changed: version of the library: 0.9.1 now

git-svn-id: svn://ttmath.org/publicrep/ttmath/trunk@286 e52654a7-88a9-db11-a3e9-0013d4bc506e

CHANGELOG

@@ -1,3 +1,239 @@
+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)
+    * fixed:   global function Mod(x) didn't set an ErrorCode object
+    * fixed:   global function Round() didn't test a carry
+               now it sets ErrorCode object
+    * changed: function Sin(x) to Sin(x, ErrorCode * err=0)
+               when x was very big the function returns zero
+               now it sets ErrorCode object to err_overflow
+               and the result has a NaN flag set
+               the same is to Cos() function
+    * changed: PrepareSin(x) is using Big::Mod() now when reducing 2PI period
+               should be a little accurate especially on a very big 'x'
+    * changed: uint Mul(const UInt<value_size> & ss2, uint algorithm = 100)
+               void MulBig(const UInt<value_size> & ss2, UInt<value_size*2> & result, uint algorithm = 100)
+               those methods by default use MulFastest() and MulFastestBig()
+    * changed: changed a little Mul2Big() to cooperate with Mul3Big()
+    * added:   uint UInt::Mul3(const UInt<value_size> & ss2)
+               void UInt::Mul3Big(const UInt<value_size> & ss2, UInt<value_size*2> & result)
+               a new multiplication algorithm: Karatsuba multiplication,
+               on a vector UInt<100> with all items different from zero this algorithm is faster
+               about 3 times than Mul2Big(), and on a vector UInt<1000> with all items different from
+               zero this algorithm is faster more than 5 times than Mul2Big()
+               (measured on 32bit platform with GCC 4.3.3 with -O3 and -DTTMATH_RELEASE)
+    * added:   uint MulFastest(const UInt<value_size> & ss2)
+               void MulFastestBig(const UInt<value_size> & ss2, UInt<value_size*2> & result)
+               those methods are trying to select the fastest multiplication algorithm
+    * added:   uint AddVector(const uint * ss1, const uint * ss2, uint ss1_size, uint ss2_size, uint * result)
+               uint SubVector(const uint * ss1, const uint * ss2, uint ss1_size, uint ss2_size, uint * result)
+               three forms: asm x86, asm x86_64, no-asm
+               those methods are used by the Karatsuba multiplication algorithm
+    * added:   to Big<> class: support for NaN flag (Not a Number)
+               bool Big::IsNan() - returns true if the NaN flag is set
+               void Big::SetNan() - sets the NaN flag
+               The NaN flag is set by default after creating an object:
+                 Big<1, 2> a;    // NaN is set (it means the object has not a valid number)
+                 std::cout << a; // cout gives "NaN"
+                 a = 123;        // now NaN is not set
+                 std::cout << a; // cout gives "123"
+               The NaN is set if there was a carry during calculations
+                 a.Mul(very_big_value); // a will have a NaN set
+               The NaN is set if an argument is NaN too
+                 b.SetNan();
+                 a.Add(b);  // a will have NaN because b has NaN too
+               If you try to do something on a NaN object, the result is a NaN too
+                 a.SetNan();
+                 a.Add(2);  // a is still a NaN 
+               The NaN is set if you use incorrect arguments
+                 a.Ln(-10); // a will have the NaN flag
+               The only way to clear the NaN flag is to assign a correct value or other correct object,
+               supposing 'a' has NaN flag, to remove the flag you can either:
+                 a = 10;
+                 a.FromInt(30);
+                 a.SetOne(); 
+                 a.FromBig(other_object_without_nan);
+                 etc.
+               
+               
+Version 0.8.4 (2009.05.08):
+    * fixed:   UInt::DivInt() didn't check whether the divisor is zero
+               there was a hardware interruption when the divisor was zero
+               (now the method returns one)
+    * fixed:   the problem with GCC optimization on x86_64
+               sometimes when using -O2 or -O3 GCC doesn't set correctly
+               the stack pointer (actually the stack is used for other things)
+               and you can't use instructions like push/pop in assembler code.
+               All the asm code in x86_64 have been rewritten, now instructions
+               push/pop are not used, other thing which have access to stack 
+               (like "m" (mask) constraints in Rcl2 and Rcr2) have also gone away,
+               now the library works well with -O2 and -O3 and the asm code
+               is a little faster
+    * added:   UInt::PrintLog(const char * msg, std::ostream & output)
+               used (for debugging purposes) by macro TTMATH_LOG(msg)
+               (it is used in nearly all methods in UInt class)
+    * added:   macro TTMATH_DEBUG_LOG: when defined then TTMATH_LOG() 
+               put some debug information (to std::cout)
+    * added:   ttmathuint_x86.h, ttmathuint_x86_64.h, ttmathuint_noasm.h, 
+               all the methods which are using assembler code have been 
+               rewritten to no-asm forms, now we have:
+               1. asm for x86      file: ttmathuint_x86.h
+               2. asm for x86_64   file: ttmathuint_x86_64.h
+               3. no asm           file: ttmathuint_noasm.h
+                  (it's used when macro TTMATH_NOASM is defined)
+               The third form can be used on x86 and x86_64 as well and
+               on other platforms with a little effort.
+                              
+
 Version 0.8.3 (2009.04.06):
     * fixed:   RclMoveAllWords() and RcrMoveAllWords() sometimes didn't return
                the proper carry, (when 'bits' was greater than or equal to 'value_size')

COPYRIGHT

@@ -1,4 +1,4 @@
-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

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@slimaczek.pl>
-Project pages: http://ttmath.slimaczek.pl
-http://sourceforge.net/projects/ttmath
+Author: Tomasz Sowa <t.sowa@ttmath.org>
+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 -DCONSTANTSGENERATOR
+CFLAGS = -s -O2 -DTTMATH_CONSTANTSGENERATOR
 name   = gen
 
 

constgen/main.cpp

@@ -91,7 +91,7 @@ void CalcE()
 	ttmath::Big<1,400> e;
 	ttmath::uint steps;
 
-	// macro CONSTANTSGENERATOR has to be defined	
+	// macro TTMATH_CONSTANTSGENERATOR has to be defined	
 	e.ExpSurrounding0(1, &steps);
 	std::cout << "---------------- e ----------------" << std::endl;
 	e.mantissa.PrintTable(std::cout);
@@ -105,7 +105,7 @@ void CalcLn(int x)
 	ttmath::Big<1,400> ln;
 	ttmath::uint steps;
 
-	// macro CONSTANTSGENERATOR has to be defined	
+	// macro TTMATH_CONSTANTSGENERATOR has to be defined	
 	ln.LnSurrounding1(x, &steps);
 	std::cout << "---------------- ln(" << x << ") ----------------" << std::endl;
 	ln.mantissa.PrintTable(std::cout);

samples/Makefile

@@ -8,7 +8,7 @@ CFLAGS = -Wall -pedantic -s -O2 -I..
 	$(CC) -c $(CFLAGS) $<
 
 
-all: uint int big parser
+all: uint int big big2 parser
 
 
 uint: uint.o
@@ -20,6 +20,9 @@ int: int.o
 big: big.o
 	$(CC) -o big $(CFLAGS) big.o
 
+big2: big2.o
+	$(CC) -o big2 $(CFLAGS) big2.o
+
 parser: parser.o
 	$(CC) -o parser $(CFLAGS) parser.o
 
@@ -27,6 +30,7 @@ parser: parser.o
 uint.o:		uint.cpp
 int.o:		int.cpp
 big.o:		big.cpp
+big2.o:		big2.cpp
 parser.o:	parser.cpp
 
 
@@ -36,6 +40,7 @@ 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
 	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;
@@ -86,12 +89,12 @@ b = 98767878.124322
 a + b = 98891334.667778
 a - b = -98644421.580866
 a * b = 12193540837712.2708
-a / b = 0.0012499665458095765
+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 = 1.624801256066640878e+646457012
+b = 456.319999999999993
+a + b = 1.624801256066640878e+646457012
+a - b = 1.624801256066640878e+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 = 0.001249966545809576460596448526166860913
+Calculating with a carry
+a = 2.3495345545711177736883282090959505003e+2776511644261678604
+b = 456.3199999999999931787897367030382156
+a + b = 2.3495345545711177736883282090959505003e+2776511644261678604
+a - b = 2.3495345545711177736883282090959505003e+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
 */
+
+
+/*
+the result (on 64 bit platform):
+-897705014.5257310676097719585259773124
+*/

ttmath/ttmathint.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>
+ * Author: Tomasz Sowa <t.sowa@ttmath.org>
  */
 
 /* 
@@ -468,6 +468,64 @@ public:
 	}
 
 
+	/*!
+		division this = this / ss2  (ss2 is int)
+		returned values:
+			0 - ok
+			1 - division by zero
+
+		for example: (result means 'this')
+			 20 /  3 --> result:  6   remainder:  2
+			-20 /  3 --> result: -6   remainder: -2
+			 20 / -3 --> result: -6   remainder:  2
+			-20 / -3 --> result:  6   remainder: -2
+
+		in other words: this(old) = ss2 * this(new)(result) + remainder
+	*/
+	uint DivInt(sint ss2, sint * remainder = 0)
+	{
+	bool ss1_is_sign, ss2_is_sign;
+
+		ss1_is_sign = IsSign();
+
+		/*
+			we don't have to test the carry from Abs as well as in Mul
+		*/
+		Abs();
+
+		if( ss2 < 0 )
+		{
+			ss2 = -ss2;
+			ss2_is_sign = true;
+		}
+		else
+		{
+			ss2_is_sign = false;
+		}
+
+		uint rem;
+		uint c = UInt<value_size>::DivInt((uint)ss2, &rem);
+
+		if( ss1_is_sign != ss2_is_sign )
+			SetSign();
+
+		if( remainder )
+		{
+			if( ss1_is_sign )
+				*remainder = -sint(rem);
+			else
+				*remainder = sint(rem);
+		}
+
+	return c;
+	}
+
+
+	uint DivInt(sint ss2, sint & remainder)
+	{
+		return DivInt(ss2, &remainder);
+	}
+
 
 private:
 
@@ -514,7 +572,7 @@ public:
 			return Pow2(pow);
 
 		if( UInt<value_size>::IsZero() )
-			// if 'p' is negative then
+			// if 'pow' is negative then
 			// 'this' must be different from zero
 			return 2;
 
@@ -534,6 +592,7 @@ public:
 	}
 
 
+
 	/*!
 	*
 	*	convertion methods
@@ -647,7 +706,7 @@ public:
 	return 0;
 	}
 
-	// converting from Int
+
 
 	/*!
 		the default assignment operator
@@ -714,7 +773,6 @@ public:
 	}
 
 
-	// converting from UInt
 
 	/*!
 		this operator converts an UInt<another_size> type to this class
@@ -759,8 +817,8 @@ public:
 		// look that 'size' we still set as 'value_size' and not as u.value_size
 		FromUInt(u);
 	}
+ 
 
-	// 
 
 
 #ifdef TTMATH_PLATFORM64
@@ -828,6 +886,15 @@ public:
 	}
 
 
+	/*!
+		a constructor for converting string to this class (with the base=10)
+	*/
+	Int(const wchar_t * s)
+	{
+		FromString(s);
+	}
+
+
 	/*!
 		a constructor for converting a string to this class (with the base=10)
 	*/
@@ -837,6 +904,15 @@ public:
 	}
 
 
+	/*!
+		a constructor for converting a string to this class (with the base=10)
+	*/
+	Int(const std::wstring & s)
+	{
+		FromString( s.c_str() );
+	}
+
+
 	/*!
 		a default constructor
 
@@ -867,10 +943,13 @@ public:
 	}
 
 
+private:
+
 	/*!	
-		this method converts the value to a string with a base equal 'b'
+		an auxiliary method for converting to a string
 	*/
-	void ToString(std::string & result, uint b = 10) const
+	template<class string_type>
+	void ToStringBase(string_type & result, uint b = 10) const
 	{
 		if( IsSign() )
 		{
@@ -886,40 +965,71 @@ public:
 		}
 	}
 
+public:
+
+	/*!	
+		this method converts the value to a string with a base equal 'b'
+	*/
+	void ToString(std::string & result, uint b = 10) const
+	{
+		return ToStringBase(result, b);
+	}
 
 
+	/*!	
+		this method converts the value to a string with a base equal 'b'
+	*/
+	void ToString(std::wstring & result, uint b = 10) const
+	{
+		return ToStringBase(result, b);
+	}
+
+
+	/*!	
+		this method converts the value to a string with a base equal 'b'
+	*/
+	std::string ToString(uint b = 10) const
+	{
+		std::string result;
+		ToStringBase(result, b);
+
+	return result;
+	}
+
+
+	/*!	
+		this method converts the value to a string with a base equal 'b'
+	*/
+	std::wstring ToWString(uint b = 10) const
+	{
+		std::wstring result;
+		ToStringBase(result, b);
+
+	return result;
+	}
+
+
+private:
 
 	/*!
-		this method converts a string into its value
-		it returns carry=1 if the value will be too big or an incorrect base 'b' is given
-
-		string is ended with a non-digit value, for example:
-			"-12" will be translated to -12
-			as well as:
-			"- 12foo" will be translated to 12 too
-
-		existing first white characters will be ommited
-		(between '-' and a first digit can be white characters too)
-
-		after_source (if exists) is pointing at the end of the parsed string
-
-		value_read (if exists) tells whether something has actually been read (at least one digit)
+		an auxiliary method for converting from a string
 	*/
-	uint FromString(const char * s, uint b = 10, const char ** after_source = 0, bool * value_read = 0)
+	template<class char_type>
+	uint FromStringBase(const char_type * s, uint b = 10, const char_type ** after_source = 0, bool * value_read = 0)
 	{
 	bool is_sign = false;
 	
-		UInt<value_size>::SkipWhiteCharacters(s);
+		Misc::SkipWhiteCharacters(s);
 
 		if( *s == '-' )
 		{
 			is_sign = true;
-			UInt<value_size>::SkipWhiteCharacters(++s);
+			Misc::SkipWhiteCharacters(++s);
 		}
 		else
 		if( *s == '+' )
 		{
-			UInt<value_size>::SkipWhiteCharacters(++s);
+			Misc::SkipWhiteCharacters(++s);
 		}
 
 		if( UInt<value_size>::FromString(s,b,after_source,value_read) )
@@ -958,13 +1068,56 @@ public:
 	}
 
 
+public:
+
+	/*!
+		this method converts a string into its value
+		it returns carry=1 if the value will be too big or an incorrect base 'b' is given
+
+		string is ended with a non-digit value, for example:
+			"-12" will be translated to -12
+			as well as:
+			"- 12foo" will be translated to -12 too
+
+		existing first white characters will be ommited
+		(between '-' and a first digit can be white characters too)
+
+		after_source (if exists) is pointing at the end of the parsed string
+
+		value_read (if exists) tells whether something has actually been read (at least one digit)
+	*/
+	uint FromString(const char * s, uint b = 10, const char ** after_source = 0, bool * value_read = 0)
+	{
+		return FromStringBase(s, b, after_source, value_read);
+	}
+
+
+	/*!
+		this method converts a string into its value
+	*/
+	uint FromString(const wchar_t * s, uint b = 10, const wchar_t ** after_source = 0, bool * value_read = 0)
+	{
+		return FromStringBase(s, b, after_source, value_read);
+	}
+
+
 	/*!
 		this method converts a string into its value
 		it returns carry=1 if the value will be too big or an incorrect base 'b' is given
 	*/
 	uint FromString(const std::string & s, uint b = 10)
 	{
-		return FromString( s.c_str() );
+		return FromString( s.c_str(), b );
+	}
+
+
+	/*!
+		this method converts a string into its value
+		it returns carry=1 if the value will be too big or an incorrect base 'b' is given
+	*/
+	uint FromString(const std::wstring & s, uint b = 10)
+	{
+		return FromString( s.c_str(), b );
 	}
 
 
@@ -979,6 +1132,17 @@ public:
 	}
 
 
+	/*!
+		this operator converts a string into its value (with base = 10)
+	*/
+	Int<value_size> & operator=(const wchar_t * s)
+	{
+		FromString(s);
+
+	return *this;
+	}
+
+
 	/*!
 		this operator converts a string into its value (with base = 10)
 	*/
@@ -990,6 +1154,15 @@ public:
 	}
 
 
+	/*!
+		this operator converts a string into its value (with base = 10)
+	*/
+	Int<value_size> & operator=(const std::wstring & s)
+	{
+		FromString( s.c_str() );
+
+	return *this;
+	}
 
 
 	/*!
@@ -1269,9 +1442,15 @@ public:
 	*
 	*/
 
-	friend std::ostream & operator<<(std::ostream & s, const Int<value_size> & l)
+private:
+
+	/*!
+		an auxiliary method for outputing to standard streams
+	*/
+	template<class ostream_type, class string_type>
+	static ostream_type & OutputToStream(ostream_type & s, const Int<value_size> & l)
 	{
-	std::string ss;
+	string_type ss;
 
 		l.ToString(ss);
 		s << ss;
@@ -1281,12 +1460,40 @@ public:
 
 
 
-	friend std::istream & operator>>(std::istream & s, Int<value_size> & l)
+public:
+
+
+	/*!
+		output to standard streams
+	*/
+	friend std::ostream & operator<<(std::ostream & s, const Int<value_size> & l)
 	{
-	std::string ss;
+		return OutputToStream<std::ostream, std::string>(s, l);
+	}
+
+
+	/*!
+		output to standard streams
+	*/
+	friend std::wostream & operator<<(std::wostream & s, const Int<value_size> & l)
+	{
+		return OutputToStream<std::wostream, std::wstring>(s, l);
+	}
+
+
+
+private:
+
+	/*!
+		an auxiliary method for converting from a string
+	*/
+	template<class istream_type, class string_type, class char_type>
+	static istream_type & InputFromStream(istream_type & s, Int<value_size> & l)
+	{
+	string_type ss;
 	
-	// char for operator>>
-	unsigned char z;
+	// char or wchar_t for operator>>
+	char_type z;
 	
 		// operator>> omits white characters if they're set for ommiting
 		s >> z;
@@ -1298,10 +1505,10 @@ public:
 		}
 
 		// we're reading only digits (base=10)
-		while( s.good() && UInt<value_size>::CharToDigit(z, 10)>=0 )
+		while( s.good() && Misc::CharToDigit(z, 10)>=0 )
 		{
 			ss += z;
-			z = s.get();
+			z = static_cast<char_type>(s.get());
 		}
 
 		// we're leaving the last readed character
@@ -1313,9 +1520,28 @@ public:
 	return s;
 	}
 
+
+public:
+
+	/*!
+		input from standard streams
+	*/
+	friend std::istream & operator>>(std::istream & s, Int<value_size> & l)
+	{
+		return InputFromStream<std::istream, std::string, char>(s, l);
+	}
+
+
+	/*!
+		input from standard streams
+	*/
+	friend std::wistream & operator>>(std::wistream & s, Int<value_size> & l)
+	{
+		return InputFromStream<std::wistream, std::wstring, wchar_t>(s, l);
+	}
+
 };
 
 } // namespace
 
-
 #endif

ttmath/ttmathmisc.h

@@ -0,0 +1,243 @@
+/*
+ * This file is a part of TTMath Bignum Library
+ * and is distributed under the (new) 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 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;
+}
+
+
+/*!
+	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());
+}
+
+
+/*
+ *
+ *	AddString(result, str)
+ *	result += str
+ *
+ */
+
+
+/*!
+	result += str
+*/
+static void AddString(std::string & result, const char * str)
+{
+	result += str;
+}
+
+
+/*!
+	result += str
+*/
+static void AddString(std::wstring & result, const char * str)
+{
+	for( ; *str ; ++str )
+		result += *str;
+}
+
+
+
+/*
+	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

@@ -1,7 +1,7 @@
 /*
  * This file is a part of TTMath Mathematical Library
  * and is distributed under the (new) BSD licence.
- * Author: Tomasz Sowa <t.sowa@slimaczek.pl>
+ * Author: Tomasz Sowa <t.sowa@ttmath.org>
  */
 
 /* 
@@ -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
@@ -73,6 +74,7 @@ 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
@@ -97,7 +99,7 @@ public:
 		if 'can_be_digit' is true that means when the 'c' is a digit this 
 		method returns true otherwise it returns false
 	*/
-	static bool CorrectCharacter(int c, bool can_be_digit)
+	static bool CorrectCharacter(wchar_t c, bool can_be_digit)
 	{
 		if( (c>='a' && c<='z') || (c>='A' && c<='Z') )
 			return true;
@@ -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 +148,22 @@ public:
 	}
 
 
+	/*!
+		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);
+	}
+
+
 	/*!
 		this method adds one object (variable of function) into the table
 	*/
@@ -165,6 +184,23 @@ public:
 	}
 
 
+	/*!
+		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);
+	}
+
+
 	/*!
 		this method returns 'true' if the table is empty
 	*/
@@ -222,6 +258,23 @@ public:
 	}
 
 
+	/*!
+		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);
+	}
+
+
 	/*!
 		this method changes the name of a specific object
 	*/
@@ -253,6 +306,23 @@ public:
 	}
 
 
+	/*!
+		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);
+	}
+
+
 	/*!
 		this method deletes an object
 	*/
@@ -272,6 +342,22 @@ public:
 	}
 
 
+	/*!
+		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);
+	}	
+		
+		
 	/*!
 		this method gets the value of a specific object
 	*/
@@ -294,6 +380,24 @@ public:
 	}
 
 
+	/*!
+		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;
+	}
+
+
 	/*!
 		this method gets the value of a specific object
 		(this version is used for not copying the whole string)
@@ -317,6 +421,23 @@ public:
 	}
 
 
+	/*!
+		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);
+	}
+
+
 	/*!
 		this method gets the value and the number of parameters
 		of a specific object
@@ -342,6 +463,25 @@ public:
 	}
 
 
+	/*!
+		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;
+	}
+
+
 	/*!
 		this method sets the value and the number of parameters
 		of a specific object
@@ -368,6 +508,25 @@ public:
 	}
 
 
+	/*!
+		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);
+	}
+
+
 	/*!
 		this method returns a pointer into the table
 	*/
@@ -380,6 +539,7 @@ public:
 private:
 
 	Table table;
+	std::string str_tmp1, str_tmp2;
 
 }; // end of class Objects
 
@@ -426,11 +586,11 @@ 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 +647,118 @@ 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
+	in a second version of Gamma() 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/ttmathparser.h

@@ -1,11 +1,11 @@
 /*
- * This file is a part of TTMath Mathematical Library
+ * This file is a part of TTMath Bignum Library
  * and is distributed under the (new) BSD licence.
- * Author: Tomasz Sowa <t.sowa@slimaczek.pl>
+ * 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
@@ -45,13 +45,14 @@
     \brief A mathematical parser
 */
 
-#include <fstream>
+#include <cstdio>
 #include <vector>
 #include <map>
 #include <set>
 
 #include "ttmath.h"
 #include "ttmathobjects.h"
+#include "ttmathmisc.h"
 
 
 
@@ -109,8 +110,7 @@ namespace ttmath
 	for example a correct input string can be:
 		"1"
 		"2.1234"
-		"2,1234"    (they are the same, we can either use a comma or a dot in values)
-		            (look at the macro TTMATH_COMMA_CHARACTER_2)
+		"2,1234"    (they are the same, by default we can either use a comma or a dot)
 		"1 + 2"
 		"(1 + 2) * 3"
 		"pi"
@@ -137,7 +137,6 @@ namespace ttmath
 template<class ValueType>
 class Parser
 {
-
 private:
 
 /*!
@@ -160,14 +159,20 @@ private:
 			none,add,sub,mul,div,pow,lt,gt,let,get,eq,neq,lor,land,shortmul
 		};
 
+		enum Assoc
+		{
+			right,		// right-associative
+			non_right	// associative or left-associative
+		};
 
-		Type GetType()     const { return type; }
-		int  GetPriority() const { return priority; }
-
+		Type  GetType()     const { return type; }
+		int   GetPriority() const { return priority; }
+		Assoc GetAssoc()    const { return assoc; }
 
 		void SetType(Type t)
 		{
-			type = t;
+			type  = t;
+			assoc = non_right;
 
 			switch( type )
 			{		
@@ -201,6 +206,7 @@ private:
 
 			case pow:
 				priority = 14;
+				assoc    = right;
 				break;
 
 			default:
@@ -209,15 +215,15 @@ private:
 			}
 		}
 
-		MatOperator(): type(none), priority(0)
+		MatOperator(): type(none), priority(0), assoc(non_right)
 		{
 		}
 
 	private:
 
-		Type type;
-		int  priority;
-
+		Type  type;
+		int   priority;
+		Assoc assoc;
 	}; // end of MatOperator class
 
 
@@ -311,6 +317,7 @@ ErrorCode error;
 
 /*!
 	pointer to the currently reading char
+	it's either char* or wchar_t*
 
 	when an error has occured it may be used to count the index of the wrong character
 */
@@ -397,7 +404,7 @@ typedef void (ValueType::*pfunction_var)();
 
 	this map consists of:
 		std::string - function's name
-		pfunction   - pointer to specific function
+		pfunction - pointer to specific function
 */
 typedef std::map<std::string, pfunction> FunctionsTable;
 FunctionsTable functions_table;
@@ -418,7 +425,7 @@ OperatorsTable operators_table;
 	table of mathematic variables
 
 	this map consists of:
-		std::string   - variable's name
+		std::string     - variable's name
 		pfunction_var - pointer to specific function which returns value of variable
 */
 typedef std::map<std::string, pfunction_var> VariablesTable;
@@ -426,10 +433,44 @@ VariablesTable variables_table;
 
 
 /*!
-	you can't calculate the factorial if the argument is greater than 'factorial_max'
-	default value is zero which means there are not any limitations
+	some coefficients used when calculating the gamma (or factorial) function
 */
-ValueType factorial_max;
+CGamma<ValueType> cgamma;
+
+
+/*!
+	temporary object for a whole string when Parse(std::wstring) is used
+*/
+std::string wide_to_ansi;
+
+
+/*!
+	group character (used when parsing)
+	default zero (not used)
+*/
+int group;
+
+
+/*!
+	characters used as a comma
+	default: '.' and ','
+	comma2 can be zero (it means it is not used)
+*/
+int comma, comma2;
+
+
+/*!
+	an additional character used as a separator between function parameters
+	(semicolon is used always)
+*/
+int param_sep;
+
+
+/*!
+	true if something was calculated (at least one mathematical operator was used or a function or a variable)
+*/
+bool calculated;
+
 
 
 /*!
@@ -505,7 +546,8 @@ void RecurrenceParsingVariablesOrFunction_DeleteName(bool variable, const std::s
 	(there can be a recurrence here therefore we're using 'visited_variables'
 	and 'visited_functions' sets to make a stop condition)
 */
-ValueType RecurrenceParsingVariablesOrFunction(bool variable, const std::string & name, const char * new_string, FunctionLocalVariables * local_variables = 0)
+ValueType RecurrenceParsingVariablesOrFunction(bool variable, const std::string & name, const char * new_string,
+											   FunctionLocalVariables * local_variables = 0)
 {
 	RecurrenceParsingVariablesOrFunction_CheckStopCondition(variable, name);
 	RecurrenceParsingVariablesOrFunction_AddName(variable, name);
@@ -559,6 +601,7 @@ bool GetValueOfUserDefinedVariable(const std::string & variable_name,ValueType &
 		return false;
 
 	result = RecurrenceParsingVariablesOrFunction(true, variable_name, string_value);
+	calculated = true;
 
 return true;
 }
@@ -607,6 +650,7 @@ ValueType result;
 		Error( err_unknown_variable );
 
 	(result.*(i->second))();
+	calculated = true;
 
 return result;
 }
@@ -674,6 +718,20 @@ return result;
 }
 
 
+void Gamma(int sindex, int amount_of_args, ValueType & result)
+{
+	if( amount_of_args != 1 )
+		Error( err_improper_amount_of_arguments );
+
+	ErrorCode err;
+	
+	result = ttmath::Gamma(stack[sindex].value, cgamma, &err, pstop_calculating);
+
+	if(err != err_ok)
+		Error( err );
+}
+
+
 /*!
 	factorial
 	result = 1 * 2 * 3 * 4 * .... * x
@@ -684,11 +742,8 @@ void Factorial(int sindex, int amount_of_args, ValueType & result)
 		Error( err_improper_amount_of_arguments );
 
 	ErrorCode err;
-	
-	if( !factorial_max.IsZero() && stack[sindex].value > factorial_max )
-		Error( err_too_big_factorial );
 
-	result = ttmath::Factorial(stack[sindex].value, &err, pstop_calculating);
+	result = ttmath::Factorial(stack[sindex].value, cgamma, &err, pstop_calculating);
 
 	if(err != err_ok)
 		Error( err );
@@ -708,7 +763,11 @@ void Sin(int sindex, int amount_of_args, ValueType & result)
 	if( amount_of_args != 1 )
 		Error( err_improper_amount_of_arguments );
 
-	result = ttmath::Sin( ConvertAngleToRad(stack[sindex].value) );
+	ErrorCode err;
+	result = ttmath::Sin( ConvertAngleToRad(stack[sindex].value), &err );
+
+	if(err != err_ok)
+		Error( err );
 }
 
 void Cos(int sindex, int amount_of_args, ValueType & result)
@@ -716,7 +775,11 @@ void Cos(int sindex, int amount_of_args, ValueType & result)
 	if( amount_of_args != 1 )
 		Error( err_improper_amount_of_arguments );
 
-	result = ttmath::Cos( ConvertAngleToRad(stack[sindex].value) );
+	ErrorCode err;
+	result = ttmath::Cos( ConvertAngleToRad(stack[sindex].value), &err );
+
+	if(err != err_ok)
+		Error( err );
 }
 
 void Tan(int sindex, int amount_of_args, ValueType & result)
@@ -757,7 +820,10 @@ void Round(int sindex, int amount_of_args, ValueType & result)
 	if( amount_of_args != 1 )
 		Error( err_improper_amount_of_arguments );
 
-	result = ttmath::Round(stack[sindex].value);
+	result = stack[sindex].value;
+
+	if( result.Round() )
+		Error( err_overflow );
 }
 
 
@@ -973,7 +1039,7 @@ void Not(int sindex, int amount_of_args, ValueType & result)
 
 void DegToRad(int sindex, int amount_of_args, ValueType & result)
 {
-	ErrorCode err;
+	ErrorCode err = err_ok;
 
 	if( amount_of_args == 1 )
 	{
@@ -1052,7 +1118,7 @@ void RadToGrad(int sindex, int amount_of_args, ValueType & result)
 
 void DegToGrad(int sindex, int amount_of_args, ValueType & result)
 {
-	ErrorCode err;
+	ErrorCode err = err_ok;
 
 	if( amount_of_args == 1 )
 	{
@@ -1333,6 +1399,45 @@ void Avg(int sindex, int amount_of_args, ValueType & result)
 }	
 
 
+void Frac(int sindex, int amount_of_args, ValueType & result)
+{
+	if( amount_of_args != 1 )
+		Error( err_improper_amount_of_arguments );
+
+	result = stack[sindex].value;
+	result.RemainFraction();
+}
+
+
+
+
+/*!
+	we use such a method because 'wvsprintf' is not everywhere defined
+*/
+void Sprintf(char * buffer, int par)
+{
+char buf[30]; // char, not wchar_t etc.
+int i;
+
+	#ifdef _MSC_VER
+	#pragma warning( disable: 4996 )
+	//warning C4996: 'sprintf': This function or variable may be unsafe.
+	#endif
+
+	sprintf(buf, "%d", par);
+	for(i=0 ; buf[i] != 0 ; ++i)
+		buffer[i] = buf[i];
+
+	buffer[i] = 0;
+
+	#ifdef _MSC_VER
+	#pragma warning( default: 4996 )
+	#endif
+}
+
+
+
+
 /*!
 	this method returns the value from a user-defined function
 
@@ -1357,20 +1462,23 @@ bool GetValueOfUserDefinedFunction(const std::string & function_name, int amount
 
 	if( amount_of_args > 0 )
 	{
-		char buffer[20];
+		char buffer[30];
 
 		// x = x1
-		sprintf(buffer,"x");
+		buffer[0] = 'x';
+		buffer[1] = 0;
 		local_variables.insert( std::make_pair(buffer, stack[sindex].value) );
 
 		for(int i=0 ; i<amount_of_args ; ++i)
 		{
-			sprintf(buffer,"x%d",i+1);
+			buffer[0] = 'x';
+			Sprintf(buffer+1, i+1);
 			local_variables.insert( std::make_pair(buffer, stack[sindex + i*2].value) );
 		}
 	}
 
 	stack[sindex-1].value = RecurrenceParsingVariablesOrFunction(false, function_name, string_value, &local_variables);
+	calculated = true;
 
 return true;
 }
@@ -1403,6 +1511,7 @@ void CallFunction(const std::string & function_name, int amount_of_args, int sin
 		calling the specify function
 	*/
 	(this->*(i->second))(sindex, amount_of_args, stack[sindex-1].value);
+	calculated = true;
 }
 
 
@@ -1417,10 +1526,14 @@ void CallFunction(const std::string & function_name, int amount_of_args, int sin
 */
 void InsertFunctionToTable(const char * function_name, pfunction pf)
 {
-	functions_table.insert( std::make_pair(std::string(function_name), pf));
+	std::string str;
+	Misc::AssignString(str, function_name);
+
+	functions_table.insert( std::make_pair(str, pf) );
 }
 
 
+
 /*!
 	inserting a function to the variables' table
 	(this function returns value of variable)
@@ -1430,7 +1543,10 @@ void InsertFunctionToTable(const char * function_name, pfunction pf)
 */
 void InsertVariableToTable(const char * variable_name, pfunction_var pf)
 {
-	variables_table.insert( std::make_pair(std::string(variable_name), pf));
+	std::string str;
+	Misc::AssignString(str, variable_name);
+
+	variables_table.insert( std::make_pair(str, pf) );
 }
 
 
@@ -1439,20 +1555,18 @@ void InsertVariableToTable(const char * variable_name, pfunction_var pf)
 */
 void CreateFunctionsTable()
 {
-	/*
-		names of functions should consist of small letters
-	*/
+	InsertFunctionToTable("gamma",		&Parser<ValueType>::Gamma);
 	InsertFunctionToTable("factorial",	&Parser<ValueType>::Factorial);
 	InsertFunctionToTable("abs",   		&Parser<ValueType>::Abs);
 	InsertFunctionToTable("sin",   		&Parser<ValueType>::Sin);
 	InsertFunctionToTable("cos",   		&Parser<ValueType>::Cos);
 	InsertFunctionToTable("tan",   		&Parser<ValueType>::Tan);
-	InsertFunctionToTable("tg",		  	&Parser<ValueType>::Tan);
+	InsertFunctionToTable("tg",			&Parser<ValueType>::Tan);
 	InsertFunctionToTable("cot",  		&Parser<ValueType>::Cot);
 	InsertFunctionToTable("ctg",  		&Parser<ValueType>::Cot);
 	InsertFunctionToTable("int",	   	&Parser<ValueType>::Int);
 	InsertFunctionToTable("round",	 	&Parser<ValueType>::Round);
-	InsertFunctionToTable("ln",	    	&Parser<ValueType>::Ln);
+	InsertFunctionToTable("ln",			&Parser<ValueType>::Ln);
 	InsertFunctionToTable("log",	   	&Parser<ValueType>::Log);
 	InsertFunctionToTable("exp",	   	&Parser<ValueType>::Exp);
 	InsertFunctionToTable("max",	   	&Parser<ValueType>::Max);
@@ -1500,6 +1614,7 @@ void CreateFunctionsTable()
 	InsertFunctionToTable("bxor",		&Parser<ValueType>::BitXor);
 	InsertFunctionToTable("sum",		&Parser<ValueType>::Sum);
 	InsertFunctionToTable("avg",		&Parser<ValueType>::Avg);
+	InsertFunctionToTable("frac",		&Parser<ValueType>::Frac);
 }
 
 
@@ -1508,9 +1623,6 @@ void CreateFunctionsTable()
 */
 void CreateVariablesTable()
 {
-	/*
-		names of variables should consist of small letters
-	*/
 	InsertVariableToTable("pi", &ValueType::SetPi);
 	InsertVariableToTable("e",  &ValueType::SetE);
 }
@@ -1555,7 +1667,7 @@ int character;
 
 	do
 	{
-		result   += character;
+		result   += static_cast<char>( character );
 		character = * ++pstring;
 	}
 	while(	(character>='a' && character<='z') ||
@@ -1634,6 +1746,8 @@ return is_it_name_of_function;
 }
 
 
+
+
 /*!
 	we're reading a numerical value directly from the string
 */
@@ -1641,9 +1755,15 @@ void ReadValue(Item & result, int reading_base)
 {
 const char * new_stack_pointer;
 bool value_read;
+Conv conv;
 
-	int carry = result.value.FromString(pstring, reading_base, &new_stack_pointer, &value_read);
-	pstring   = new_stack_pointer;
+	conv.base   = reading_base;
+	conv.comma  = comma;
+	conv.comma2 = comma2;
+	conv.group  = group;
+
+	uint carry = result.value.FromString(pstring, conv, &new_stack_pointer, &value_read);
+	pstring    = new_stack_pointer;
 
 	if( carry )
 		Error( err_overflow );
@@ -1653,43 +1773,18 @@ bool value_read;
 }
 
 
-/*!
-	this method converts the character ascii c into the value in range <0;base-1>
-
-	if the character is incorrect for this base the funcion will return -1
-*/
-int CharToDigit(int c, int 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 c;
-}
-
-
 /*!
 	this method returns true if 'character' is a proper first digit for the value (or a comma -- can be first too)
 */
 bool ValueStarts(int character, int base)
 {
-	if( character == TTMATH_COMMA_CHARACTER_1 )
+	if( character == comma )
 		return true;
 
-	if( TTMATH_COMMA_CHARACTER_2 != 0 && character == TTMATH_COMMA_CHARACTER_2 )
+	if( comma2!=0 && character==comma2 )
 		return true;
 
-	if( CharToDigit(character, base) != -1 )
+	if( Misc::CharToDigit(character, base) != -1 )
 		return true;
 
 return false;
@@ -1800,21 +1895,21 @@ int  character;
 		this value is from a variable or directly from the string
 	*/
 	result.type = Item::numerical_value;
-
+	
 	if( result.sign )
 	{
 		result.value.ChangeSign();
 		result.sign = false;
 	}
-
+	
 
 return 0;
 }
 
 
-void InsertOperatorToTable(const std::string & name, typename MatOperator::Type type)
+void InsertOperatorToTable(const char * name, typename MatOperator::Type type)
 {
-	operators_table.insert( std::make_pair(name, type) );
+	operators_table.insert( std::make_pair(std::string(name), type) );
 }
 
 
@@ -1823,19 +1918,19 @@ void InsertOperatorToTable(const std::string & name, typename MatOperator::Type
 */
 void CreateMathematicalOperatorsTable()
 {
-	InsertOperatorToTable(std::string("||"), MatOperator::lor);
-	InsertOperatorToTable(std::string("&&"), MatOperator::land);
-	InsertOperatorToTable(std::string("!="), MatOperator::neq);
-	InsertOperatorToTable(std::string("=="), MatOperator::eq);
-	InsertOperatorToTable(std::string(">="), MatOperator::get);
-	InsertOperatorToTable(std::string("<="), MatOperator::let);
-	InsertOperatorToTable(std::string(">"),  MatOperator::gt);
-	InsertOperatorToTable(std::string("<"),  MatOperator::lt);
-	InsertOperatorToTable(std::string("-"),  MatOperator::sub);
-	InsertOperatorToTable(std::string("+"),  MatOperator::add);
-	InsertOperatorToTable(std::string("/"),  MatOperator::div);
-	InsertOperatorToTable(std::string("*"),  MatOperator::mul);
-	InsertOperatorToTable(std::string("^"),  MatOperator::pow);
+	InsertOperatorToTable("||", MatOperator::lor);
+	InsertOperatorToTable("&&", MatOperator::land);
+	InsertOperatorToTable("!=", MatOperator::neq);
+	InsertOperatorToTable("==", MatOperator::eq);
+	InsertOperatorToTable(">=", MatOperator::get);
+	InsertOperatorToTable("<=", MatOperator::let);
+	InsertOperatorToTable(">",  MatOperator::gt);
+	InsertOperatorToTable("<",  MatOperator::lt);
+	InsertOperatorToTable("-",  MatOperator::sub);
+	InsertOperatorToTable("+",  MatOperator::add);
+	InsertOperatorToTable("/",  MatOperator::div);
+	InsertOperatorToTable("*",  MatOperator::mul);
+	InsertOperatorToTable("^",  MatOperator::pow);
 }
 
 
@@ -1850,7 +1945,7 @@ bool IsSubstring(const std::string & str1, const std::string & str2)
 	if( str2.length() > str1.length() )
 		return false;
 
-	for(std::string::size_type i=0 ; i<str2.length() ; ++i)
+	for(typename std::string::size_type i=0 ; i<str2.length() ; ++i)
 		if( str1[i] != str2[i] )
 			return false;
 
@@ -1892,6 +1987,31 @@ typename OperatorsTable::iterator iter_old, iter_new;
 }
 
 
+/*!
+	this method makes a calculation for the percentage operator
+	e.g.
+	1000-50% = 1000-(1000*0,5) = 500
+*/
+void OperatorPercentage()
+{
+	if( stack_index < 3										||
+		stack[stack_index-1].type != Item::numerical_value	||
+		stack[stack_index-2].type != Item::mat_operator		||
+		stack[stack_index-3].type != Item::numerical_value	)
+		Error(err_percent_from);
+
+	++pstring;
+	SkipWhiteCharacters();
+
+	uint c = 0;
+	c += stack[stack_index-1].value.Div(100);
+	c += stack[stack_index-1].value.Mul(stack[stack_index-3].value);
+
+	if( c )
+		Error(err_overflow);
+}
+
+
 /*!
 	this method reads a mathematic operators
 	or the final bracket or the semicolon operator
@@ -1903,7 +2023,11 @@ typename OperatorsTable::iterator iter_old, iter_new;
 int ReadOperator(Item & result)
 {
 	SkipWhiteCharacters();
-	
+
+	if( *pstring == '%' )
+		OperatorPercentage();
+
+
 	if( *pstring == 0 )
 		return 1;
 	else
@@ -1913,7 +2037,7 @@ int ReadOperator(Item & result)
 		++pstring;
 	}
 	else
-	if( *pstring == ';' )
+	if( *pstring == ';' || (param_sep!=0 && *pstring==param_sep) )
 	{
 		result.type = Item::semicolon;
 		++pstring;
@@ -1943,7 +2067,9 @@ return 0;
 void MakeStandardMathematicOperation(ValueType & value1, typename MatOperator::Type mat_operator,
 									const ValueType & value2)
 {
-int res;
+uint res;
+
+	calculated = true;
 
 	switch( mat_operator )
 	{
@@ -2006,7 +2132,6 @@ int res;
 
 		break;
 
-
 	default:
 		/*
 			on the stack left an unknown operator but we had to recognize its before
@@ -2035,10 +2160,20 @@ void TryRollingUpStackWithOperatorPriority()
 {
 	while(	stack_index>=4 &&
 			stack[stack_index-4].type == Item::numerical_value &&
-			stack[stack_index-3].type == Item::mat_operator &&
+			stack[stack_index-3].type == Item::mat_operator    &&
 			stack[stack_index-2].type == Item::numerical_value &&
-			stack[stack_index-1].type == Item::mat_operator &&
-			stack[stack_index-3].moperator.GetPriority() >= stack[stack_index-1].moperator.GetPriority()
+			stack[stack_index-1].type == Item::mat_operator    &&
+			(
+				(
+					// the first operator has greater priority
+					stack[stack_index-3].moperator.GetPriority() > stack[stack_index-1].moperator.GetPriority()
+				) ||
+				(
+					// or both operators have the same priority and the first operator is not right associative
+					stack[stack_index-3].moperator.GetPriority() == stack[stack_index-1].moperator.GetPriority() &&
+					stack[stack_index-3].moperator.GetAssoc()    == MatOperator::non_right
+				)
+			)
 		 )
 	{
 		MakeStandardMathematicOperation(stack[stack_index-4].value,
@@ -2221,7 +2356,7 @@ int index;
 		stack[index-1].value.ChangeSign();
 
 	stack[index-1].type = Item::numerical_value;
-	
+
 
 	/*
 		the pointer of the stack will be pointing on the next (non-existing now) element
@@ -2387,13 +2522,16 @@ public:
 Parser(): default_stack_size(100)
 {
 	pstop_calculating = 0;
-	puser_variables = 0;
-	puser_functions = 0;
+	puser_variables   = 0;
+	puser_functions   = 0;
 	pfunction_local_variables = 0;
-	base = 10;
-	deg_rad_grad = 1;
-	error = err_ok;
-	factorial_max.SetZero();
+	base              = 10;
+	deg_rad_grad      = 1;
+	error             = err_ok;
+	group             = 0;
+	comma             = '.';
+	comma2            = ',';
+	param_sep         = 0;
 
 	CreateFunctionsTable();
 	CreateVariablesTable();
@@ -2410,10 +2548,13 @@ Parser<ValueType> & operator=(const Parser<ValueType> & p)
 	puser_variables   = p.puser_variables;
 	puser_functions   = p.puser_functions;
 	pfunction_local_variables = 0;
-	base = p.base;
-	deg_rad_grad = p.deg_rad_grad;
-	error = err_ok;
-	factorial_max = p.factorial_max;
+	base              = p.base;
+	deg_rad_grad      = p.deg_rad_grad;
+	error             = p.error;
+	group             = p.group;
+	comma             = p.comma;
+	comma2            = p.comma2;
+	param_sep         = p.param_sep;
 
 	/*
 		we don't have to call 'CreateFunctionsTable()' etc.
@@ -2440,7 +2581,8 @@ Parser(const Parser<ValueType> & p): default_stack_size(p.default_stack_size)
 
 
 /*!
-	the new base of mathematic system		
+	the new base of mathematic system
+	default is 10
 */
 void SetBase(int b)
 {
@@ -2496,14 +2638,36 @@ void SetFunctions(const Objects * pf)
 
 
 /*!
-	you will not be allowed to calculate the factorial 
-	if its argument is greater than 'm'
-	there'll be: ErrorCode::err_too_big_factorial
-	default 'factorial_max' is zero which means you can calculate what you want to
+	setting the group character
+	default zero (not used)
 */
-void SetFactorialMax(const ValueType & m)
+void SetGroup(int g)
 {
-	factorial_max = m;
+	group = g;
+}
+
+
+/*!
+	setting the main comma operator and the additional comma operator
+	the additional operator can be zero (which means it is not used)
+	default are: '.' and ','
+*/
+void SetComma(int c, int c2 = 0)
+{
+	comma  = c;
+	comma2 = c2;
+}
+
+
+/*!
+	setting an additional character which is used as a parameters separator
+	the main parameters separator is a semicolon (is used always)
+
+	this character is used also as a global separator
+*/
+void SetParamSep(int s)
+{
+	param_sep = s;
 }
 
 
@@ -2515,7 +2679,8 @@ ErrorCode Parse(const char * str)
 	stack_index  = 0;
 	pstring      = str;
 	error        = err_ok;
-	
+	calculated   = false;
+
 	stack.resize( default_stack_size );
 
 	try
@@ -2525,6 +2690,7 @@ ErrorCode Parse(const char * str)
 	catch(ErrorCode c)
 	{
 		error = c;
+		calculated = false;
 	}
 
 	NormalizeStack();
@@ -2533,11 +2699,55 @@ return error;
 }
 
 
+/*!
+	the main method using for parsing string
+*/
+ErrorCode Parse(const std::string & str)
+{
+	return Parse(str.c_str());
+}
 
 
+/*!
+	the main method using for parsing string
+*/
+ErrorCode Parse(const wchar_t * str)
+{
+	Misc::AssignString(wide_to_ansi, str);
+
+return Parse(wide_to_ansi.c_str());	
+}
+
+
+/*!
+	the main method using for parsing string
+*/
+ErrorCode Parse(const std::wstring & str)
+{
+	return Parse(str.c_str());
+}
+
+
+/*!
+	this method returns true is something was calculated
+	(at least one mathematical operator was used or a function or variable)
+	e.g. true if the string to Parse() looked like this:
+	"1+1"
+	"2*3"
+	"sin(5)"
+
+	if the string was e.g. "678" the result is false
+*/
+bool Calculated()
+{
+	return calculated;
+}
+
 
 };
 
+
+
 } // namespace
 
 

ttmath/ttmaththreads.h

@@ -0,0 +1,250 @@
+/*
+ * This file is a part of TTMath Bignum Library
+ * and is distributed under the (new) 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
+*/
+
+
+/*
+	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)
+*/
+
+
+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;
+
+
+	/*!
+		objects of this class are used to synchronize
+	*/
+	class ThreadLock
+	{
+	public:
+
+		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>
+ * 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
@@ -54,7 +54,7 @@
 
 #include <stdexcept>
 #include <sstream>
-
+#include <vector>
 
 /*!
 	the version of the library
@@ -63,8 +63,8 @@
 	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		3
+#define TTMATH_MINOR_VER		9
+#define TTMATH_REVISION_VER		1
 #define TTMATH_PRERELEASE_VER	0
 
 
@@ -83,7 +83,7 @@
 	 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	
+	if TTMATH_RELEASE is not set then TTMATH_DEBUG is set automatically
 */
 #ifndef TTMATH_RELEASE
 	#define TTMATH_DEBUG
@@ -112,6 +112,15 @@ namespace ttmath
 
 
 
+/*!
+	another compilers than MS VC or GCC by default use no asm version (TTMATH_NOASM)
+*/
+#if !defined _MSC_VER && !defined __GNUC__
+	#define TTMATH_NOASM
+#endif
+
+
+
 #ifdef TTMATH_PLATFORM32
 
 	/*!
@@ -120,6 +129,20 @@ namespace ttmath
 	typedef unsigned int uint;
 	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
+	*/
+	#ifdef TTMATH_NOASM
+		typedef unsigned long long int ulint;
+	#endif
+
 	/*!
 		how many bits there are in the uint type
 	*/
@@ -143,13 +166,35 @@ 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
+		typedef unsigned long uint;
+		typedef signed   long 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
+	*/
+	//typedef unsigned long long int ulint;
 
 	/*!
 		how many bits there are in the uint type
@@ -174,27 +219,28 @@ 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
 
 
 
@@ -214,10 +260,59 @@ namespace ttmath
 
 
 
+/*!
+	this is a limit when calculating Karatsuba multiplication
+	if the size of a vector is smaller than TTMATH_USE_KARATSUBA_MULTIPLICATION_FROM_SIZE
+	the Karatsuba algorithm will use standard schoolbook multiplication
+*/
+#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
+	#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
 	*/
@@ -248,11 +343,132 @@ 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
 	};
 
 
+	/*!
+		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 print 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;
+
+
+		/*!
+		*/
+		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_exp    = 0;
+		}
+	};
+
+
+
 	/*!
 		this simple class can be used in multithreading model
 		(you can write your own class derived from this one)
@@ -306,7 +522,7 @@ namespace ttmath
 		can throw an exception 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_REFERENCE_ASSERT
 		was used)
 
@@ -326,16 +542,16 @@ namespace ttmath
 			foo.Add(foo);
 		but there are only few methods which can do that
 	*/
-	class ReferenceError : public std::logic_error, ExceptionInfo
+	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)
 		{
 		}
 
@@ -354,20 +570,20 @@ 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)
 	*/
-	class RuntimeError : public std::runtime_error, ExceptionInfo
+	class RuntimeError : public std::runtime_error, public ExceptionInfo
 	{
 	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)
 		{
 		}
 
@@ -407,7 +623,24 @@ namespace ttmath
 	#endif
 
 
+
+	#ifdef TTMATH_DEBUG_LOG
+		#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_x86_64_msvc.asm

@@ -0,0 +1,548 @@
+;
+; This file is a part of TTMath Bignum Library
+; and is distributed under the (new) BSD licence.
+; Author: Christian Kaiser <chk@online.de>
+;
+
+; 
+; Copyright (c) 2009, Christian Kaiser
+; 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 /Zd /Zi ttmathuint_x86_64_msvc.asm
+; compile without debug info: ml64.exe ttmathuint_x86_64_msvc.asm
+; this create ttmathuint_x86_64_msvc.obj file which can be linked with your program
+;
+
+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
+        ; [esp+0x28] = nValue2
+
+		xor		rax, rax			; return value
+		mov		r11, rcx			; table
+		sub		rdx, r8				; rdx = remaining count of uints
+		mov		r10, [esp+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
+        ; [esp+0x28] = result
+
+		mov		r10, [esp+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
+        ; [esp+0x28] = result
+
+		mov		r10, [esp+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