( !IsNameCorrect(name) )
+			return err_incorrect_name;
+
+		Misc::AssignString(str_tmp1, name);
+		ErrorCode err = GetValue(str_tmp1, str_tmp2);
+		Misc::AssignString(value, str_tmp2);
+
+	return err;
+	}
+
+#endif
+
+
 	/*!
 		this method gets the value of a specific object
 		(this version is used for not copying the whole string)
@@ -317,6 +450,27 @@ public:
 	}
 
 
+#ifndef TTMATH_DONT_USE_WCHAR
+
+	/*!
+		this method gets the value of a specific object
+		(this version is used for not copying the whole string)
+	*/
+	ErrorCode GetValue(const std::wstring & name, const char ** value)
+	{
+		// we should check whether the name (in wide characters) are correct
+		// before calling AssignString() function
+		if( !IsNameCorrect(name) )
+			return err_incorrect_name;
+
+		Misc::AssignString(str_tmp1, name);
+
+	return GetValue(str_tmp1, value);
+	}
+
+#endif
+
+
 	/*!
 		this method gets the value and the number of parameters
 		of a specific object
@@ -330,7 +484,7 @@ public:
 
 		if( i == table.end() )
 		{
-			value.empty();
+			value.clear();
 			*param = 0;
 			return err_unknown_object;
 		}
@@ -342,6 +496,29 @@ public:
 	}
 
 
+#ifndef TTMATH_DONT_USE_WCHAR
+
+	/*!
+		this method gets the value and the number of parameters
+		of a specific object
+	*/
+	ErrorCode GetValueAndParam(const std::wstring & name, std::wstring & value, int * param)
+	{
+		// we should check whether the name (in wide characters) are correct
+		// before calling AssignString() function
+		if( !IsNameCorrect(name) )
+			return err_incorrect_name;
+
+		Misc::AssignString(str_tmp1, name);
+		ErrorCode err = GetValueAndParam(str_tmp1, str_tmp2, param);
+		Misc::AssignString(value, str_tmp2);
+
+	return err;
+	}
+
+#endif
+
+
 	/*!
 		this method sets the value and the number of parameters
 		of a specific object
@@ -368,6 +545,31 @@ public:
 	}
 
 
+#ifndef TTMATH_DONT_USE_WCHAR
+
+
+	/*!
+		this method sets the value and the number of parameters
+		of a specific object
+		(this version is used for not copying the whole string
+		but in fact we make one copying during AssignString())
+	*/
+	ErrorCode GetValueAndParam(const std::wstring & name, const char ** value, int * param)
+	{
+		// we should check whether the name (in wide characters) are correct
+		// before calling AssignString() function
+		if( !IsNameCorrect(name) )
+			return err_incorrect_name;
+
+		Misc::AssignString(str_tmp1, name);
+
+	return GetValueAndParam(str_tmp1, value, param);
+	}
+
+
+#endif
+
+
 	/*!
 		this method returns a pointer into the table
 	*/
@@ -380,6 +582,7 @@ public:
 private:
 
 	Table table;
+	std::string str_tmp1, str_tmp2;
 
 }; // end of class Objects
 
@@ -426,11 +629,11 @@ public:
 	
 	/*!
 		default constructor
-		default max size of the History's container is 10 items
+		default max size of the History's container is 15 items
 	*/
 	History()
 	{
-		buffer_max_size = 10;
+		buffer_max_size = 15;
 	}
 
 
@@ -487,10 +690,121 @@ public:
 	return false;
 	}
 
+
+	/*!
+		this methods deletes an item
+
+		we assume that there is only one item with the 'key'
+		(this methods removes the first one)
+	*/
+	bool Remove(const ValueType & key)
+	{
+		typename buffer_type::iterator i = buffer.begin();
+
+		for( ; i != buffer.end() ; ++i )
+		{
+			if( i->key == key )
+			{
+				buffer.erase(i);
+				return true;
+			}
+		}
+
+	return false;
+	}
+
+
 }; // end of class History
 
 
 
+/*!
+	this is an auxiliary class used when calculating Gamma() or Factorial()
+
+	in multithreaded environment you can provide an object of this class to
+	the Gamma() or Factorial() function, e.g;
+
+		typedef Big<1, 3> MyBig;
+		MyBig x = 123456;
+		CGamma<MyBig> cgamma;
+		std::cout << Gamma(x, cgamma);
+
+	each thread should have its own CGamma<> object
+
+	in a single-thread environment a CGamma<> object is a static variable
+	and you don't have to explicitly use it, e.g.
+
+		typedef Big<1, 3> MyBig;
+		MyBig x = 123456;
+		std::cout << Gamma(x);
+*/
+template<class ValueType>
+struct CGamma
+{
+	/*!
+		this table holds factorials
+			1
+			1
+			2
+			6
+			24
+			120
+			720
+			.......
+	*/
+	std::vector<ValueType> fact;
+
+
+	/*!
+		this table holds Bernoulli numbers
+			1
+			-0.5
+			0.166666666666666666666666667
+			0
+			-0.0333333333333333333333333333
+			0
+			0.0238095238095238095238095238
+			0
+			-0.0333333333333333333333333333
+			0
+			0.075757575757575757575757576
+			.....
+	*/
+	std::vector<ValueType> bern;
+
+
+	/*!
+		here we store some calculated values
+		(this is for speeding up, if the next argument of Gamma() or Factorial()
+		is in the 'history' then the result we are not calculating but simply
+		return from the 'history' object)
+	*/
+	History<ValueType> history;
+
+
+	/*!
+		this method prepares some coefficients: factorials and Bernoulli numbers
+		stored in 'fact' and 'bern' objects
+		
+		how many values should be depends on the size of the mantissa - if
+		the mantissa is larger then we must calculate more values
+		    for a mantissa which consists of 256 bits (8 words on a 32bit platform)
+			we have to calculate about 30 values (the size of fact and bern will be 30),
+			and for a 2048 bits mantissa we have to calculate 306 coefficients
+
+		you don't have to call this method, these coefficients will be automatically calculated
+		when they are needed
+
+		you must note that calculating these coefficients is a little time-consuming operation,
+		(especially when the mantissa is large) and first call to Gamma() or Factorial()
+		can take more time than next calls, and in the end this is the point when InitAll()
+		comes in handy: you can call this method somewhere at the beginning of your program
+	*/
+	void InitAll();
+	// definition is in ttmath.h
+};
+
+
 
 
 } // namespace

ttmath/ttmaththreads.h

@@ -0,0 +1,252 @@
+/*
+ * This file is a part of TTMath Bignum Library
+ * and is distributed under the 3-Clause BSD Licence.
+ * Author: Tomasz Sowa <t.sowa@ttmath.org>
+ */
+
+/* 
+ * Copyright (c) 2006-2009, Tomasz Sowa
+ * All rights reserved.
+ * 
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are met:
+ * 
+ *  * Redistributions of source code must retain the above copyright notice,
+ *    this list of conditions and the following disclaimer.
+ *    
+ *  * Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in the
+ *    documentation and/or other materials provided with the distribution.
+ *    
+ *  * Neither the name Tomasz Sowa nor the names of contributors to this
+ *    project may be used to endorse or promote products derived
+ *    from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+ * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+ * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
+ * THE POSSIBILITY OF SUCH DAMAGE.
+ */
+
+
+
+#ifndef headerfilettmaththreads
+#define headerfilettmaththreads
+
+#include "ttmathtypes.h"
+
+#ifdef TTMATH_WIN32_THREADS
+#include <windows.h>
+#include <cstdio>
+#endif
+
+#ifdef TTMATH_POSIX_THREADS
+#include <pthread.h>
+#endif
+
+
+
+/*!
+	\file ttmaththreads.h
+    \brief Some objects used in multithreads environment
+*/
+
+
+namespace ttmath
+{
+
+
+#ifdef TTMATH_WIN32_THREADS
+
+	/*
+		we use win32 threads
+	*/
+
+
+	/*!
+		in multithreads environment you should use TTMATH_MULTITHREADS_HELPER macro
+		somewhere in *.cpp file
+
+		(at the moment in win32 this macro does nothing)
+	*/
+	#define TTMATH_MULTITHREADS_HELPER
+
+
+	/*!
+		objects of this class are used to synchronize
+	*/
+	class ThreadLock
+	{
+		HANDLE mutex_handle;
+
+
+		void CreateName(char * buffer) const
+		{
+			#ifdef _MSC_VER
+			#pragma warning (disable : 4996)
+			// warning C4996: 'sprintf': This function or variable may be unsafe. Consider using sprintf_s instead.
+			#endif
+
+			sprintf(buffer, "TTMATH_LOCK_%ul", (unsigned long)GetCurrentProcessId());
+
+			#ifdef _MSC_VER
+			#pragma warning (default : 4996)
+			#endif
+		}
+
+
+	public:
+
+		bool Lock()
+		{
+		char buffer[50];
+
+			CreateName(buffer);
+			mutex_handle = CreateMutexA(0, false, buffer);
+
+			if( mutex_handle == 0 )
+				return false;
+
+			WaitForSingleObject(mutex_handle, INFINITE);
+
+		return true;
+		}
+
+
+		ThreadLock()
+		{
+			mutex_handle = 0;
+		}
+
+
+		~ThreadLock()
+		{
+			if( mutex_handle != 0 )
+			{
+				ReleaseMutex(mutex_handle);
+				CloseHandle(mutex_handle);
+			}
+		}
+	};
+
+#endif  // #ifdef TTMATH_WIN32_THREADS
+
+
+
+
+
+#ifdef TTMATH_POSIX_THREADS
+
+	/*
+		we use posix threads
+	*/
+
+
+	/*!
+		in multithreads environment you should use TTMATH_MULTITHREADS_HELPER macro
+		somewhere in *.cpp file
+		(this macro defines a pthread_mutex_t object used by TTMath library)
+	*/
+	#define TTMATH_MULTITHREADS_HELPER                          \
+	namespace ttmath                                            \
+	{                                                           \
+	pthread_mutex_t ttmath_mutex = PTHREAD_MUTEX_INITIALIZER;   \
+	}
+
+
+	/*!
+		ttmath_mutex will be defined by TTMATH_MULTITHREADS_HELPER macro 
+	*/
+	extern pthread_mutex_t ttmath_mutex;
+
+
+	/*!
+		\brief objects of this class are used to synchronize
+
+		this is a simple skeleton of a program in multithreads environment:
+
+			#define TTMATH_MULTITHREADS
+			#include<ttmath/ttmath.h>
+
+			TTMATH_MULTITHREADS_HELPER
+
+			int main()
+			{
+			[...]
+			}
+
+		make sure that macro TTMATH_MULTITHREADS is defined and (somewhere in *.cpp file)
+		use TTMATH_MULTITHREADS_HELPER macro (outside of any classes/functions/namespaces scope)
+	*/
+	class ThreadLock
+	{
+	public:
+
+		/*!
+ 	 		lock the current thread
+
+ 	 		it uses a global mutex created by TTMATH_MULTITHREADS_HELPER macro
+		*/
+		bool Lock()
+		{
+			if( pthread_mutex_lock(&ttmath_mutex) != 0 )
+				return false;
+
+		return true;
+		}
+
+
+		~ThreadLock()
+		{
+			pthread_mutex_unlock(&ttmath_mutex);
+		}
+	};
+
+#endif // #ifdef TTMATH_POSIX_THREADS
+
+
+
+
+#if !defined(TTMATH_POSIX_THREADS) && !defined(TTMATH_WIN32_THREADS)
+
+	/*!
+		we don't use win32 and pthreads
+	*/
+
+	/*!
+	*/
+	#define TTMATH_MULTITHREADS_HELPER
+
+
+	/*!
+		objects of this class are used to synchronize
+		actually we don't synchronize, the method Lock() returns always 'false'
+	*/
+	class ThreadLock
+	{
+	public:
+
+		bool Lock()
+		{
+			return false;
+		}
+	};
+
+
+#endif // #if !defined(TTMATH_POSIX_THREADS) && !defined(TTMATH_WIN32_THREADS)
+
+
+
+
+
+} // namespace
+
+#endif
+

ttmath/ttmathtypes.h

@@ -1,11 +1,11 @@
 /*
  * This file is a part of TTMath Bignum Library
- * and is distributed under the (new) BSD licence.
- * Author: Tomasz Sowa <t.sowa@slimaczek.pl>
+ * and is distributed under the 3-Clause BSD Licence.
+ * Author: Tomasz Sowa <t.sowa@ttmath.org>
  */
 
 /* 
- * Copyright (c) 2006-2009, Tomasz Sowa
+ * Copyright (c) 2006-2019, Tomasz Sowa
  * All rights reserved.
  * 
  * Redistribution and use in source and binary forms, with or without
@@ -54,63 +54,123 @@
 
 #include <stdexcept>
 #include <sstream>
+#include <vector>
 
-
-/*!
-	the version of the library
-
-	TTMATH_PRERELEASE_VER is either zero or one
-	if zero that means this is the release version of the library
-*/
-#define TTMATH_MAJOR_VER		0
-#define TTMATH_MINOR_VER		8
-#define TTMATH_REVISION_VER		5
-#define TTMATH_PRERELEASE_VER	0
-
-
-/*!
-	TTMATH_DEBUG
-	this macro enables further testing during writing your code
-	you don't have to define it in a release mode
-
-	if this macro is set then macros TTMATH_ASSERT and TTMATH_REFERENCE_ASSERT
-	are set as well	and these macros can throw an exception if a condition in it
-	is not fulfilled (look at the definition of TTMATH_ASSERT and TTMATH_REFERENCE_ASSERT)
-
-	TTMATH_RELEASE
-	if you are confident that your code is perfect you can define TTMATH_RELEASE
-	macro for example by using -D option in gcc
-	 gcc -DTTMATH_RELEASE -o myprogram myprogram.cpp 
-	or by defining this macro in your code before using any header files of this library
-
-	if TTMATH_RELEASE is not set then TTMATH_DEBUG is set automatically
-*/
-#ifndef TTMATH_RELEASE
-	#define TTMATH_DEBUG
+#ifndef _MSC_VER
+#include <stdint.h>
+// for uint64_t and int64_t on a 32 bit platform
 #endif
 
 
 
+/*!
+	the major version of the library
+
+	the version present to the end user is constructed in this way:
+
+		TTMATH_MAJOR_VER.TTMATH_MINOR_VER.TTMATH_REVISION_VER.[prerelease if TTMATH_PRERELEASE_VER==1]
+*/
+#define TTMATH_MAJOR_VER		0
+
+/*!
+	the minor version of the library
+
+	the version present to the end user is constructed in this way:
+
+		TTMATH_MAJOR_VER.TTMATH_MINOR_VER.TTMATH_REVISION_VER.[prerelease if TTMATH_PRERELEASE_VER==1]
+*/
+#define TTMATH_MINOR_VER		9
+
+/*!
+	the revision version of the library
+
+	the version present to the end user is constructed in this way:
+
+		TTMATH_MAJOR_VER.TTMATH_MINOR_VER.TTMATH_REVISION_VER.[prerelease if TTMATH_PRERELEASE_VER==1]
+*/
+#define TTMATH_REVISION_VER		4
+
+/*!
+	TTMATH_PRERELEASE_VER is either zero or one
+	zero means that this is the release version of the library
+	(one means something like beta)
+
+	the version present to the end user is constructed in this way:
+
+		TTMATH_MAJOR_VER.TTMATH_MINOR_VER.TTMATH_REVISION_VER.[prerelease if TTMATH_PRERELEASE_VER==1]
+*/
+#define TTMATH_PRERELEASE_VER	1
+
+
+
+/*!
+	you can define a platform explicitly by defining either
+	TTMATH_PLATFORM32 or TTMATH_PLATFORM64 macro
+*/
+#if !defined TTMATH_PLATFORM32 && !defined TTMATH_PLATFORM64
+
+	#if !defined _M_X64 && !defined __x86_64__
+
+		/*
+			other platforms than x86 and amd64 are not recognized at the moment
+			so you should set TTMATH_PLATFORMxx manually
+		*/
+
+		// we're using a 32bit platform
+		#define TTMATH_PLATFORM32
+
+	#else
+
+		//	we're using a 64bit platform
+		#define TTMATH_PLATFORM64
+
+	#endif
+
+#endif
+
+
+/*!
+	asm version of the library is available by default only for:
+	x86 and amd64 platforms and for Microsoft Visual and GCC compilers
+
+	but you can force using asm version (the same asm as for Microsoft Visual)
+	by defining TTMATH_FORCEASM macro
+	you have to be sure that your compiler accept such an asm format
+*/
+#ifndef TTMATH_FORCEASM
+
+	#if !defined __i386__  && !defined _X86_ && !defined  _M_IX86 && !defined __x86_64__  && !defined _M_X64
+		/*!
+			x86 architecture:
+			__i386__    defined by GNU C
+			_X86_  	    defined by MinGW32
+			_M_IX86     defined by Visual Studio, Intel C/C++, Digital Mars and Watcom C/C++
+
+			amd64 architecture:
+			__x86_64__  defined by GNU C, CLANG (LLVM) and Sun Studio
+			_M_X64  	defined by Visual Studio
+
+			asm version is available only for x86 or amd64 platforms
+		*/
+		#define TTMATH_NOASM
+	#endif
+
+
+
+	#if !defined _MSC_VER && !defined __GNUC__
+		/*!
+			another compilers than MS VC or GCC or CLANG (LLVM) by default use no asm version
+			(CLANG defines __GNUC__ too)
+		*/
+		#define TTMATH_NOASM
+	#endif
+
+#endif
+
+
 namespace ttmath
 {
 
-#if !defined _M_X64 && !defined __x86_64__
-
-	/*!
-		we're using a 32bit platform
-	*/
-	#define TTMATH_PLATFORM32
-
-#else
-
-	/*!
-		we're using a 64bit platform
-	*/
-	#define TTMATH_PLATFORM64
-
-#endif
-
-
 
 #ifdef TTMATH_PLATFORM32
 
@@ -121,17 +181,18 @@ namespace ttmath
 	typedef signed   int sint;
 
 	/*!
-		this type is twice bigger than uint
-		(64bit on a 32bit platforms)
-
-		although C++ Standard - ANSI ISO IEC 14882:2003 doesn't define such a type (long long) 
-		but it is defined in C99 and in upcoming C++0x /3.9.1 (2)/ and many compilers support it
-
-		this type is used in UInt::MulTwoWords and UInt::DivTwoWords when macro TTMATH_NOASM is defined
-		but only on a 32bit platform
+		on 32 bit platform ulint and slint will be equal 64 bits
 	*/
-	#ifdef TTMATH_NOASM
+	#ifdef _MSC_VER
+		// long long on MS Windows (Visual and GCC mingw compilers) have 64 bits
+		// stdint.h is not available on Visual Studio prior to VS 2010 version
 		typedef unsigned long long int ulint;
+		typedef signed   long long int slint;
+	#else
+		// we do not use 'long' here because there is a difference in unix and windows
+		// environments: in unix 'long' has 64 bits but in windows it has only 32 bits
+		typedef uint64_t ulint;
+		typedef int64_t  slint;
 	#endif
 
 	/*!
@@ -157,22 +218,34 @@ namespace ttmath
 	*/
 	#define TTMATH_BUILTIN_VARIABLES_SIZE 256u
 
+	/*!
+		this macro returns the number of machine words 
+		capable to hold min_bits bits
+		e.g. TTMATH_BITS(128) returns 4
+	*/
+	#define TTMATH_BITS(min_bits) ((min_bits-1)/32 + 1)
+
 #else
 
-	/*!
-		on 64bit platforms one word (uint, sint) will be equal 64bits
-	*/
-	typedef unsigned long uint;
-	typedef signed   long sint;
+	#ifdef _MSC_VER
+		/* in VC 'long' type has 32 bits, __int64 is VC extension */
+		typedef unsigned __int64 uint;
+		typedef signed   __int64 sint;
+	#else
+		/*!
+			on 64bit platforms one word (uint, sint) will be equal 64bits
+		*/
+		typedef uint64_t uint;
+
+		/*!
+			on 64bit platforms one word (uint, sint) will be equal 64bits
+		*/
+		typedef int64_t sint;
+	#endif 
 
 	/*!
-		on 64bit platform we do not define ulint
-		sizeof(long long) is 8 (64bit) but we need 128bit
-
-		on 64 bit platform (when there is defined TTMATH_NOASM macro)
-		methods UInt::MulTwoWords and UInt::DivTwoWords are using other algorithms than those on 32 bit
+		on 64bit platforms we do not define ulint and slint
 	*/
-	//typedef unsigned long long int ulint;
 
 	/*!
 		how many bits there are in the uint type
@@ -197,27 +270,28 @@ namespace ttmath
 	*/
 	#define TTMATH_BUILTIN_VARIABLES_SIZE 128ul
 
+	/*!
+		this macro returns the number of machine words 
+		capable to hold min_bits bits
+		e.g. TTMATH_BITS(128) returns 2
+	*/
+	#define TTMATH_BITS(min_bits) ((min_bits-1)/64 + 1)
+
 #endif
 }
 
 
+#if defined(TTMATH_MULTITHREADS) && !defined(TTMATH_MULTITHREADS_NOSYNC)
+	#if !defined(TTMATH_POSIX_THREADS) && !defined(TTMATH_WIN32_THREADS)
 
-/*!
-	characters which represent the comma operator
+		#if defined(_WIN32)
+			#define TTMATH_WIN32_THREADS
+		#elif defined(unix) || defined(__unix__) || defined(__unix)
+			#define TTMATH_POSIX_THREADS
+		#endif
 
-	TTMATH_COMMA_CHARACTER_1 is used in reading (parsing) and in writing (default, can be overwritten in ToString() function)
-	TTMATH_COMMA_CHARACTER_2 can be used in reading as an auxiliary comma character
-	that means you can input values for example 1.2345 and 1,2345 as well
-
-	if you don't want it just put 0 there e.g.
-		#define TTMATH_COMMA_CHARACTER_2 0
-	then only TTMATH_COMMA_CHARACTER_1 will be used
-
-	don't put there any special character which is used by the parser
-	(for example a semicolon ';' shouldn't be there)
-*/
-#define TTMATH_COMMA_CHARACTER_1 '.'
-#define TTMATH_COMMA_CHARACTER_2 ','
+	#endif
+#endif
 
 
 
@@ -242,16 +316,54 @@ namespace ttmath
 	if the size of a vector is smaller than TTMATH_USE_KARATSUBA_MULTIPLICATION_FROM_SIZE
 	the Karatsuba algorithm will use standard schoolbook multiplication
 */
-#ifdef __GNUC__
-#define TTMATH_USE_KARATSUBA_MULTIPLICATION_FROM_SIZE 3
+#ifdef TTMATH_DEBUG_LOG
+	// if TTMATH_DEBUG_LOG is defined then we should use the same size regardless of the compiler
+	#define TTMATH_USE_KARATSUBA_MULTIPLICATION_FROM_SIZE 3
 #else
-#define TTMATH_USE_KARATSUBA_MULTIPLICATION_FROM_SIZE 5
+	#ifdef __GNUC__
+		#define TTMATH_USE_KARATSUBA_MULTIPLICATION_FROM_SIZE 3
+	#else
+		#define TTMATH_USE_KARATSUBA_MULTIPLICATION_FROM_SIZE 5
+	#endif
 #endif
 
 
+/*!
+	this is a special value used when calculating the Gamma(x) function
+	if x is greater than this value then the Gamma(x) will be calculated using
+	some kind of series
+
+	don't use smaller values than about 100
+*/
+#define TTMATH_GAMMA_BOUNDARY 2000
+
+
+
+
+
 namespace ttmath
 {
 
+	/*!
+		lib type codes:
+		-  asm_vc_32   - with asm code designed for Microsoft Visual C++ (32 bits)
+		-  asm_gcc_32  - with asm code designed for GCC (32 bits)
+		-  asm_vc_64   - with asm for VC (64 bit)
+		-  asm_gcc_64  - with asm for GCC (64 bit)
+		-  no_asm_32   - pure C++ version (32 bit) - without any asm code
+		-  no_asm_64   - pure C++ version (64 bit) - without any asm code
+	*/
+	enum LibTypeCode
+	{
+	  asm_vc_32 = 0,
+	  asm_gcc_32,
+	  asm_vc_64,
+	  asm_gcc_64,
+	  no_asm_32,
+	  no_asm_64
+	};
+
+
 	/*!
 		error codes
 	*/
@@ -282,11 +394,142 @@ namespace ttmath
 		err_object_exists,
 		err_unknown_object,
 		err_still_calculating,
-		err_too_big_factorial,
-		err_in_short_form_used_function
+		err_in_short_form_used_function,
+		err_percent_from,
+		err_assignment_requires_variable
 	};
 
 
+	/*!
+		this struct is used when converting to/from a string
+		/temporarily only in Big::ToString() and Big::FromString()/
+	*/
+	struct Conv
+	{
+		/*!
+			base (radix) on which the value will be shown (or read)
+			default: 10
+		*/
+		uint base;
+
+
+		/*!
+			used only in Big::ToString()
+			if true the value will be always shown in the scientific mode, e.g: 123e+30
+			default: false
+		*/
+		bool scient;
+
+
+		/*!
+			used only in Big::ToString()
+			if scient is false then the value will be printed in the scientific mode
+			only if the exponent is greater than scien_from
+			default: 15
+		*/
+		sint scient_from;
+
+
+		/*!
+			if 'base_round' is true and 'base' is different from 2, 4, 8, or 16
+			and the result value is not an integer then we make an additional rounding
+			(after converting the last digit from the result is skipped)
+			default: true
+
+			e.g.
+
+				Conv c;
+				c.base_round = false;
+				Big<1, 1> a = "0.1";                       // decimal input
+				std::cout << a.ToString(c) << std::endl;   // the result is: 0.099999999
+		*/
+		bool base_round;
+
+
+		/*!
+			used only in Big::ToString()
+			tells how many digits after comma are possible
+			default: -1 which means all digits are printed
+
+			set it to zero if you want integer value only
+
+			for example when the value is:
+				12.345678 and 'round' is 4
+			then the result will be 
+				12.3457   (the last digit was rounded)
+		*/
+		sint round;
+
+
+		/*!
+			if true that not mattered digits in the mantissa will be cut off
+			(zero characters at the end -- after the comma operator)
+			e.g. 1234,78000 will be: 1234,78
+			default: true
+		*/
+		bool trim_zeroes;
+
+
+		/*!
+			the main comma operator (used when reading and writing)
+			default is a dot '.'
+		*/
+		uint comma;
+
+
+		/*!
+			additional comma operator (used only when reading) 
+			if you don't want it just set it to zero
+			default is a comma ','
+
+			this allowes you to convert from a value:
+			123.45 as well as from 123,45
+		*/
+		uint comma2;
+
+
+		/*!
+			it sets the character which is used for grouping
+			if group=' ' then: 1234,56789 will be printed as: 1 234,567 89
+
+			if you don't want grouping just set it to zero (which is default)
+		*/
+		uint group;
+
+
+		/*!
+			how many digits should be grouped (it is used if 'group' is non zero)
+			default: 3
+		*/
+		uint group_digits;
+
+
+		/*!
+		*/
+		uint group_exp; // not implemented yet
+
+
+
+
+		Conv()
+		{
+			// default values
+			base         = 10;
+			scient       = false;
+			scient_from  = 15;
+			base_round   = true;
+			round        = -1;
+			trim_zeroes  = true;
+			comma        = '.';
+			comma2       = ',';
+			group        = 0;
+			group_digits = 3;
+			group_exp    = 0;
+		}
+	};
+
+
+
 	/*!
 		this simple class can be used in multithreading model
 		(you can write your own class derived from this one)
@@ -339,37 +582,24 @@ namespace ttmath
 		In the library is used macro TTMATH_REFERENCE_ASSERT which
 		can throw an exception of this type
 
+		** from version 0.9.2 this macro is removed from all methods
+		   in public interface so you don't have to worry about it **
+
 		If you compile with gcc you can get a small benefit 
-		from using method Where() (it returns std::string with
+		from using method Where() (it returns std::string) with
 		the name and the line of a file where the macro TTMATH_REFERENCE_ASSERT
 		was used)
-
-		What is the 'reference' error?
-		Some kind of methods use a reference as their argument to another object,
-		and the another object not always can be the same which is calling, e.g.
-			Big<1,2> foo(10);
-			foo.Mul(foo); // this is incorrect
-		above method Mul is making something more with 'this' object and 
-		'this' cannot be passed as the argument because the result will be undefined
-
-		macro TTMATH_REFERENCE_ASSERT helps us to solve the above problem
-
-		note! some methods can use 'this' object as the argument
-		for example this code is correct:
-			UInt<2> foo(10);
-			foo.Add(foo);
-		but there are only few methods which can do that
 	*/
 	class ReferenceError : public std::logic_error, public ExceptionInfo
 	{
 	public:
 
-		ReferenceError() : std::logic_error ("reference error")
+		ReferenceError() : std::logic_error("reference error")
 		{
 		}
 
 		ReferenceError(const char * f, int l) :
-							std::logic_error ("reference error"), ExceptionInfo(f,l)
+							std::logic_error("reference error"), ExceptionInfo(f,l)
 		{
 		}
 
@@ -388,7 +618,7 @@ namespace ttmath
 		of this type
 
 		if you compile with gcc you can get a small benefit 
-		from using method Where() (it returns std::string with
+		from using method Where() (it returns std::string) with
 		the name and the line of a file where the macro TTMATH_ASSERT
 		was used)
 	*/
@@ -396,12 +626,12 @@ namespace ttmath
 	{
 	public:
 
-		RuntimeError() : std::runtime_error ("internal error")
+		RuntimeError() : std::runtime_error("internal error")
 		{
 		}
 
 		RuntimeError(const char * f, int l) :
-						std::runtime_error ("internal error"), ExceptionInfo(f,l)
+						std::runtime_error("internal error"), ExceptionInfo(f,l)
 		{
 		}
 
@@ -414,8 +644,21 @@ namespace ttmath
 
 
 	/*!
-		look at the description of macros TTMATH_RELEASE and TTMATH_DEBUG
+		TTMATH_DEBUG
+		this macro enables further testing during writing your code
+		you don't have to define it in a release mode
+
+		if this macro is set then macros TTMATH_ASSERT and TTMATH_REFERENCE_ASSERT
+		are set as well	and these macros can throw an exception if a condition in it
+		is not fulfilled (look at the definition of TTMATH_ASSERT and TTMATH_REFERENCE_ASSERT)
+
+		TTMATH_DEBUG is set automatically if DEBUG or _DEBUG are defined
 	*/
+	#if defined DEBUG || defined _DEBUG
+		#define TTMATH_DEBUG
+	#endif
+
+
 	#ifdef TTMATH_DEBUG
 
 		#if defined(__FILE__) && defined(__LINE__)
@@ -443,18 +686,22 @@ namespace ttmath
 
 
 	#ifdef TTMATH_DEBUG_LOG
-
-		#define TTMATH_LOG(msg) \
-			PrintLog(msg, std::cout);		
-
+		#define TTMATH_LOG(msg)                             PrintLog(msg, std::cout);
+		#define TTMATH_LOGC(msg, carry)                     PrintLog(msg, carry, std::cout);
+		#define TTMATH_VECTOR_LOG(msg, vector, len)         PrintVectorLog(msg, std::cout, vector, len);
+		#define TTMATH_VECTOR_LOGC(msg, carry, vector, len) PrintVectorLog(msg, carry, std::cout, vector, len);
 	#else
-
 		#define TTMATH_LOG(msg)
-
+		#define TTMATH_LOGC(msg, carry)
+		#define TTMATH_VECTOR_LOG(msg, vector, len)
+		#define TTMATH_VECTOR_LOGC(msg, carry, vector, len)
 	#endif
 
 
+
+
 } // namespace
 
 
 #endif
+

ttmath/ttmathuint_noasm.h

@@ -1,11 +1,11 @@
 /*
  * This file is a part of TTMath Bignum Library
- * and is distributed under the (new) BSD licence.
- * Author: Tomasz Sowa <t.sowa@slimaczek.pl>
+ * and is distributed under the 3-Clause BSD Licence.
+ * Author: Tomasz Sowa <t.sowa@ttmath.org>
  */
 
 /* 
- * Copyright (c) 2006-2009, Tomasz Sowa
+ * Copyright (c) 2006-2010, Tomasz Sowa
  * All rights reserved.
  * 
  * Redistribution and use in source and binary forms, with or without
@@ -39,18 +39,69 @@
 #define headerfilettmathuint_noasm
 
 
-#ifdef TTMATH_NOASM
-
 /*!
 	\file ttmathuint_noasm.h
-    \brief template class UInt<uint> with methods without any assembler code
+    \brief template class UInt<uint> with methods without any assembler code (used for no-asm version of ttmath)
 
 	this file is included at the end of ttmathuint.h
 */
 
+#ifdef TTMATH_NOASM
+
+
 
 namespace ttmath
 {
+
+	/*!
+		returning the string represents the currect type of the library
+		we have following types:
+		-  asm_vc_32   - with asm code designed for Microsoft Visual C++ (32 bits)
+		-  asm_gcc_32  - with asm code designed for GCC (32 bits)
+		-  asm_vc_64   - with asm for VC (64 bit)
+		-  asm_gcc_64  - with asm for GCC (64 bit)
+		-  no_asm_32   - pure C++ version (32 bit) - without any asm code
+		-  no_asm_64   - pure C++ version (64 bit) - without any asm code
+	*/
+	template<uint value_size>
+	const char * UInt<value_size>::LibTypeStr()
+	{
+		#ifdef TTMATH_PLATFORM32
+			static const char info[] = "no_asm_32";
+		#endif		
+
+		#ifdef TTMATH_PLATFORM64
+			static const char info[] = "no_asm_64";
+		#endif
+
+	return info;
+	}
+
+	
+	/*!
+		returning the currect type of the library
+	*/
+	template<uint value_size>
+	LibTypeCode UInt<value_size>::LibType()
+	{
+		#ifdef TTMATH_PLATFORM32
+			LibTypeCode info = no_asm_32;
+		#endif		
+
+		#ifdef TTMATH_PLATFORM64
+			LibTypeCode info = no_asm_64;
+		#endif
+
+	return info;
+	}
+
+
+	/*!
+		this method adds two words together
+		returns carry
+
+		this method is created only when TTMATH_NOASM macro is defined
+	*/
 	template<uint value_size>
 	uint UInt<value_size>::AddTwoWords(uint a, uint b, uint carry, uint * result)
 	{
@@ -95,7 +146,7 @@ namespace ttmath
 		for(i=0 ; i<value_size ; ++i)
 			c = AddTwoWords(table[i], ss2.table[i], c, &table[i]);
 
-		TTMATH_LOG("UInt::Add")
+		TTMATH_LOGC("UInt::Add", c)
 	
 	return c;
 	}
@@ -106,12 +157,17 @@ namespace ttmath
 		and returns a carry (if it was)
 
 		if we've got (value_size=3):
+
 			table[0] = 10;
 			table[1] = 30;
-			table[2] = 5;	
+			table[2] = 5;
+
 		and we call:
+
 			AddInt(2,1)
+
 		then it'll be:
+
 			table[0] = 10;
 			table[1] = 30 + 2;
 			table[2] = 5;
@@ -131,7 +187,7 @@ namespace ttmath
 		for(i=index+1 ; i<value_size && c ; ++i)
 			c = AddTwoWords(table[i], 0, c, &table[i]);
 
-		TTMATH_LOG("UInt::AddInt")
+		TTMATH_LOGC("UInt::AddInt", c)
 	
 	return c;
 	}
@@ -149,17 +205,23 @@ namespace ttmath
 		x1 - lower word, x2 - higher word
 
 		for example if we've got value_size equal 4 and:
+
 			table[0] = 3
 			table[1] = 4
 			table[2] = 5
 			table[3] = 6
+
 		then let
+
 			x1 = 10
 			x2 = 20
+
 		and
+
 			index = 1
 
 		the result of this method will be:
+
 			table[0] = 3
 			table[1] = 4 + x1 = 14
 			table[2] = 5 + x2 = 25
@@ -184,7 +246,7 @@ namespace ttmath
 		for(i=index+2 ; i<value_size && c ; ++i)
 			c = AddTwoWords(table[i], 0, c, &table[i]);
 
-		TTMATH_LOG("UInt::AddTwoInts")
+		TTMATH_LOGC("UInt::AddTwoInts", c)
 	
 	return c;
 	}
@@ -195,19 +257,20 @@ namespace ttmath
 		this static method addes one vector to the other
 		'ss1' is larger in size or equal to 'ss2'
 
-		ss1 points to the first (larger) vector
-		ss2 points to the second vector
-		ss1_size - size of the ss1 (and size of the result too)
-		ss2_size - size of the ss2
-		result - is the result vector (which has size the same as ss1: ss1_size)
+		-  ss1 points to the first (larger) vector
+		-  ss2 points to the second vector
+		-  ss1_size - size of the ss1 (and size of the result too)
+		-  ss2_size - size of the ss2
+		-  result - is the result vector (which has size the same as ss1: ss1_size)
+
+			Example:  ss1_size is 5, ss2_size is 3
+			ss1:      ss2:   result (output):
+			  5        1         5+1
+			  4        3         4+3
+			  2        7         2+7
+			  6                  6
+			  9                  9
 
-		Example:  ss1_size is 5, ss2_size is 3
-		ss1:      ss2:   result (output):
-		  5        1         5+1
-		  4        3         4+3
-		  2        7         2+7
-		  6                  6
-		  9                  9
 	  of course the carry is propagated and will be returned from the last item
 	  (this method is used by the Karatsuba multiplication algorithm)
 	*/
@@ -224,7 +287,7 @@ namespace ttmath
 		for( ; i<ss1_size ; ++i)
 			c = AddTwoWords(ss1[i], 0, c, &result[i]);
 
-		TTMATH_LOG("UInt::AddVector")
+		TTMATH_VECTOR_LOGC("UInt::AddVector", c, result, ss1_size)
 
 	return c;
 	}
@@ -232,6 +295,12 @@ namespace ttmath
 
 
 
+	/*!
+		this method subtractes one word from the other
+		returns carry
+
+		this method is created only when TTMATH_NOASM macro is defined
+	*/
 	template<uint value_size>
 	uint UInt<value_size>::SubTwoWords(uint a, uint b, uint carry, uint * result)
 	{
@@ -273,7 +342,7 @@ namespace ttmath
 		for(i=0 ; i<value_size ; ++i)
 			c = SubTwoWords(table[i], ss2.table[i], c, &table[i]);
 
-		TTMATH_LOG("UInt::Sub")
+		TTMATH_LOGC("UInt::Sub", c)
 
 	return c;
 	}
@@ -286,12 +355,17 @@ namespace ttmath
 		and returns a carry (if it was)
 
 		if we've got (value_size=3):
+
 			table[0] = 10;
 			table[1] = 30;
 			table[2] = 5;	
+
 		and we call:
+
 			SubInt(2,1)
+
 		then it'll be:
+
 			table[0] = 10;
 			table[1] = 30 - 2;
 			table[2] = 5;
@@ -311,7 +385,7 @@ namespace ttmath
 		for(i=index+1 ; i<value_size && c ; ++i)
 			c = SubTwoWords(table[i], 0, c, &table[i]);
 
-		TTMATH_LOG("UInt::SubInt")
+		TTMATH_LOGC("UInt::SubInt", c)
 	
 	return c;
 	}
@@ -321,19 +395,19 @@ namespace ttmath
 		this static method subtractes one vector from the other
 		'ss1' is larger in size or equal to 'ss2'
 
-		ss1 points to the first (larger) vector
-		ss2 points to the second vector
-		ss1_size - size of the ss1 (and size of the result too)
-		ss2_size - size of the ss2
-		result - is the result vector (which has size the same as ss1: ss1_size)
+		-  ss1 points to the first (larger) vector
+		-  ss2 points to the second vector
+		-  ss1_size - size of the ss1 (and size of the result too)
+		-  ss2_size - size of the ss2
+		-  result - is the result vector (which has size the same as ss1: ss1_size)
 
-		Example:  ss1_size is 5, ss2_size is 3
-		ss1:      ss2:   result (output):
-		  5        1         5-1
-		  4        3         4-3
-		  2        7         2-7
-		  6                  6-1  (the borrow from previous item)
-		  9                  9
+			Example:  ss1_size is 5, ss2_size is 3
+			ss1:      ss2:   result (output):
+			  5        1         5-1
+			  4        3         4-3
+			  2        7         2-7
+			  6                  6-1  (the borrow from previous item)
+			  9                  9
 		                 return (carry): 0
 	  of course the carry (borrow) is propagated and will be returned from the last item
 	  (this method is used by the Karatsuba multiplication algorithm)
@@ -351,7 +425,7 @@ namespace ttmath
 		for( ; i<ss1_size ; ++i)
 			c = SubTwoWords(ss1[i], 0, c, &result[i]);
 
-		TTMATH_LOG("UInt::SubVector")
+		TTMATH_VECTOR_LOGC("UInt::SubVector", c, result, ss1_size)
 
 	return c;
 	}
@@ -385,7 +459,7 @@ namespace ttmath
 			c        = new_c;
 		}
 
-		TTMATH_LOG("UInt::Rcl2_one")
+		TTMATH_LOGC("UInt::Rcl2_one", c)
 
 	return c;
 	}
@@ -424,7 +498,9 @@ namespace ttmath
 			c        = new_c;
 		}
 
-		TTMATH_LOG("UInt::Rcr2_one")
+		c = (c != 0)? 1 : 0;
+
+		TTMATH_LOGC("UInt::Rcr2_one", c)
 
 	return c;
 	}
@@ -462,7 +538,7 @@ namespace ttmath
 			c        = new_c;
 		}
 
-		TTMATH_LOG("UInt::Rcl2")
+		TTMATH_LOGC("UInt::Rcl2", (c & 1))
 
 	return (c & 1);
 	}
@@ -501,18 +577,19 @@ namespace ttmath
 			c        = new_c;
 		}
 
-		TTMATH_LOG("UInt::Rcr2")
+		c = (c & TTMATH_UINT_HIGHEST_BIT) ? 1 : 0;
 
-	return (c & TTMATH_UINT_HIGHEST_BIT) ? 1 : 0;
+		TTMATH_LOGC("UInt::Rcr2", c)
+
+	return c;
 	}
 
 
 
 
-	/*
+	/*!
 		this method returns the number of the highest set bit in x
 		if the 'x' is zero this method returns '-1'
-
 	*/
 	template<uint value_size>
 	sint UInt<value_size>::FindLeadingBitInWord(uint x)
@@ -533,18 +610,41 @@ namespace ttmath
 
 
 
+	/*!
+		this method returns the number of the highest set bit in x
+		if the 'x' is zero this method returns '-1'
+	*/
+	template<uint value_size>
+	sint UInt<value_size>::FindLowestBitInWord(uint x)
+	{
+		if( x == 0 )
+			return -1;
+
+		uint bit = 0;
+		
+		while( (x & 1) == 0 )
+		{
+			x = x >> 1;
+			++bit;
+		}
+
+	return bit;
+	}
+
 
 
 	/*!
 		this method sets a special bit in the 'value'
 		and returns the last state of the bit (zero or one)
 
-		bit is from <0,63>
+		bit is from <0,TTMATH_BITS_PER_UINT-1>
 
 		e.g.
-		 uint x = 100;
-		 uint bit = SetBitInWord(x, 3);
-		 now: x = 108 and bit = 0
+
+			uint x = 100;
+			uint bit = SetBitInWord(x, 3);
+
+		now: x = 108 and bit = 0
 	*/
 	template<uint value_size>
 	uint UInt<value_size>::SetBitInWord(uint & value, uint bit)
@@ -553,7 +653,7 @@ namespace ttmath
 
 		uint mask = 1;
 
-		if( bit > 1 )
+		if( bit > 0 )
 			mask = mask << bit;
 
 		uint last = value & mask;
@@ -577,10 +677,11 @@ namespace ttmath
 
 	/*!
 		multiplication: result_high:result_low = a * b
-		result_high - higher word of the result
-		result_low  - lower word of the result
+		-  result_high - higher word of the result
+		-  result_low  - lower word of the result
 	
 		this methos never returns a carry
+
 		this method is used in the second version of the multiplication algorithms
 	*/
 	template<uint value_size>
@@ -741,11 +842,11 @@ namespace ttmath
 			temp1.u_.high = a_.u_.low;
 			temp1.u_.low  = b_.u_.high;
 
-			res_.u_.high  = temp1.u / c;
-			temp2.u_.high = temp1.u % c;
+			res_.u_.high  = (unsigned int)(temp1.u / c);
+			temp2.u_.high = (unsigned int)(temp1.u % c);
 			temp2.u_.low  = b_.u_.low;
 			
-			res_.u_.low  = temp2.u / c;
+			res_.u_.low  = (unsigned int)(temp2.u / c);
 			*rest        = temp2.u % c;
 
 			*r = res_.u;

ttmath/ttmathuint_x86.h

@@ -1,7 +1,7 @@
 /*
  * This file is a part of TTMath Bignum Library
- * and is distributed under the (new) BSD licence.
- * Author: Tomasz Sowa <t.sowa@slimaczek.pl>
+ * and is distributed under the 3-Clause BSD Licence.
+ * Author: Tomasz Sowa <t.sowa@ttmath.org>
  */
 
 /* 
@@ -35,16 +35,10 @@
  * THE POSSIBILITY OF SUCH DAMAGE.
  */
 
-
-
 #ifndef headerfilettmathuint_x86
 #define headerfilettmathuint_x86
 
 
-#ifndef TTMATH_NOASM
-#ifdef TTMATH_PLATFORM32
-
-
 /*!
 	\file ttmathuint_x86.h
     \brief template class UInt<uint> with assembler code for 32bit x86 processors
@@ -53,6 +47,12 @@
 */
 
 
+#ifndef TTMATH_NOASM
+#ifdef TTMATH_PLATFORM32
+
+
+
+
 
 /*!
     \brief a namespace for the TTMath library
@@ -60,6 +60,51 @@
 namespace ttmath
 {
 
+	/*!
+		returning the string represents the currect type of the library
+
+		we have following types:
+		-  asm_vc_32   - with asm code designed for Microsoft Visual C++ (32 bits)
+		-  asm_gcc_32  - with asm code designed for GCC (32 bits)
+		-  asm_vc_64   - with asm for VC (64 bit)
+		-  asm_gcc_64  - with asm for GCC (64 bit)
+		-  no_asm_32   - pure C++ version (32 bit) - without any asm code
+		-  no_asm_64   - pure C++ version (64 bit) - without any asm code
+	*/
+	template<uint value_size>
+	const char * UInt<value_size>::LibTypeStr()
+	{
+		#ifndef __GNUC__
+			static const char info[] = "asm_vc_32";
+		#endif		
+
+		#ifdef __GNUC__
+			static const char info[] = "asm_gcc_32";
+		#endif
+
+	return info;
+	}
+
+
+	/*!
+		returning the currect type of the library
+	*/
+	template<uint value_size>
+	LibTypeCode UInt<value_size>::LibType()
+	{
+		#ifndef __GNUC__
+			LibTypeCode info = asm_vc_32;
+		#endif		
+
+		#ifdef __GNUC__
+			LibTypeCode info = asm_gcc_32;
+		#endif
+
+	return info;
+	}
+
+
+
 	/*!
 	*
 	*	basic mathematic functions
@@ -152,7 +197,7 @@ namespace ttmath
 				: "cc", "memory" );
 		#endif
 
-		TTMATH_LOG("UInt::Add")
+		TTMATH_LOGC("UInt::Add", c)
 
 	return c;
 	}
@@ -166,12 +211,17 @@ namespace ttmath
 		e.g.
 
 		if we've got (value_size=3):
+
 			table[0] = 10;
 			table[1] = 30;
-			table[2] = 5;	
+			table[2] = 5;
+
 		and we call:
+
 			AddInt(2,1)
+
 		then it'll be:
+
 			table[0] = 10;
 			table[1] = 30 + 2;
 			table[2] = 5;
@@ -253,7 +303,7 @@ namespace ttmath
 
 		#endif
 	
-		TTMATH_LOG("UInt::AddInt")
+		TTMATH_LOGC("UInt::AddInt", c)
 
 	return c;
 	}
@@ -270,17 +320,23 @@ namespace ttmath
 		x1 - lower word, x2 - higher word
 
 		for example if we've got value_size equal 4 and:
+
 			table[0] = 3
 			table[1] = 4
 			table[2] = 5
 			table[3] = 6
+
 		then let
+
 			x1 = 10
 			x2 = 20
+
 		and
+
 			index = 1
 
 		the result of this method will be:
+
 			table[0] = 3
 			table[1] = 4 + x1 = 14
 			table[2] = 5 + x2 = 25
@@ -374,7 +430,7 @@ namespace ttmath
 
 		#endif
 
-		TTMATH_LOG("UInt::AddTwoInts")
+		TTMATH_LOGC("UInt::AddTwoInts", c)
 	
 	return c;
 	}
@@ -506,7 +562,7 @@ namespace ttmath
 
 		#endif
 
-		TTMATH_LOG("UInt::AddVector")
+		TTMATH_VECTOR_LOGC("UInt::AddVector", c, result, ss1_size)
 
 	return c;
 	}
@@ -594,7 +650,7 @@ namespace ttmath
 
 		#endif
 
-		TTMATH_LOG("UInt::Sub")
+		TTMATH_LOGC("UInt::Sub", c)
 
 	return c;
 	}
@@ -609,12 +665,17 @@ namespace ttmath
 		e.g.
 
 		if we've got (value_size=3):
+
 			table[0] = 10;
 			table[1] = 30;
 			table[2] = 5;	
+
 		and we call:
+
 			SubInt(2,1)
+
 		then it'll be:
+
 			table[0] = 10;
 			table[1] = 30 - 2;
 			table[2] = 5;
@@ -696,7 +757,7 @@ namespace ttmath
 
 		#endif
 		
-		TTMATH_LOG("UInt::SubInt")
+		TTMATH_LOGC("UInt::SubInt", c)
 	
 	return c;
 	}
@@ -834,7 +895,7 @@ namespace ttmath
 
 		#endif
 
-		TTMATH_LOG("UInt::SubVector")
+		TTMATH_VECTOR_LOGC("UInt::SubVector", c, result, ss1_size)
 
 	return c;
 	}
@@ -912,7 +973,7 @@ namespace ttmath
 
 		#endif
 
-		TTMATH_LOG("UInt::Rcl2_one")
+		TTMATH_LOGC("UInt::Rcl2_one", c)
 
 	return c;
 	}
@@ -984,7 +1045,7 @@ namespace ttmath
 
 		#endif
 
-		TTMATH_LOG("UInt::Rcr2_one")
+		TTMATH_LOGC("UInt::Rcr2_one", c)
 
 	return c;
 	}
@@ -1118,7 +1179,7 @@ namespace ttmath
 
 		#endif
 
-		TTMATH_LOG("UInt::Rcl2")
+		TTMATH_LOGC("UInt::Rcl2", c)
 
 	return c;
 	}
@@ -1252,7 +1313,7 @@ namespace ttmath
 
 		#endif
 
-		TTMATH_LOG("UInt::Rcr2")
+		TTMATH_LOGC("UInt::Rcr2", c)
 
 	return c;
 	}
@@ -1309,6 +1370,50 @@ namespace ttmath
 
 
 
+	/*
+		this method returns the number of the smallest set bit in one 32-bit word
+		if the 'x' is zero this method returns '-1'
+	*/
+	template<uint value_size>
+	sint UInt<value_size>::FindLowestBitInWord(uint x)
+	{
+	sint result;
+
+		#ifndef __GNUC__
+			__asm
+			{
+				push eax
+				push edx
+
+				mov edx,-1
+				bsf eax,[x]
+				cmovz eax,edx
+				mov [result], eax
+
+				pop edx
+				pop eax
+			}
+		#endif
+
+
+		#ifdef __GNUC__
+		uint dummy;
+
+				__asm__ (
+
+				"movl $-1, %1          \n"
+				"bsfl %2, %0           \n"
+				"cmovz %1, %0          \n"
+
+				: "=r" (result), "=&r" (dummy)
+				: "r" (x)
+				: "cc" );
+
+		#endif
+
+	return result;
+	}
+
 
 
 	/*!
@@ -1317,9 +1422,10 @@ namespace ttmath
 
 		bit is from <0,31>
 		e.g.
-		 uint x = 100;
-		 uint bit = SetBitInWord(x, 3);
-		 now: x = 108 and bit = 0
+
+			uint x = 100;
+			uint bit = SetBitInWord(x, 3);
+			now: x = 108 and bit = 0
 	*/
 	template<uint value_size>
 	uint UInt<value_size>::SetBitInWord(uint & value, uint bit)

ttmath/ttmathuint_x86_64.h

@@ -1,11 +1,11 @@
 /*
  * This file is a part of TTMath Bignum Library
- * and is distributed under the (new) BSD licence.
- * Author: Tomasz Sowa <t.sowa@slimaczek.pl>
+ * and is distributed under the 3-Clause BSD Licence.
+ * Author: Tomasz Sowa <t.sowa@ttmath.org>
  */
 
 /* 
- * Copyright (c) 2006-2009, Tomasz Sowa
+ * Copyright (c) 2006-2010, Tomasz Sowa
  * All rights reserved.
  * 
  * Redistribution and use in source and binary forms, with or without
@@ -52,9 +52,91 @@
 */
 
 
+/*!
+	\file ttmathuint_x86_64_msvc.asm
+	\brief some asm routines for x86_64 when using Microsoft compiler
+
+	this file should be first compiled:
+	- compile with debug info:    ml64.exe /c /Zd /Zi ttmathuint_x86_64_msvc.asm
+	- compile without debug info: ml64.exe /c ttmathuint_x86_64_msvc.asm
+
+	this creates ttmathuint_x86_64_msvc.obj file which can be linked with your program
+
+	(you can use win64_assemble.bat file from ttmath subdirectory)
+*/
+
+
+#ifndef __GNUC__
+#include <intrin.h>
+#endif
+
+
 namespace ttmath
 {
 
+	#ifndef __GNUC__
+
+		extern "C"
+			{
+			uint __fastcall ttmath_adc_x64(uint* p1, const uint* p2, uint nSize, uint c);
+			uint __fastcall ttmath_addindexed_x64(uint* p1, uint nSize, uint nPos, uint nValue);
+			uint __fastcall ttmath_addindexed2_x64(uint* p1, uint nSize, uint nPos, uint nValue1, uint nValue2);
+			uint __fastcall ttmath_addvector_x64(const uint * ss1, const uint * ss2, uint ss1_size, uint ss2_size, uint * result);
+			uint __fastcall ttmath_sbb_x64(uint* p1, const uint* p2, uint nSize, uint c);
+			uint __fastcall ttmath_subindexed_x64(uint* p1, uint nSize, uint nPos, uint nValue);
+			uint __fastcall ttmath_subvector_x64(const uint * ss1, const uint * ss2, uint ss1_size, uint ss2_size, uint * result);
+			uint __fastcall ttmath_rcl_x64(uint* p1, uint nSize, uint nLowestBit);
+			uint __fastcall ttmath_rcr_x64(uint* p1, uint nSize, uint nLowestBit);
+			uint __fastcall ttmath_div_x64(uint* pnValHi, uint* pnValLo, uint nDiv);
+			uint __fastcall ttmath_rcl2_x64(uint* p1, uint nSize, uint nBits, uint c);
+			uint __fastcall ttmath_rcr2_x64(uint* p1, uint nSize, uint nBits, uint c);
+			};
+	#endif
+
+
+	/*!
+		returning the string represents the currect type of the library
+		we have following types:
+		  asm_vc_32   - with asm code designed for Microsoft Visual C++ (32 bits)
+		  asm_gcc_32  - with asm code designed for GCC (32 bits)
+		  asm_vc_64   - with asm for VC (64 bit)
+		  asm_gcc_64  - with asm for GCC (64 bit)
+		  no_asm_32   - pure C++ version (32 bit) - without any asm code
+		  no_asm_64   - pure C++ version (64 bit) - without any asm code
+	*/
+	template<uint value_size>
+	const char * UInt<value_size>::LibTypeStr()
+	{
+		#ifndef __GNUC__
+			static const char info[] = "asm_vc_64";
+		#endif		
+
+		#ifdef __GNUC__
+			static const char info[] = "asm_gcc_64";
+		#endif
+
+	return info;
+	}
+
+
+	/*!
+		returning the currect type of the library
+	*/
+	template<uint value_size>
+	LibTypeCode UInt<value_size>::LibType()
+	{
+		#ifndef __GNUC__
+			LibTypeCode info = asm_vc_64;
+		#endif		
+
+		#ifdef __GNUC__
+			LibTypeCode info = asm_gcc_64;
+		#endif
+
+	return info;
+	}
+
+
 	/*!
 	*
 	*	basic mathematic functions
@@ -83,7 +165,7 @@ namespace ttmath
 		// this algorithm doesn't require it
 
 		#ifndef __GNUC__
-			#error "another compiler than GCC is currently not supported in 64bit mode"
+			c = ttmath_adc_x64(p1,p2,b,c);
 		#endif
 
 		#ifdef __GNUC__
@@ -113,7 +195,7 @@ namespace ttmath
 
 		#endif
 
-		TTMATH_LOG("UInt::Add")
+		TTMATH_LOGC("UInt::Add", c)
 	
 	return c;
 	}
@@ -128,12 +210,17 @@ namespace ttmath
 
 
 		if we've got (value_size=3):
+
 			table[0] = 10;
 			table[1] = 30;
-			table[2] = 5;	
+			table[2] = 5;
+
 		and we call:
+
 			AddInt(2,1)
+
 		then it'll be:
+
 			table[0] = 10;
 			table[1] = 30 + 2;
 			table[2] = 5;
@@ -150,9 +237,10 @@ namespace ttmath
 		TTMATH_ASSERT( index < value_size )
 
 		#ifndef __GNUC__
-			#error "another compiler than GCC is currently not supported in 64bit mode"
+			c = ttmath_addindexed_x64(p1,b,index,value);
 		#endif
 
+
 		#ifdef __GNUC__
 		uint dummy, dummy2;
 	
@@ -179,7 +267,7 @@ namespace ttmath
 
 		#endif
 
-		TTMATH_LOG("UInt::AddInt")
+		TTMATH_LOGC("UInt::AddInt", c)
 	
 	return c;
 	}
@@ -197,17 +285,23 @@ namespace ttmath
 		x1 - lower word, x2 - higher word
 
 		for example if we've got value_size equal 4 and:
+
 			table[0] = 3
 			table[1] = 4
 			table[2] = 5
 			table[3] = 6
+
 		then let
+
 			x1 = 10
 			x2 = 20
+
 		and
+
 			index = 1
 
 		the result of this method will be:
+
 			table[0] = 3
 			table[1] = 4 + x1 = 14
 			table[2] = 5 + x2 = 25
@@ -228,9 +322,10 @@ namespace ttmath
 		TTMATH_ASSERT( index < value_size - 1 )
 
 		#ifndef __GNUC__
-			#error "another compiler than GCC is currently not supported in 64bit mode"
+			c = ttmath_addindexed2_x64(p1,b,index,x1,x2);
 		#endif
 
+
 		#ifdef __GNUC__
 		uint dummy, dummy2;
 
@@ -261,7 +356,7 @@ namespace ttmath
 
 		#endif
 
-		TTMATH_LOG("UInt::AddTwoInts")
+		TTMATH_LOGC("UInt::AddTwoInts", c)
 
 	return c;
 	}
@@ -272,19 +367,19 @@ namespace ttmath
 		this static method addes one vector to the other
 		'ss1' is larger in size or equal to 'ss2'
 
-		ss1 points to the first (larger) vector
-		ss2 points to the second vector
-		ss1_size - size of the ss1 (and size of the result too)
-		ss2_size - size of the ss2
-		result - is the result vector (which has size the same as ss1: ss1_size)
+		-  ss1 points to the first (larger) vector
+		-  ss2 points to the second vector
+		-  ss1_size - size of the ss1 (and size of the result too)
+		-  ss2_size - size of the ss2
+		-  result - is the result vector (which has size the same as ss1: ss1_size)
 
-		Example:  ss1_size is 5, ss2_size is 3
-		ss1:      ss2:   result (output):
-		  5        1         5+1
-		  4        3         4+3
-		  2        7         2+7
-		  6                  6
-		  9                  9
+			Example:  ss1_size is 5, ss2_size is 3
+			ss1:      ss2:   result (output):
+		  	  5        1         5+1
+			  4        3         4+3
+			  2        7         2+7
+			  6                  6
+			  9                  9
 	  of course the carry is propagated and will be returned from the last item
 	  (this method is used by the Karatsuba multiplication algorithm)
 	*/
@@ -293,15 +388,16 @@ namespace ttmath
 	{
 		TTMATH_ASSERT( ss1_size >= ss2_size )
 
-		uint rest = ss1_size - ss2_size;
 		uint c;
 
 		#ifndef __GNUC__
-			#error "another compiler than GCC is currently not supported in 64bit mode"
+			 c = ttmath_addvector_x64(ss1, ss2, ss1_size, ss2_size, result);
 		#endif
 
+
 		#ifdef __GNUC__
 		uint dummy1, dummy2, dummy3;	
+		uint rest = ss1_size - ss2_size;
 			
 			//	this part should be compiled with gcc
 		
@@ -343,15 +439,13 @@ namespace ttmath
 
 		#endif
 
-		TTMATH_LOG("UInt::AddVector")
+		TTMATH_VECTOR_LOGC("UInt::AddVector", c, result, ss1_size)
 
 	return c;
 	}
 
 
 
-
-
 	/*!
 		this method's subtracting ss2 from the 'this' and subtracting
 		carry if it has been defined
@@ -369,14 +463,14 @@ namespace ttmath
 	uint * p1 = table;
 	const uint * p2 = ss2.table;
 	
-
 		// we don't have to use TTMATH_REFERENCE_ASSERT here
 		// this algorithm doesn't require it
 
 		#ifndef __GNUC__
-			#error "another compiler than GCC is currently not supported in 64bit mode"
+			c = ttmath_sbb_x64(p1,p2,b,c);
 		#endif
 
+
 		#ifdef __GNUC__
 		uint dummy, dummy2;
 
@@ -401,12 +495,13 @@ namespace ttmath
 
 		#endif
 
-		TTMATH_LOG("UInt::Sub")
+		TTMATH_LOGC("UInt::Sub", c)
 
 	return c;
 	}
 
 
+
 	/*!
 		this method subtracts one word (at a specific position)
 		and returns a carry (if it was)
@@ -414,12 +509,17 @@ namespace ttmath
 		***this method is created only on a 64bit platform***
 
 		if we've got (value_size=3):
+
 			table[0] = 10;
 			table[1] = 30;
 			table[2] = 5;	
+
 		and we call:
+
 			SubInt(2,1)
+
 		then it'll be:
+
 			table[0] = 10;
 			table[1] = 30 - 2;
 			table[2] = 5;
@@ -432,15 +532,17 @@ namespace ttmath
 	uint b = value_size;
 	uint * p1 = table;
 	uint c;
-	uint dummy, dummy2;
 
 		TTMATH_ASSERT( index < value_size )
 
 		#ifndef __GNUC__
-			#error "another compiler than GCC is currently not supported in 64bit mode"
+			c = ttmath_subindexed_x64(p1,b,index,value);
 		#endif
 
+
 		#ifdef __GNUC__
+			uint dummy, dummy2;
+
 			__asm__ __volatile__(
 			
 				"subq %%rdx, %%rcx 				\n"
@@ -464,31 +566,29 @@ namespace ttmath
 
 		#endif
 
-		TTMATH_LOG("UInt::SubInt")
+		TTMATH_LOGC("UInt::SubInt", c)
 
 	return c;
 	}
 
 
-
-
 	/*!
 		this static method subtractes one vector from the other
 		'ss1' is larger in size or equal to 'ss2'
 
-		ss1 points to the first (larger) vector
-		ss2 points to the second vector
-		ss1_size - size of the ss1 (and size of the result too)
-		ss2_size - size of the ss2
-		result - is the result vector (which has size the same as ss1: ss1_size)
+		-  ss1 points to the first (larger) vector
+		-  ss2 points to the second vector
+		-  ss1_size - size of the ss1 (and size of the result too)
+		-  ss2_size - size of the ss2
+		-  result - is the result vector (which has size the same as ss1: ss1_size)
 
-		Example:  ss1_size is 5, ss2_size is 3
-		ss1:      ss2:   result (output):
-		  5        1         5-1
-		  4        3         4-3
-		  2        7         2-7
-		  6                  6-1  (the borrow from previous item)
-		  9                  9
+			Example:  ss1_size is 5, ss2_size is 3
+			ss1:      ss2:   result (output):
+			  5        1         5-1
+			  4        3         4-3
+			  2        7         2-7
+			  6                  6-1  (the borrow from previous item)
+			  9                  9
 		               return (carry): 0
 	  of course the carry (borrow) is propagated and will be returned from the last item
 	  (this method is used by the Karatsuba multiplication algorithm)
@@ -498,20 +598,20 @@ namespace ttmath
 	{
 		TTMATH_ASSERT( ss1_size >= ss2_size )
 
-		uint rest = ss1_size - ss2_size;
 		uint c;
 
 		#ifndef __GNUC__
-			#error "another compiler than GCC is currently not supported in 64bit mode"
+			c = ttmath_subvector_x64(ss1, ss2, ss1_size, ss2_size, result);
 		#endif
 
-		#ifdef __GNUC__
 
-		/*
-			the asm code is nearly the same as in AddVector
-			only two instructions 'adc' are changed to 'sbb'
-		*/
+		#ifdef __GNUC__
+		
+		//	the asm code is nearly the same as in AddVector
+		//	only two instructions 'adc' are changed to 'sbb'
+		
 		uint dummy1, dummy2, dummy3;
+		uint rest = ss1_size - ss2_size;
 
 			__asm__ __volatile__(
 				"mov %%rdx, %%r8					\n"
@@ -551,13 +651,12 @@ namespace ttmath
 
 		#endif
 
-		TTMATH_LOG("UInt::SubVector")
+		TTMATH_VECTOR_LOGC("UInt::SubVector", c, result, ss1_size)
 
 	return c;
 	}
 
 
-
 	/*!
 		this method moves all bits into the left hand side
 		return value <- this <- c
@@ -577,12 +676,13 @@ namespace ttmath
 	{
 	sint b = value_size;
 	uint * p1 = table;
-	
+
 
 		#ifndef __GNUC__
-			#error "another compiler than GCC is currently not supported in 64bit mode"
+			c = ttmath_rcl_x64(p1,b,c);
 		#endif
 
+
 		#ifdef __GNUC__
 		uint dummy, dummy2;
 
@@ -606,7 +706,7 @@ namespace ttmath
 	
 		#endif
 
-		TTMATH_LOG("UInt::Rcl2_one")
+		TTMATH_LOGC("UInt::Rcl2_one", c)
 
 	return c;
 	}
@@ -634,9 +734,10 @@ namespace ttmath
 	
 
 		#ifndef __GNUC__
-			#error "another compiler than GCC is currently not supported in 64bit mode"
+			c = ttmath_rcr_x64(p1,b,c);
 		#endif
 
+
 		#ifdef __GNUC__
 		uint dummy;
 
@@ -658,7 +759,7 @@ namespace ttmath
 
 		#endif
 
-		TTMATH_LOG("UInt::Rcr2_one")
+		TTMATH_LOGC("UInt::Rcr2_one", c)
 
 	return c;
 	}
@@ -689,9 +790,10 @@ namespace ttmath
 
 
 		#ifndef __GNUC__
-			#error "another compiler than GCC is currently not supported in 64bit mode"
+			c = ttmath_rcl2_x64(p1,b,bits,c);
 		#endif
 
+
 		#ifdef __GNUC__
 		uint dummy, dummy2, dummy3;
 
@@ -731,7 +833,7 @@ namespace ttmath
 
 		#endif
 
-		TTMATH_LOG("UInt::Rcl2")
+		TTMATH_LOGC("UInt::Rcl2", c)
 
 	return c;
 	}
@@ -758,14 +860,16 @@ namespace ttmath
 
 	sint b = value_size;
 	uint * p1 = table;
-	uint dummy, dummy2, dummy3;
+
 
 		#ifndef __GNUC__
-			#error "another compiler than GCC is currently not supported in 64bit mode"
+			c = ttmath_rcr2_x64(p1,b,bits,c);
 		#endif
 
 
 		#ifdef __GNUC__
+			uint dummy, dummy2, dummy3;
+
 			__asm__  __volatile__(
 
 			"movq %%rcx, %%rsi				\n"
@@ -805,7 +909,7 @@ namespace ttmath
 
 		#endif
 
-		TTMATH_LOG("UInt::Rcr2")
+		TTMATH_LOGC("UInt::Rcr2", c)
 
 	return c;
 	}
@@ -822,11 +926,19 @@ namespace ttmath
 	{
 	sint result;
 
-
+	
 		#ifndef __GNUC__
-			#error "another compiler than GCC is currently not supported in 64bit mode"
+
+			unsigned long nIndex = 0;
+
+			if( _BitScanReverse64(&nIndex,x) == 0 )
+				result = -1;
+			else
+				result = nIndex;
+
 		#endif
 
+
 		#ifdef __GNUC__
 		uint dummy;
 
@@ -847,6 +959,50 @@ namespace ttmath
 	}
 
 
+	/*
+		this method returns the number of the highest set bit in one 64-bit word
+		if the 'x' is zero this method returns '-1'
+
+		***this method is created only on a 64bit platform***
+	*/
+	template<uint value_size>
+	sint UInt<value_size>::FindLowestBitInWord(uint x)
+	{
+	sint result;
+
+	
+		#ifndef __GNUC__
+
+			unsigned long nIndex = 0;
+
+			if( _BitScanForward64(&nIndex,x) == 0 )
+				result = -1;
+			else
+				result = nIndex;
+
+		#endif
+
+
+		#ifdef __GNUC__
+		uint dummy;
+
+				__asm__ (
+
+				"movq $-1, %1          \n"
+				"bsfq %2, %0           \n"
+				"cmovz %1, %0          \n"
+
+				: "=r" (result), "=&r" (dummy)
+				: "r" (x)
+				: "cc" );
+
+		#endif
+
+
+	return result;
+	}
+
+
 	/*!
 		this method sets a special bit in the 'value'
 		and returns the last state of the bit (zero or one)
@@ -870,9 +1026,10 @@ namespace ttmath
 
 
 		#ifndef __GNUC__
-			#error "another compiler than GCC is currently not supported in 64bit mode"
+			old_bit = _bittestandset64((__int64*)&value,bit) != 0;
 		#endif
 
+
 		#ifdef __GNUC__
 
 			__asm__ (
@@ -903,8 +1060,8 @@ namespace ttmath
 
 	/*!
 		multiplication: result_high:result_low = a * b
-		result_high - higher word of the result
-		result_low  - lower word of the result
+		-  result_high - higher word of the result
+		-  result_low  - lower word of the result
 	
 		this methos never returns a carry
 		this method is used in the second version of the multiplication algorithms
@@ -924,10 +1081,12 @@ namespace ttmath
 	uint result1_;
 	uint result2_;
 
+
 		#ifndef __GNUC__
-			#error "another compiler than GCC is currently not supported in 64bit mode"
+			result1_ = _umul128(a,b,&result2_);
 		#endif
 
+
 		#ifdef __GNUC__
 
 		__asm__ (
@@ -981,10 +1140,16 @@ namespace ttmath
 
 		TTMATH_ASSERT( c != 0 )
 
+
 		#ifndef __GNUC__
-			#error "another compiler than GCC is currently not supported in 64bit mode"
+
+			ttmath_div_x64(&a,&b,c);
+			r_    = a;
+			rest_ = b;
+			
 		#endif
 
+
 		#ifdef __GNUC__
 		
 			__asm__ (
@@ -1002,7 +1167,6 @@ namespace ttmath
 		*rest = rest_;
 	}
 
-
 } //namespace
 
 

ttmath/ttmathuint_x86_64_msvc.asm

@@ -0,0 +1,551 @@
+;
+; This file is a part of TTMath Bignum Library
+; and is distributed under the 3-Clause BSD Licence.
+; Author: Christian Kaiser <chk@online.de>, Tomasz Sowa <t.sowa@ttmath.org>
+;
+
+; 
+; Copyright (c) 2009-2017, Christian Kaiser, Tomasz Sowa
+; All rights reserved.
+; 
+; Redistribution and use in source and binary forms, with or without
+; modification, are permitted provided that the following conditions are met:
+; 
+;  * Redistributions of source code must retain the above copyright notice,
+;    this list of conditions and the following disclaimer.
+;    
+;  * Redistributions in binary form must reproduce the above copyright
+;    notice, this list of conditions and the following disclaimer in the
+;    documentation and/or other materials provided with the distribution.
+;    
+;  * Neither the name Christian Kaiser nor the names of contributors to this
+;    project may be used to endorse or promote products derived
+;    from this software without specific prior written permission.
+;
+; THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+; AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+; IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+; ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+; LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+; CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+; SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+; INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+; CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+; ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
+; THE POSSIBILITY OF SUCH DAMAGE.
+;
+
+;
+; compile with debug info:    ml64.exe /c /Zd /Zi ttmathuint_x86_64_msvc.asm
+; compile without debug info: ml64.exe /c ttmathuint_x86_64_msvc.asm
+; this creates ttmathuint_x86_64_msvc.obj file which can be linked with your program
+;
+
+; doxygen info is put to ttmathuint_x86_64.h file
+
+
+PUBLIC	ttmath_adc_x64
+PUBLIC	ttmath_addindexed_x64
+PUBLIC	ttmath_addindexed2_x64
+PUBLIC	ttmath_addvector_x64
+
+PUBLIC	ttmath_sbb_x64
+PUBLIC	ttmath_subindexed_x64
+PUBLIC	ttmath_subvector_x64
+
+PUBLIC	ttmath_rcl_x64
+PUBLIC	ttmath_rcr_x64
+
+PUBLIC	ttmath_rcl2_x64
+PUBLIC	ttmath_rcr2_x64
+
+PUBLIC	ttmath_div_x64
+
+;
+; Microsoft x86_64 convention: http://msdn.microsoft.com/en-us/library/9b372w95.aspx
+;
+;	"rax, rcx, rdx, r8-r11 are volatile."
+;	"rbx, rbp, rdi, rsi, r12-r15 are nonvolatile."
+;
+
+
+.CODE
+
+
+        ALIGN       8
+
+;----------------------------------------
+
+ttmath_adc_x64				PROC
+        ; rcx = p1
+        ; rdx = p2
+        ; r8 = nSize
+        ; r9 = nCarry
+
+        xor		rax, rax
+        xor		r11, r11
+        sub		rax, r9		; sets CARRY if r9 != 0
+
+		ALIGN 16
+ loop1:
+		mov		rax,qword ptr [rdx + r11 * 8]
+		adc		qword ptr [rcx + r11 * 8], rax
+		lea		r11, [r11+1]
+		dec		r8
+		jnz		loop1
+
+		setc	al
+		movzx	rax, al
+
+		ret
+
+ttmath_adc_x64				ENDP
+
+;----------------------------------------
+
+        ALIGN       8
+
+;----------------------------------------
+
+ttmath_addindexed_x64	PROC
+
+        ; rcx = p1
+        ; rdx = nSize
+        ; r8 = nPos
+        ; r9 = nValue
+
+		xor		rax, rax			; rax = result
+		sub		rdx, r8				; rdx = remaining count of uints
+
+		add		qword ptr [rcx + r8 * 8], r9
+		jc		next1
+
+		ret
+
+next1:
+		mov		r9, 1
+
+		ALIGN 16
+loop1:
+		dec		rdx
+		jz		done_with_cy
+		lea		r8, [r8+1]
+		add		qword ptr [rcx + r8 * 8], r9
+		jc		loop1
+
+		ret
+
+done_with_cy:
+		lea		rax, [rax+1]		; rax = 1
+
+		ret
+
+ttmath_addindexed_x64	ENDP
+
+;----------------------------------------
+
+        ALIGN       8
+
+;----------------------------------------
+
+ttmath_addindexed2_x64	PROC
+
+        ; rcx = p1 (pointer)
+        ; rdx = b  (value size)
+        ; r8 = nPos
+        ; r9 = nValue1
+        ; [rsp+0x28] = nValue2
+
+		xor		rax, rax			; return value
+		mov		r11, rcx			; table
+		sub		rdx, r8				; rdx = remaining count of uints
+		mov		r10, [rsp+028h]		; r10 = nValue2
+
+		add		qword ptr [r11 + r8 * 8], r9
+		lea		r8, [r8+1]
+		lea		rdx, [rdx-1]
+		adc		qword ptr [r11 + r8 * 8], r10
+		jc		next
+		ret
+
+		ALIGN 16
+loop1:
+		lea		r8, [r8+1]
+		add		qword ptr [r11 + r8 * 8], 1
+		jc		next
+		ret
+
+next:
+		dec		rdx					; does not modify CY too...
+		jnz		loop1
+		lea		rax, [rax+1]
+		ret
+
+ttmath_addindexed2_x64	ENDP
+
+
+
+;----------------------------------------
+
+        ALIGN       8
+
+;----------------------------------------
+
+
+ttmath_addvector_x64				PROC
+        ; rcx = ss1
+        ; rdx = ss2
+        ; r8 = ss1_size
+        ; r9 = ss2_size
+        ; [rsp+0x28] = result
+
+		mov		r10, [rsp+028h]
+		sub		r8, r9
+        xor		r11, r11				; r11=0, cf=0
+
+		ALIGN 16
+ loop1:
+		mov		rax, qword ptr [rcx + r11 * 8]
+		adc		rax, qword ptr [rdx + r11 * 8]
+		mov		qword ptr [r10 + r11 * 8], rax
+		inc		r11
+		dec		r9
+		jnz		loop1
+
+		adc		r9, r9					; r9 has the cf state
+
+		or		r8, r8
+		jz		done
+
+		neg		r9						; setting cf from r9
+		mov		r9, 0					; don't use xor here (cf is used)
+ loop2:
+		mov		rax, qword ptr [rcx + r11 * 8]
+		adc		rax, r9
+		mov		qword ptr [r10 + r11 * 8], rax
+		inc		r11
+		dec		r8
+		jnz		loop2
+
+		adc		r8, r8
+		mov		rax, r8
+		
+		ret
+
+done:
+		mov		rax, r9
+		ret
+
+ttmath_addvector_x64				ENDP
+
+
+;----------------------------------------
+
+        ALIGN       8
+
+;----------------------------------------
+
+ttmath_sbb_x64				PROC
+
+        ; rcx = p1
+        ; rdx = p2
+        ; r8 = nCount
+        ; r9 = nCarry
+
+        xor		rax, rax
+        xor		r11, r11
+        sub		rax, r9				; sets CARRY if r9 != 0
+
+		ALIGN 16
+ loop1:
+		mov		rax,qword ptr [rdx + r11 * 8]
+		sbb		qword ptr [rcx + r11 * 8], rax
+		lea		r11, [r11+1]
+		dec		r8
+		jnz		loop1
+
+		setc	al
+		movzx	rax, al
+
+		ret
+
+ttmath_sbb_x64				ENDP
+
+;----------------------------------------
+
+        ALIGN       8
+
+;----------------------------------------
+
+ttmath_subindexed_x64	PROC
+        ; rcx = p1
+        ; rdx = nSize
+        ; r8 = nPos
+        ; r9 = nValue
+
+		sub		rdx, r8				; rdx = remaining count of uints
+
+		ALIGN 16
+loop1:
+		sub		qword ptr [rcx + r8 * 8], r9
+		jnc		done
+
+		lea		r8, [r8+1]
+		mov		r9, 1
+		dec		rdx
+		jnz		loop1
+
+		mov		rax, 1
+		ret
+
+done:
+		xor		rax, rax
+		ret
+
+ttmath_subindexed_x64	ENDP
+
+
+
+;----------------------------------------
+
+        ALIGN       8
+
+;----------------------------------------
+
+;	the same asm code as in addvector_x64 only two instructions 'adc' changed to 'sbb'
+
+ttmath_subvector_x64				PROC
+        ; rcx = ss1
+        ; rdx = ss2
+        ; r8 = ss1_size
+        ; r9 = ss2_size
+        ; [rsp+0x28] = result
+
+		mov		r10, [rsp+028h]
+		sub		r8, r9
+        xor		r11, r11				; r11=0, cf=0
+
+		ALIGN 16
+ loop1:
+		mov		rax, qword ptr [rcx + r11 * 8]
+		sbb		rax, qword ptr [rdx + r11 * 8]
+		mov		qword ptr [r10 + r11 * 8], rax
+		inc		r11
+		dec		r9
+		jnz		loop1
+
+		adc		r9, r9					; r9 has the cf state
+
+		or		r8, r8
+		jz		done
+
+		neg		r9						; setting cf from r9
+		mov		r9, 0					; don't use xor here (cf is used)
+ loop2:
+		mov		rax, qword ptr [rcx + r11 * 8]
+		sbb		rax, r9
+		mov		qword ptr [r10 + r11 * 8], rax
+		inc		r11
+		dec		r8
+		jnz		loop2
+
+		adc		r8, r8
+		mov		rax, r8
+		
+		ret
+
+done:
+		mov		rax, r9
+		ret
+
+ttmath_subvector_x64				ENDP
+
+
+
+
+;----------------------------------------
+
+        ALIGN       8
+
+;----------------------------------------
+
+ttmath_rcl_x64	PROC
+        ; rcx = p1
+        ; rdx = b
+        ; r8 = nLowestBit
+
+		mov		r11, rcx			; table
+		xor		r10, r10
+		neg		r8					; CY set if r8 <> 0
+
+		ALIGN 16
+loop1:
+		rcl		qword ptr [r11 + r10 * 8], 1
+		lea		r10, [r10+1]
+		dec		rdx
+		jnz		loop1
+
+		setc	al
+		movzx	rax, al
+
+        ret
+
+ttmath_rcl_x64	ENDP
+
+;----------------------------------------
+
+        ALIGN       8
+
+;----------------------------------------
+
+ttmath_rcr_x64	PROC
+        ; rcx = p1
+        ; rdx = nSize
+        ; r8 = nLowestBit
+
+		xor		r10, r10
+		neg		r8					; CY set if r8 <> 0
+
+		ALIGN 16
+loop1:
+		rcr		qword ptr -8[rcx + rdx * 8], 1
+		dec		rdx
+		jnz		loop1
+
+		setc	al
+		movzx	rax, al
+
+        ret
+
+ttmath_rcr_x64	ENDP
+
+;----------------------------------------
+
+        ALIGN       8
+
+;----------------------------------------
+
+ttmath_div_x64	PROC
+
+        ; rcx = &Hi
+        ; rdx = &Lo
+        ; r8 = nDiv
+
+        mov		r11, rcx
+        mov		r10, rdx
+
+        mov		rdx, qword ptr [r11]
+        mov		rax, qword ptr [r10]
+        div		r8
+        mov		qword ptr [r10], rdx ; remainder
+        mov		qword ptr [r11], rax ; value
+
+        ret
+
+ttmath_div_x64	ENDP
+
+;----------------------------------------
+
+        ALIGN       8
+
+;----------------------------------------
+
+ttmath_rcl2_x64	PROC
+        ; rcx = p1
+        ; rdx = nSize
+        ; r8 = bits
+        ; r9 = c
+
+        push	rbx
+
+        mov		r10, rcx	; r10 = p1
+        xor		rax, rax
+
+        mov		rcx, 64
+        sub		rcx, r8
+
+        mov		r11, -1
+        shr		r11, cl		; r11 = mask
+
+		mov		rcx, r8		; rcx = count of bits
+
+		mov		rbx, rax	; rbx = old value = 0
+		or		r9, r9
+		cmovnz	rbx, r11	; if (c) then old value = mask
+
+        mov		r9, rax		; r9 = index (0..nSize-1)
+
+		ALIGN 16
+loop1:
+		rol		qword ptr [r10+r9*8], cl
+		mov		rax, qword ptr [r10+r9*8]
+		and		rax, r11
+		xor		qword ptr [r10+r9*8], rax
+		or		qword ptr [r10+r9*8], rbx
+		mov		rbx, rax
+
+		lea		r9, [r9+1]
+		dec		rdx
+
+		jnz		loop1
+
+		and		rax, 1
+		pop		rbx
+        ret
+
+ttmath_rcl2_x64	ENDP
+
+;----------------------------------------
+
+        ALIGN       8
+
+;----------------------------------------
+
+ttmath_rcr2_x64	PROC
+        ; rcx = p1
+        ; rdx = nSize
+        ; r8 = bits
+        ; r9 = c
+
+        push	rbx
+        mov		r10, rcx	; r10 = p1
+        xor		rax, rax
+
+        mov		rcx, 64
+        sub		rcx, r8
+
+        mov		r11, -1
+        shl		r11, cl		; r11 = mask
+
+		mov		rcx, r8		; rcx = count of bits
+
+		mov		rbx, rax	; rbx = old value = 0
+		or		r9, r9
+		cmovnz	rbx, r11	; if (c) then old value = mask
+
+        mov		r9, rdx		; r9 = index (0..nSize-1)
+		lea		r9, [r9-1]
+
+		ALIGN 16
+loop1:
+		ror		qword ptr [r10+r9*8], cl
+		mov		rax, qword ptr [r10+r9*8]
+		and		rax, r11
+		xor		qword ptr [r10+r9*8], rax
+		or		qword ptr [r10+r9*8], rbx
+		mov		rbx, rax
+
+		lea		r9, [r9-1]
+		dec		rdx
+
+		jnz		loop1
+
+		rol		rax, 1
+		and		rax, 1
+		pop		rbx
+
+        ret
+
+ttmath_rcr2_x64	ENDP
+
+END

ttmath/win64_assemble.bat

@@ -0,0 +1,9 @@
+rem make sure this is a proper path to the 64 bit assembler
+"C:\Program Files (x86)\Microsoft Visual Studio 14.0\VC\bin\amd64\ml64.exe" /c ttmathuint_x86_64_msvc.asm
+rem ml64.exe will produce ttmathuint_x86_64_msvc.obj which should be added (linked) to your project
+
+rem or you can assemble with debug info
+rem ml64.exe /c /Zd /Zi ttmathuint_x86_64_msvc.asm
+
+rem be nice, most Windows users just click on the file
+pause