changed: version of the library: 0.9.0 now

git-svn-id: svn://ttmath.org/publicrep/ttmath/trunk@251 e52654a7-88a9-db11-a3e9-0013d4bc506e

CHANGELOG

@@ -1,15 +1,5 @@
-Version 0.9.0 prerelease (2009.07.16):
+Version 0.9.0 (2009.11.25):
-    * added:   support for wide characters (wchar_t)
+    * added:   support for wide characters (wchar_t, std::wstring)
-               wide characters are used when macro TTMATH_USE_WCHAR is defined
-               this macro is defined automatically when there is macro UNICODE or _UNICODE defined
-               some types have been changed
-                 char               -> tt_char
-                 std::string        -> tt_string
-                 std::ostringstream -> tt_ostringstream
-                 std::ostream       -> tt_ostream
-                 std::istream       -> tt_istream
-               normally tt_char is equal char but when you are using wide characters then tt_char will be wchar_t (and so on)
-               (all typedef's are in ttmathtypes.h)
     * added:   Big::IsInteger()
                returns true if the value is integer (without fraction)
                (NaN flag is not checked)
@@ -17,10 +7,117 @@ Version 0.9.0 prerelease (2009.07.16):
     * 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):

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

ttmath/ttmath.h

@@ -48,8 +48,13 @@
 #ifdef _MSC_VER
 //warning C4127: conditional expression is constant
 #pragma warning( disable: 4127 )
+//warning C4702: unreachable code
+#pragma warning( disable: 4702 )
+//warning C4800: forcing value to bool 'true' or 'false' (performance warning)
+#pragma warning( disable: 4800 )
 #endif
 
+
 #include "ttmathbig.h"
 #include "ttmathobjects.h"
 
@@ -1840,20 +1845,7 @@ namespace ttmath
 		return ValueType(); // NaN is set by default
 		}
 
-		if( x.IsZero() )
+		uint c = x.Sqrt();
-		{
-			// Sqrt(0) = 0
-			if( err )
-				*err = err_ok;
-
-		return x;
-		}
-
-		ValueType pow;
-		pow.Set05();
-
-		// PowFrac can return only a carry because x is greater than zero
-		uint c = x.PowFrac(pow);
 
 		if( err )
 			*err = c ? err_overflow : err_ok;
@@ -1866,146 +1858,189 @@ namespace ttmath
 	namespace auxiliaryfunctions
 	{
 
-		template<class ValueType>
+	template<class ValueType>
-		bool RootCheckIndexSign(ValueType & x, const ValueType & index, ErrorCode * err)
+	bool RootCheckIndexSign(ValueType & x, const ValueType & index, ErrorCode * err)
+	{
+		if( index.IsSign() )
 		{
-			if( index.IsSign() )
+			// index cannot be negative
-			{
+			if( err )
-				// index cannot be negative
+				*err = err_improper_argument;
-				if( err )
-					*err = err_improper_argument;
 
-				x.SetNan();
+			x.SetNan();
 
-			return true;
+		return true;
-			}
-
-		return false;
 		}
 
-
+	return false;
-		template<class ValueType>
+	}
-		bool RootCheckIndexZero(ValueType & x, const ValueType & index, ErrorCode * err)
-		{
-			if( index.IsZero() )
-			{
-				if( x.IsZero() )
-				{
-					// there isn't root(0;0) - we assume it's not defined
-					if( err )
-						*err = err_improper_argument;
-
-					x.SetNan();
-
-				return true;
-				}
-		
-				// root(x;0) is 1 (if x!=0)
-				x.SetOne();
-
-				if( err )
-					*err = err_ok;
-
-			return true;
-			}
-
-		return false;
-		}
 
 
-		template<class ValueType>
+	template<class ValueType>
-		bool RootCheckIndexOne(const ValueType & index, ErrorCode * err)
+	bool RootCheckIndexZero(ValueType & x, const ValueType & index, ErrorCode * err)
-		{
+	{
-			ValueType one;
+		if( index.IsZero() )
-			one.SetOne();
-
-			if( index == one )
-			{
-				//root(x;1) is x
-				// we do it because if we used the PowFrac function
-				// we would lose the precision
-				if( err )
-					*err = err_ok;
-
-			return true;
-			}
-
-		return false;
-		}
-
-
-		template<class ValueType>
-		bool RootCheckIndexFrac(ValueType & x, const ValueType & index, ErrorCode * err)
-		{
-			if( !index.IsInteger() )
-			{
-				// index must be integer
-				if( err )
-					*err = err_improper_argument;
-
-				x.SetNan();
-
-			return true;
-			}
-
-		return false;
-		}
-
-
-		template<class ValueType>
-		bool RootCheckXZero(ValueType & x, ErrorCode * err)
 		{
 			if( x.IsZero() )
 			{
-				// root(0;index) is zero (if index!=0)
+				// there isn't root(0;0) - we assume it's not defined
-				// RootCheckIndexZero() must be called beforehand
-				x.SetZero();
-
 				if( err )
-					*err = err_ok;
+					*err = err_improper_argument;
+
+				x.SetNan();
 
 			return true;
 			}
+	
+			// root(x;0) is 1 (if x!=0)
+			x.SetOne();
 
-		return false;
+			if( err )
-		}
+				*err = err_ok;
-
+
-
+		return true;
-		template<class ValueType>
-		bool RootCheckIndex(ValueType & x, const ValueType & index, ErrorCode * err, bool * change_sign)
-		{
-			*change_sign = false;
-
-			if( index.Mod2() )
-			{
-				// index is odd (1,3,5...)
-				if( x.IsSign() )
-				{
-					*change_sign = true;
-					x.Abs();
-				}
-			}
-			else
-			{
-				// index is even
-				// x cannot be negative
-				if( x.IsSign() )
-				{
-					if( err )
-						*err = err_improper_argument;
-
-					x.SetNan();
-
-					return true;
-				}
-			}
-
-		return false;
 		}
 
+	return false;
 	}
 
 
+	template<class ValueType>
+	bool RootCheckIndexOne(const ValueType & index, ErrorCode * err)
+	{
+		ValueType one;
+		one.SetOne();
+
+		if( index == one )
+		{
+			//root(x;1) is x
+			// we do it because if we used the PowFrac function
+			// we would lose the precision
+			if( err )
+				*err = err_ok;
+
+		return true;
+		}
+
+	return false;
+	}
+
+
+	template<class ValueType>
+	bool RootCheckIndexTwo(ValueType & x, const ValueType & index, ErrorCode * err)
+	{
+		if( index == 2 )
+		{
+			x = Sqrt(x, err);
+
+		return true;
+		}
+
+	return false;
+	}
+
+
+	template<class ValueType>
+	bool RootCheckIndexFrac(ValueType & x, const ValueType & index, ErrorCode * err)
+	{
+		if( !index.IsInteger() )
+		{
+			// index must be integer
+			if( err )
+				*err = err_improper_argument;
+
+			x.SetNan();
+
+		return true;
+		}
+
+	return false;
+	}
+
+
+	template<class ValueType>
+	bool RootCheckXZero(ValueType & x, ErrorCode * err)
+	{
+		if( x.IsZero() )
+		{
+			// root(0;index) is zero (if index!=0)
+			// RootCheckIndexZero() must be called beforehand
+			x.SetZero();
+
+			if( err )
+				*err = err_ok;
+
+		return true;
+		}
+
+	return false;
+	}
+
+
+	template<class ValueType>
+	bool RootCheckIndex(ValueType & x, const ValueType & index, ErrorCode * err, bool * change_sign)
+	{
+		*change_sign = false;
+
+		if( index.Mod2() )
+		{
+			// index is odd (1,3,5...)
+			if( x.IsSign() )
+			{
+				*change_sign = true;
+				x.Abs();
+			}
+		}
+		else
+		{
+			// index is even
+			// x cannot be negative
+			if( x.IsSign() )
+			{
+				if( err )
+					*err = err_improper_argument;
+
+				x.SetNan();
+
+				return true;
+			}
+		}
+
+	return false;
+	}
+
+
+	template<class ValueType>
+	uint RootCorrectInteger(ValueType & old_x, ValueType & x, const ValueType & index)
+	{
+		if( !old_x.IsInteger() || x.IsInteger() || !index.exponent.IsSign() )
+			return 0;
+
+		// old_x is integer,
+		// x is not integer,
+		// index is relatively small (index.exponent<0 or index.exponent<=0)
+		// (because we're using a special powering algorithm Big::PowUInt())
+
+		uint c = 0;
+
+		ValueType temp(x);
+		c += temp.Round();
+
+		ValueType temp_round(temp);
+		c += temp.PowUInt(index);
+
+		if( temp == old_x )
+			x = temp_round;
+
+	return (c==0)? 0 : 1;
+	}
+
+
+
+	} // namespace auxiliaryfunctions 
+
+
+
 	/*!
 		indexth Root of x
 		index must be integer and not negative <0;1;2;3....)
@@ -2036,30 +2071,33 @@ namespace ttmath
 		if( RootCheckIndexSign(x, index, err) ) return x;
 		if( RootCheckIndexZero(x, index, err) ) return x;
 		if( RootCheckIndexOne (   index, err) ) return x;
+		if( RootCheckIndexTwo (x, index, err) ) return x;
 		if( RootCheckIndexFrac(x, index, err) ) return x;
 		if( RootCheckXZero    (x,        err) ) return x;
 
 		// index integer and index!=0
 		// x!=0
 
-		uint c = 0;
+		ValueType old_x(x);
 		bool change_sign;
+
 		if( RootCheckIndex(x, index, err, &change_sign ) ) return x;
 
-		ValueType newindex;
+		ValueType temp;
-		newindex.SetOne();
+		uint c = 0;
-		c += newindex.Div(index);
+
-		c += x.PowFrac(newindex); // here can only be a carry
+		// we're using the formula: root(x ; n) = exp( ln(x) / n )
+		c += temp.Ln(x);
+		c += temp.Div(index);
+		c += x.Exp(temp);
 
 		if( change_sign )
 		{
-			// the value of x should be different from zero
+			// x is different from zero
-			// (x is actually tested by RootCheckXZero)
-			TTMATH_ASSERT( x.IsZero() == false )
-
 			x.SetSign();
 		}
 
+		c += RootCorrectInteger(old_x, x, index);
 
 		if( err )
 			*err = c ? err_overflow : err_ok;
@@ -2681,7 +2719,9 @@ namespace ttmath
 		   x! = gamma(x+1)
 	*/
 	template<class ValueType>
-	ValueType Factorial2(ValueType x, CGamma<ValueType> * cgamma = 0, ErrorCode * err = 0,
+	ValueType Factorial2(ValueType x,
+						 CGamma<ValueType> * cgamma = 0,
+						 ErrorCode * err = 0,
 						 const volatile StopCalculating * stop = 0)
 	{
 	ValueType result, one;
@@ -2715,7 +2755,7 @@ namespace ttmath
 		it's multithread safe, you should create a CGamma<> object and use it whenever you call the Factorial()
 		e.g.
 			typedef Big<1,2> MyBig;
-			MyBig x=234, y=345.53;
+			MyBig x=234, y=54345;
 			CGamma<MyBig> cgamma;
 			std::cout << Factorial(x, cgamma) << std::endl;
 			std::cout << Factorial(y, cgamma) << std::endl;
@@ -2742,7 +2782,7 @@ namespace ttmath
 	template<class ValueType>
 	ValueType Factorial(const ValueType & x, ErrorCode * err = 0)
 	{
-		return auxiliaryfunctions::Factorial2(x, 0, err, 0);
+		return auxiliaryfunctions::Factorial2(x, (CGamma<ValueType>*)0, err, 0);
 	}
 
 
@@ -2784,8 +2824,12 @@ namespace ttmath
 
 
 #ifdef _MSC_VER
-#pragma warning( default: 4127 )
 //warning C4127: conditional expression is constant
+#pragma warning( default: 4127 )
+//warning C4702: unreachable code
+#pragma warning( default: 4702 )
+//warning C4800: forcing value to bool 'true' or 'false' (performance warning)
+#pragma warning( default: 4800 )
 #endif
 
 #endif

ttmath/ttmathint.h

@@ -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
@@ -821,7 +880,16 @@ public:
 	/*!
 		a constructor for converting string to this class (with the base=10)
 	*/
-	Int(const tt_char * s)
+	Int(const char * s)
+	{
+		FromString(s);
+	}
+
+
+	/*!
+		a constructor for converting string to this class (with the base=10)
+	*/
+	Int(const wchar_t * s)
 	{
 		FromString(s);
 	}
@@ -830,7 +898,16 @@ public:
 	/*!
 		a constructor for converting a string to this class (with the base=10)
 	*/
-	Int(const tt_string & s)
+	Int(const std::string & s)
+	{
+		FromString( s.c_str() );
+	}
+
+
+	/*!
+		a constructor for converting a string to this class (with the base=10)
+	*/
+	Int(const std::wstring & s)
 	{
 		FromString( s.c_str() );
 	}
@@ -866,10 +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(tt_string & result, uint b = 10) const
+	template<class string_type>
+	void ToStringBase(string_type & result, uint b = 10) const
 	{
 		if( IsSign() )
 		{
@@ -885,42 +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
+		an auxiliary method for converting from a string
-		string is given either as 'const char *' or 'const wchar_t *'
-
-		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 tt_char * s, uint b = 10, const tt_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) )
@@ -959,11 +1068,54 @@ 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 tt_string & s, uint b = 10)
+	uint FromString(const std::string & s, uint b = 10)
+	{
+		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 );
 	}
@@ -972,7 +1124,7 @@ public:
 	/*!
 		this operator converts a string into its value (with base = 10)
 	*/
-	Int<value_size> & operator=(const tt_char * s)
+	Int<value_size> & operator=(const char * s)
 	{
 		FromString(s);
 
@@ -983,7 +1135,18 @@ public:
 	/*!
 		this operator converts a string into its value (with base = 10)
 	*/
-	Int<value_size> & operator=(const tt_string & s)
+	Int<value_size> & operator=(const wchar_t * s)
+	{
+		FromString(s);
+
+	return *this;
+	}
+
+
+	/*!
+		this operator converts a string into its value (with base = 10)
+	*/
+	Int<value_size> & operator=(const std::string & s)
 	{
 		FromString( s.c_str() );
 
@@ -991,6 +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;
+	}
 
 
 	/*!
@@ -1270,14 +1442,15 @@ public:
 	*
 	*/
 
-	/*!
+private:
-		output for standard streams
 
-		tt_ostream is either std::ostream or std::wostream
+	/*!
+		an auxiliary method for outputing to standard streams
 	*/
-	friend tt_ostream & operator<<(tt_ostream & s, const Int<value_size> & l)
+	template<class ostream_type, class string_type>
+	static ostream_type & OutputToStream(ostream_type & s, const Int<value_size> & l)
 	{
-	tt_string ss;
+	string_type ss;
 
 		l.ToString(ss);
 		s << ss;
@@ -1286,17 +1459,41 @@ public:
 	}
 
 
-	/*!
-		input from standard streams
 
-		tt_istream is either std::istream or std::wistream
+public:
+
+
+	/*!
+		output to standard streams
 	*/
-	friend tt_istream & operator>>(tt_istream & s, Int<value_size> & l)
+	friend std::ostream & operator<<(std::ostream & s, const Int<value_size> & l)
 	{
-	tt_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;
 	
-	// tt_char for operator>>
+	// char or wchar_t for operator>>
-	tt_char z;
+	char_type z;
 	
 		// operator>> omits white characters if they're set for ommiting
 		s >> z;
@@ -1308,10 +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 = static_cast<tt_char>(s.get());
+			z = static_cast<char_type>(s.get());
 		}
 
 		// we're leaving the last readed character
@@ -1323,6 +1520,26 @@ 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

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

@@ -44,13 +44,13 @@
     \brief Mathematic functions.
 */
 
-#include "ttmathtypes.h"
-
 #include <string>
 #include <vector>
 #include <list>
 #include <map>
 
+#include "ttmathtypes.h"
+#include "ttmathmisc.h"
 
 
 namespace ttmath
@@ -74,19 +74,19 @@ public:
 	struct Item
 	{
 		// name of a variable of a function
-		// (either std::string or std::wstring)
+		// internally we store variables and funcions as std::string (not std::wstring even when wide characters are used)
-		tt_string value;
+		std::string value;
 
 		// number of parameters required by the function
 		// (if there's a variable this 'param' is ignored)
 		int param;
 
 		Item() {}
-		Item(const tt_string & v, int p) : value(v), param(p) {}
+		Item(const std::string & v, int p) : value(v), param(p) {}
 	};
 
 	// 'Table' is the type of our table
-	typedef std::map<tt_string, Item> Table;
+	typedef std::map<std::string, Item> Table;
 	typedef	Table::iterator Iterator;
 	typedef	Table::const_iterator CIterator;
 
@@ -99,7 +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;
@@ -114,7 +114,8 @@ public:
 	/*!
 		this method returns true if the name can be as a name of an object
 	*/
-	static bool IsNameCorrect(const tt_string & name)
+	template<class string_type>
+	static bool IsNameCorrect(const string_type & name)
 	{
 		if( name.empty() )
 			return false;
@@ -122,7 +123,7 @@ public:
 		if( !CorrectCharacter(name[0], false) )
 			return false;
 
-		tt_string::const_iterator i=name.begin();
+		typename string_type::const_iterator i = name.begin();
 
 		for(++i ; i!=name.end() ; ++i)
 			if( !CorrectCharacter(*i, true) )
@@ -135,7 +136,7 @@ public:
 	/*!
 		this method returns true if such an object is defined (name exists)
 	*/
-	bool IsDefined(const tt_string & name)
+	bool IsDefined(const std::string & name)
 	{
 		Iterator i = table.find(name);
 
@@ -147,10 +148,26 @@ 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
 	*/
-	ErrorCode Add(const tt_string & name, const tt_string & value, int param = 0)
+	ErrorCode Add(const std::string & name, const std::string & value, int param = 0)
 	{
 		if( !IsNameCorrect(name) )
 			return err_incorrect_name;
@@ -167,6 +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
 	*/
@@ -207,7 +241,7 @@ public:
 	/*!
 		this method changes the value and the number of parameters for a specific object
 	*/
-	ErrorCode EditValue(const tt_string & name, const tt_string & value, int param = 0)
+	ErrorCode EditValue(const std::string & name, const std::string & value, int param = 0)
 	{
 		if( !IsNameCorrect(name) )
 			return err_incorrect_name;
@@ -224,10 +258,27 @@ 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
 	*/
-	ErrorCode EditName(const tt_string & old_name, const tt_string & new_name)
+	ErrorCode EditName(const std::string & old_name, const std::string & new_name)
 	{
 		if( !IsNameCorrect(old_name) || !IsNameCorrect(new_name) )
 			return err_incorrect_name;
@@ -255,10 +306,27 @@ 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
 	*/
-	ErrorCode Delete(const tt_string & name)
+	ErrorCode Delete(const std::string & name)
 	{
 		if( !IsNameCorrect(name) )
 			return err_incorrect_name;
@@ -274,10 +342,26 @@ 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
 	*/
-	ErrorCode GetValue(const tt_string & name, tt_string & value) const
+	ErrorCode GetValue(const std::string & name, std::string & value) const
 	{
 		if( !IsNameCorrect(name) )
 			return err_incorrect_name;
@@ -296,11 +380,29 @@ 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)
 	*/
-	ErrorCode GetValue(const tt_string & name, const tt_char ** value) const
+	ErrorCode GetValue(const std::string & name, const char ** value) const
 	{
 		if( !IsNameCorrect(name) )
 			return err_incorrect_name;
@@ -319,11 +421,28 @@ 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
 	*/
-	ErrorCode GetValueAndParam(const tt_string & name, tt_string & value, int * param) const
+	ErrorCode GetValueAndParam(const std::string & name, std::string & value, int * param) const
 	{
 		if( !IsNameCorrect(name) )
 			return err_incorrect_name;
@@ -344,12 +463,31 @@ 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
 		(this version is used for not copying the whole string)
 	*/
-	ErrorCode GetValueAndParam(const tt_string & name, const tt_char ** value, int * param) const
+	ErrorCode GetValueAndParam(const std::string & name, const char ** value, int * param) const
 	{
 		if( !IsNameCorrect(name) )
 			return err_incorrect_name;
@@ -370,6 +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
 	*/
@@ -382,6 +539,7 @@ public:
 private:
 
 	Table table;
+	std::string str_tmp1, str_tmp2;
 
 }; // end of class Objects
 
@@ -428,7 +586,7 @@ 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()
 	{
@@ -591,8 +749,8 @@ struct CGamma
 		you don't have to call this method, these coefficients will be automatically calculated
 		when they are needed
 
-		you must note that calculating of the coefficients is a little time-consuming operation,
+		you must note that calculating these coefficients is a little time-consuming operation,
-		(especially when the mantissa is large) and first called to Gamma() or Factorial()
+		(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
 	*/

ttmath/ttmathparser.h

@@ -52,6 +52,7 @@
 
 #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)
+		"2,1234"    (they are the same, by default we can either use a comma or a dot)
-		            (look at the macro TTMATH_COMMA_CHARACTER_2)
 		"1 + 2"
 		"(1 + 2) * 3"
 		"pi"
@@ -255,7 +255,7 @@ public:
 		bool function;
 
 		// if function is true
-		tt_string function_name;
+		std::string function_name;
 
 		/*
 			the sign of value
@@ -314,7 +314,7 @@ ErrorCode error;
 
 	when an error has occured it may be used to count the index of the wrong character
 */
-const tt_char * pstring;
+const char * pstring;
 
 
 /*!
@@ -351,7 +351,7 @@ const Objects * puser_variables;
 const Objects * puser_functions;
 
 
-typedef std::map<tt_string, ValueType> FunctionLocalVariables;
+typedef std::map<std::string, ValueType> FunctionLocalVariables;
 
 /*!
 	a pointer to the local variables of a function
@@ -362,13 +362,13 @@ const FunctionLocalVariables * pfunction_local_variables;
 /*!
 	a temporary set using during parsing user defined variables
 */
-std::set<tt_string> visited_variables;
+std::set<std::string> visited_variables;
 
 
 /*!
 	a temporary set using during parsing user defined functions
 */
-std::set<tt_string> visited_functions;
+std::set<std::string> visited_functions;
 
 
 
@@ -396,10 +396,10 @@ typedef void (ValueType::*pfunction_var)();
 	table of mathematic functions
 
 	this map consists of:
-		tt_string - function's name
+		std::string - function's name
 		pfunction - pointer to specific function
 */
-typedef std::map<tt_string, pfunction> FunctionsTable;
+typedef std::map<std::string, pfunction> FunctionsTable;
 FunctionsTable functions_table;
 
 
@@ -407,10 +407,10 @@ FunctionsTable functions_table;
 	table of mathematic operators
 
 	this map consists of:
-		tt_string - operators's name
+		std::string - operators's name
 		MatOperator::Type - type of the operator
 */
-typedef std::map<tt_string, typename MatOperator::Type> OperatorsTable;
+typedef std::map<std::string, typename MatOperator::Type> OperatorsTable;
 OperatorsTable operators_table;
 
 
@@ -418,10 +418,10 @@ OperatorsTable operators_table;
 	table of mathematic variables
 
 	this map consists of:
-		tt_string     - variable's name
+		std::string     - variable's name
 		pfunction_var - pointer to specific function which returns value of variable
 */
-typedef std::map<tt_string, pfunction_var> VariablesTable;
+typedef std::map<std::string, pfunction_var> VariablesTable;
 VariablesTable variables_table;
 
 
@@ -431,6 +431,39 @@ VariablesTable variables_table;
 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;
+
 /*!
 	we're using this method for reporting an error
 */
@@ -455,7 +488,7 @@ void SkipWhiteCharacters()
 /*!
 	an auxiliary method for RecurrenceParsingVariablesOrFunction(...)
 */
-void RecurrenceParsingVariablesOrFunction_CheckStopCondition(bool variable, const tt_string & name)
+void RecurrenceParsingVariablesOrFunction_CheckStopCondition(bool variable, const std::string & name)
 {
 	if( variable )
 	{
@@ -473,7 +506,7 @@ void RecurrenceParsingVariablesOrFunction_CheckStopCondition(bool variable, cons
 /*!
 	an auxiliary method for RecurrenceParsingVariablesOrFunction(...)
 */
-void RecurrenceParsingVariablesOrFunction_AddName(bool variable, const tt_string & name)
+void RecurrenceParsingVariablesOrFunction_AddName(bool variable, const std::string & name)
 {
 	if( variable )
 		visited_variables.insert( name );
@@ -485,7 +518,7 @@ void RecurrenceParsingVariablesOrFunction_AddName(bool variable, const tt_string
 /*!
 	an auxiliary method for RecurrenceParsingVariablesOrFunction(...)
 */
-void RecurrenceParsingVariablesOrFunction_DeleteName(bool variable, const tt_string & name)
+void RecurrenceParsingVariablesOrFunction_DeleteName(bool variable, const std::string & name)
 {
 	if( variable )
 		visited_variables.erase( name );
@@ -504,7 +537,7 @@ void RecurrenceParsingVariablesOrFunction_DeleteName(bool variable, const tt_str
 	(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 tt_string & name, const tt_char * new_string,
+ValueType RecurrenceParsingVariablesOrFunction(bool variable, const std::string & name, const char * new_string,
 											   FunctionLocalVariables * local_variables = 0)
 {
 	RecurrenceParsingVariablesOrFunction_CheckStopCondition(variable, name);
@@ -548,17 +581,18 @@ public:
 /*!
 	this method returns the user-defined value of a variable
 */
-bool GetValueOfUserDefinedVariable(const tt_string & variable_name,ValueType & result)
+bool GetValueOfUserDefinedVariable(const std::string & variable_name,ValueType & result)
 {
 	if( !puser_variables )
 		return false;
 
-	const tt_char * string_value;
+	const char * string_value;
 
 	if( puser_variables->GetValue(variable_name, &string_value) != err_ok )
 		return false;
 
 	result = RecurrenceParsingVariablesOrFunction(true, variable_name, string_value);
+	calculated = true;
 
 return true;
 }
@@ -567,7 +601,7 @@ return true;
 /*!
 	this method returns the value of a local variable of a function
 */
-bool GetValueOfFunctionLocalVariable(const tt_string & variable_name, ValueType & result)
+bool GetValueOfFunctionLocalVariable(const std::string & variable_name, ValueType & result)
 {
 	if( !pfunction_local_variables )
 		return false;
@@ -589,7 +623,7 @@ return true;
 	we make an object of type ValueType then call a method which 
 	sets the correct value in it and finally we'll return the object
 */
-ValueType GetValueOfVariable(const tt_string & variable_name)
+ValueType GetValueOfVariable(const std::string & variable_name)
 {
 ValueType result;
 
@@ -600,13 +634,14 @@ ValueType result;
 		return result;
 
 
-	typename std::map<tt_string, pfunction_var>::iterator i =
+	typename std::map<std::string, pfunction_var>::iterator i =
 													variables_table.find(variable_name);
 
 	if( i == variables_table.end() )
 		Error( err_unknown_variable );
 
 	(result.*(i->second))();
+	calculated = true;
 
 return result;
 }
@@ -1355,34 +1390,56 @@ 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(tt_char * buffer, int par)
+void Sprintf(char * buffer, int par)
 {
-char buf[30]; // char, not tt_char
+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
 
 	(look at the description in 'CallFunction(...)')
 */
-bool GetValueOfUserDefinedFunction(const tt_string & function_name, int amount_of_args, int sindex)
+bool GetValueOfUserDefinedFunction(const std::string & function_name, int amount_of_args, int sindex)
 {
 	if( !puser_functions )
 		return false;
 
-	const tt_char * string_value;
+	const char * string_value;
 	int param;
 
 	if( puser_functions->GetValueAndParam(function_name, &string_value, &param) != err_ok )
@@ -1396,7 +1453,7 @@ bool GetValueOfUserDefinedFunction(const tt_string & function_name, int amount_o
 
 	if( amount_of_args > 0 )
 	{
-		tt_char buffer[30];
+		char buffer[30];
 
 		// x = x1
 		buffer[0] = 'x';
@@ -1412,6 +1469,7 @@ bool GetValueOfUserDefinedFunction(const tt_string & function_name, int amount_o
 	}
 
 	stack[sindex-1].value = RecurrenceParsingVariablesOrFunction(false, function_name, string_value, &local_variables);
+	calculated = true;
 
 return true;
 }
@@ -1430,7 +1488,7 @@ return true;
 	result will be stored in 'stack[sindex-1].value'
 	(we don't have to set the correct type of this element, it'll be set later)
 */
-void CallFunction(const tt_string & function_name, int amount_of_args, int sindex)
+void CallFunction(const std::string & function_name, int amount_of_args, int sindex)
 {
 	if( GetValueOfUserDefinedFunction(function_name, amount_of_args, sindex) )
 		return;
@@ -1444,6 +1502,7 @@ void CallFunction(const tt_string & function_name, int amount_of_args, int sinde
 		calling the specify function
 	*/
 	(this->*(i->second))(sindex, amount_of_args, stack[sindex-1].value);
+	calculated = true;
 }
 
 
@@ -1456,12 +1515,16 @@ void CallFunction(const tt_string & function_name, int amount_of_args, int sinde
 	function_name - name of the function
 	pf - pointer to the function (to the wrapper)
 */
-void InsertFunctionToTable(const tt_char * function_name, pfunction pf)
+void InsertFunctionToTable(const char * function_name, pfunction pf)
 {
-	functions_table.insert( std::make_pair(tt_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)
@@ -1469,9 +1532,12 @@ void InsertFunctionToTable(const tt_char * function_name, pfunction pf)
 	variable_name - name of the function
 	pf - pointer to the function
 */
-void InsertVariableToTable(const tt_char * variable_name, pfunction_var pf)
+void InsertVariableToTable(const char * variable_name, pfunction_var pf)
 {
-	variables_table.insert( std::make_pair(tt_string(variable_name), pf));
+	std::string str;
+	Misc::AssignString(str, variable_name);
+
+	variables_table.insert( std::make_pair(str, pf) );
 }
 
 
@@ -1480,65 +1546,66 @@ void InsertVariableToTable(const tt_char * variable_name, pfunction_var pf)
 */
 void CreateFunctionsTable()
 {
-	InsertFunctionToTable(TTMATH_TEXT("gamma"),		&Parser<ValueType>::Gamma);
+	InsertFunctionToTable("gamma",		&Parser<ValueType>::Gamma);
-	InsertFunctionToTable(TTMATH_TEXT("factorial"),	&Parser<ValueType>::Factorial);
+	InsertFunctionToTable("factorial",	&Parser<ValueType>::Factorial);
-	InsertFunctionToTable(TTMATH_TEXT("abs"),   	&Parser<ValueType>::Abs);
+	InsertFunctionToTable("abs",   		&Parser<ValueType>::Abs);
-	InsertFunctionToTable(TTMATH_TEXT("sin"),   	&Parser<ValueType>::Sin);
+	InsertFunctionToTable("sin",   		&Parser<ValueType>::Sin);
-	InsertFunctionToTable(TTMATH_TEXT("cos"),   	&Parser<ValueType>::Cos);
+	InsertFunctionToTable("cos",   		&Parser<ValueType>::Cos);
-	InsertFunctionToTable(TTMATH_TEXT("tan"),   	&Parser<ValueType>::Tan);
+	InsertFunctionToTable("tan",   		&Parser<ValueType>::Tan);
-	InsertFunctionToTable(TTMATH_TEXT("tg"),		&Parser<ValueType>::Tan);
+	InsertFunctionToTable("tg",			&Parser<ValueType>::Tan);
-	InsertFunctionToTable(TTMATH_TEXT("cot"),  		&Parser<ValueType>::Cot);
+	InsertFunctionToTable("cot",  		&Parser<ValueType>::Cot);
-	InsertFunctionToTable(TTMATH_TEXT("ctg"),  		&Parser<ValueType>::Cot);
+	InsertFunctionToTable("ctg",  		&Parser<ValueType>::Cot);
-	InsertFunctionToTable(TTMATH_TEXT("int"),	   	&Parser<ValueType>::Int);
+	InsertFunctionToTable("int",	   	&Parser<ValueType>::Int);
-	InsertFunctionToTable(TTMATH_TEXT("round"),	 	&Parser<ValueType>::Round);
+	InsertFunctionToTable("round",	 	&Parser<ValueType>::Round);
-	InsertFunctionToTable(TTMATH_TEXT("ln"),	    &Parser<ValueType>::Ln);
+	InsertFunctionToTable("ln",			&Parser<ValueType>::Ln);
-	InsertFunctionToTable(TTMATH_TEXT("log"),	   	&Parser<ValueType>::Log);
+	InsertFunctionToTable("log",	   	&Parser<ValueType>::Log);
-	InsertFunctionToTable(TTMATH_TEXT("exp"),	   	&Parser<ValueType>::Exp);
+	InsertFunctionToTable("exp",	   	&Parser<ValueType>::Exp);
-	InsertFunctionToTable(TTMATH_TEXT("max"),	   	&Parser<ValueType>::Max);
+	InsertFunctionToTable("max",	   	&Parser<ValueType>::Max);
-	InsertFunctionToTable(TTMATH_TEXT("min"),	   	&Parser<ValueType>::Min);
+	InsertFunctionToTable("min",	   	&Parser<ValueType>::Min);
-	InsertFunctionToTable(TTMATH_TEXT("asin"),   	&Parser<ValueType>::ASin);
+	InsertFunctionToTable("asin",   	&Parser<ValueType>::ASin);
-	InsertFunctionToTable(TTMATH_TEXT("acos"),   	&Parser<ValueType>::ACos);
+	InsertFunctionToTable("acos",   	&Parser<ValueType>::ACos);
-	InsertFunctionToTable(TTMATH_TEXT("atan"),   	&Parser<ValueType>::ATan);
+	InsertFunctionToTable("atan",   	&Parser<ValueType>::ATan);
-	InsertFunctionToTable(TTMATH_TEXT("atg"),	   	&Parser<ValueType>::ATan);
+	InsertFunctionToTable("atg",	   	&Parser<ValueType>::ATan);
-	InsertFunctionToTable(TTMATH_TEXT("acot"),   	&Parser<ValueType>::ACot);
+	InsertFunctionToTable("acot",   	&Parser<ValueType>::ACot);
-	InsertFunctionToTable(TTMATH_TEXT("actg"),   	&Parser<ValueType>::ACot);
+	InsertFunctionToTable("actg",   	&Parser<ValueType>::ACot);
-	InsertFunctionToTable(TTMATH_TEXT("sgn"),   	&Parser<ValueType>::Sgn);
+	InsertFunctionToTable("sgn",   		&Parser<ValueType>::Sgn);
-	InsertFunctionToTable(TTMATH_TEXT("mod"),   	&Parser<ValueType>::Mod);
+	InsertFunctionToTable("mod",   		&Parser<ValueType>::Mod);
-	InsertFunctionToTable(TTMATH_TEXT("if"),   		&Parser<ValueType>::If);
+	InsertFunctionToTable("if",   		&Parser<ValueType>::If);
-	InsertFunctionToTable(TTMATH_TEXT("or"),   		&Parser<ValueType>::Or);
+	InsertFunctionToTable("or",   		&Parser<ValueType>::Or);
-	InsertFunctionToTable(TTMATH_TEXT("and"),  		&Parser<ValueType>::And);
+	InsertFunctionToTable("and",  		&Parser<ValueType>::And);
-	InsertFunctionToTable(TTMATH_TEXT("not"),  		&Parser<ValueType>::Not);
+	InsertFunctionToTable("not",  		&Parser<ValueType>::Not);
-	InsertFunctionToTable(TTMATH_TEXT("degtorad"),	&Parser<ValueType>::DegToRad);
+	InsertFunctionToTable("degtorad",	&Parser<ValueType>::DegToRad);
-	InsertFunctionToTable(TTMATH_TEXT("radtodeg"),	&Parser<ValueType>::RadToDeg);
+	InsertFunctionToTable("radtodeg",	&Parser<ValueType>::RadToDeg);
-	InsertFunctionToTable(TTMATH_TEXT("degtodeg"),	&Parser<ValueType>::DegToDeg);
+	InsertFunctionToTable("degtodeg",	&Parser<ValueType>::DegToDeg);
-	InsertFunctionToTable(TTMATH_TEXT("gradtorad"),	&Parser<ValueType>::GradToRad);
+	InsertFunctionToTable("gradtorad",	&Parser<ValueType>::GradToRad);
-	InsertFunctionToTable(TTMATH_TEXT("radtograd"),	&Parser<ValueType>::RadToGrad);
+	InsertFunctionToTable("radtograd",	&Parser<ValueType>::RadToGrad);
-	InsertFunctionToTable(TTMATH_TEXT("degtograd"),	&Parser<ValueType>::DegToGrad);
+	InsertFunctionToTable("degtograd",	&Parser<ValueType>::DegToGrad);
-	InsertFunctionToTable(TTMATH_TEXT("gradtodeg"),	&Parser<ValueType>::GradToDeg);
+	InsertFunctionToTable("gradtodeg",	&Parser<ValueType>::GradToDeg);
-	InsertFunctionToTable(TTMATH_TEXT("ceil"),		&Parser<ValueType>::Ceil);
+	InsertFunctionToTable("ceil",		&Parser<ValueType>::Ceil);
-	InsertFunctionToTable(TTMATH_TEXT("floor"),		&Parser<ValueType>::Floor);
+	InsertFunctionToTable("floor",		&Parser<ValueType>::Floor);
-	InsertFunctionToTable(TTMATH_TEXT("sqrt"),		&Parser<ValueType>::Sqrt);
+	InsertFunctionToTable("sqrt",		&Parser<ValueType>::Sqrt);
-	InsertFunctionToTable(TTMATH_TEXT("sinh"),		&Parser<ValueType>::Sinh);
+	InsertFunctionToTable("sinh",		&Parser<ValueType>::Sinh);
-	InsertFunctionToTable(TTMATH_TEXT("cosh"),		&Parser<ValueType>::Cosh);
+	InsertFunctionToTable("cosh",		&Parser<ValueType>::Cosh);
-	InsertFunctionToTable(TTMATH_TEXT("tanh"),		&Parser<ValueType>::Tanh);
+	InsertFunctionToTable("tanh",		&Parser<ValueType>::Tanh);
-	InsertFunctionToTable(TTMATH_TEXT("tgh"),		&Parser<ValueType>::Tanh);
+	InsertFunctionToTable("tgh",		&Parser<ValueType>::Tanh);
-	InsertFunctionToTable(TTMATH_TEXT("coth"),		&Parser<ValueType>::Coth);
+	InsertFunctionToTable("coth",		&Parser<ValueType>::Coth);
-	InsertFunctionToTable(TTMATH_TEXT("ctgh"),		&Parser<ValueType>::Coth);
+	InsertFunctionToTable("ctgh",		&Parser<ValueType>::Coth);
-	InsertFunctionToTable(TTMATH_TEXT("root"),		&Parser<ValueType>::Root);
+	InsertFunctionToTable("root",		&Parser<ValueType>::Root);
-	InsertFunctionToTable(TTMATH_TEXT("asinh"),		&Parser<ValueType>::ASinh);
+	InsertFunctionToTable("asinh",		&Parser<ValueType>::ASinh);
-	InsertFunctionToTable(TTMATH_TEXT("acosh"),		&Parser<ValueType>::ACosh);
+	InsertFunctionToTable("acosh",		&Parser<ValueType>::ACosh);
-	InsertFunctionToTable(TTMATH_TEXT("atanh"),		&Parser<ValueType>::ATanh);
+	InsertFunctionToTable("atanh",		&Parser<ValueType>::ATanh);
-	InsertFunctionToTable(TTMATH_TEXT("atgh"),		&Parser<ValueType>::ATanh);
+	InsertFunctionToTable("atgh",		&Parser<ValueType>::ATanh);
-	InsertFunctionToTable(TTMATH_TEXT("acoth"),		&Parser<ValueType>::ACoth);
+	InsertFunctionToTable("acoth",		&Parser<ValueType>::ACoth);
-	InsertFunctionToTable(TTMATH_TEXT("actgh"),		&Parser<ValueType>::ACoth);
+	InsertFunctionToTable("actgh",		&Parser<ValueType>::ACoth);
-	InsertFunctionToTable(TTMATH_TEXT("bitand"),	&Parser<ValueType>::BitAnd);
+	InsertFunctionToTable("bitand",		&Parser<ValueType>::BitAnd);
-	InsertFunctionToTable(TTMATH_TEXT("bitor"),		&Parser<ValueType>::BitOr);
+	InsertFunctionToTable("bitor",		&Parser<ValueType>::BitOr);
-	InsertFunctionToTable(TTMATH_TEXT("bitxor"),	&Parser<ValueType>::BitXor);
+	InsertFunctionToTable("bitxor",		&Parser<ValueType>::BitXor);
-	InsertFunctionToTable(TTMATH_TEXT("band"),		&Parser<ValueType>::BitAnd);
+	InsertFunctionToTable("band",		&Parser<ValueType>::BitAnd);
-	InsertFunctionToTable(TTMATH_TEXT("bor"),		&Parser<ValueType>::BitOr);
+	InsertFunctionToTable("bor",		&Parser<ValueType>::BitOr);
-	InsertFunctionToTable(TTMATH_TEXT("bxor"),		&Parser<ValueType>::BitXor);
+	InsertFunctionToTable("bxor",		&Parser<ValueType>::BitXor);
-	InsertFunctionToTable(TTMATH_TEXT("sum"),		&Parser<ValueType>::Sum);
+	InsertFunctionToTable("sum",		&Parser<ValueType>::Sum);
-	InsertFunctionToTable(TTMATH_TEXT("avg"),		&Parser<ValueType>::Avg);
+	InsertFunctionToTable("avg",		&Parser<ValueType>::Avg);
+	InsertFunctionToTable("frac",		&Parser<ValueType>::Frac);
 }
 
 
@@ -1547,8 +1614,8 @@ void CreateFunctionsTable()
 */
 void CreateVariablesTable()
 {
-	InsertVariableToTable(TTMATH_TEXT("pi"), &ValueType::SetPi);
+	InsertVariableToTable("pi", &ValueType::SetPi);
-	InsertVariableToTable(TTMATH_TEXT("e"),  &ValueType::SetE);
+	InsertVariableToTable("e",  &ValueType::SetE);
 }
 
 
@@ -1574,7 +1641,7 @@ return c;
 	what should be returned is tested just by a '(' character that means if there's
 	a '(' character after a name that function returns 'true'
 */
-bool ReadName(tt_string & result)
+bool ReadName(std::string & result)
 {
 int character;
 
@@ -1591,7 +1658,7 @@ int character;
 
 	do
 	{
-		result   += static_cast<tt_char>( character );
+		result   += static_cast<char>( character );
 		character = * ++pstring;
 	}
 	while(	(character>='a' && character<='z') ||
@@ -1646,7 +1713,7 @@ return false;
 */
 bool ReadVariableOrFunction(Item & result)
 {
-tt_string name;
+std::string name;
 bool is_it_name_of_function = ReadName(name);
 
 	if( is_it_name_of_function )
@@ -1675,11 +1742,17 @@ return is_it_name_of_function;
 */
 void ReadValue(Item & result, int reading_base)
 {
-const tt_char * new_stack_pointer;
+const char * new_stack_pointer;
 bool value_read;
+Conv conv;
 
-	int carry = result.value.FromString(pstring, reading_base, &new_stack_pointer, &value_read);
+	conv.base   = base;
-	pstring   = new_stack_pointer;
+	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 );
@@ -1689,43 +1762,18 @@ 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;
@@ -1848,9 +1896,9 @@ return 0;
 }
 
 
-void InsertOperatorToTable(const tt_char * name, typename MatOperator::Type type)
+void InsertOperatorToTable(const char * name, typename MatOperator::Type type)
 {
-	operators_table.insert( std::make_pair(tt_string(name), type) );
+	operators_table.insert( std::make_pair(std::string(name), type) );
 }
 
 
@@ -1859,19 +1907,19 @@ void InsertOperatorToTable(const tt_char * name, typename MatOperator::Type type
 */
 void CreateMathematicalOperatorsTable()
 {
-	InsertOperatorToTable(TTMATH_TEXT("||"), MatOperator::lor);
+	InsertOperatorToTable("||", MatOperator::lor);
-	InsertOperatorToTable(TTMATH_TEXT("&&"), MatOperator::land);
+	InsertOperatorToTable("&&", MatOperator::land);
-	InsertOperatorToTable(TTMATH_TEXT("!="), MatOperator::neq);
+	InsertOperatorToTable("!=", MatOperator::neq);
-	InsertOperatorToTable(TTMATH_TEXT("=="), MatOperator::eq);
+	InsertOperatorToTable("==", MatOperator::eq);
-	InsertOperatorToTable(TTMATH_TEXT(">="), MatOperator::get);
+	InsertOperatorToTable(">=", MatOperator::get);
-	InsertOperatorToTable(TTMATH_TEXT("<="), MatOperator::let);
+	InsertOperatorToTable("<=", MatOperator::let);
-	InsertOperatorToTable(TTMATH_TEXT(">"),  MatOperator::gt);
+	InsertOperatorToTable(">",  MatOperator::gt);
-	InsertOperatorToTable(TTMATH_TEXT("<"),  MatOperator::lt);
+	InsertOperatorToTable("<",  MatOperator::lt);
-	InsertOperatorToTable(TTMATH_TEXT("-"),  MatOperator::sub);
+	InsertOperatorToTable("-",  MatOperator::sub);
-	InsertOperatorToTable(TTMATH_TEXT("+"),  MatOperator::add);
+	InsertOperatorToTable("+",  MatOperator::add);
-	InsertOperatorToTable(TTMATH_TEXT("/"),  MatOperator::div);
+	InsertOperatorToTable("/",  MatOperator::div);
-	InsertOperatorToTable(TTMATH_TEXT("*"),  MatOperator::mul);
+	InsertOperatorToTable("*",  MatOperator::mul);
-	InsertOperatorToTable(TTMATH_TEXT("^"),  MatOperator::pow);
+	InsertOperatorToTable("^",  MatOperator::pow);
 }
 
 
@@ -1881,12 +1929,12 @@ void CreateMathematicalOperatorsTable()
 	e.g.
 	true when str1="test" and str2="te"
 */
-bool IsSubstring(const tt_string & str1, const tt_string & str2)
+bool IsSubstring(const std::string & str1, const std::string & str2)
 {
 	if( str2.length() > str1.length() )
 		return false;
 
-	for(tt_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;
 
@@ -1899,7 +1947,7 @@ return true;
 */
 void ReadMathematicalOperator(Item & result)
 {
-tt_string oper;
+std::string oper;
 typename OperatorsTable::iterator iter_old, iter_new;
 
 	iter_old = operators_table.end();
@@ -1928,6 +1976,31 @@ 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
@@ -1939,7 +2012,11 @@ typename OperatorsTable::iterator iter_old, iter_new;
 int ReadOperator(Item & result)
 {
 	SkipWhiteCharacters();
-	
+
+	if( *pstring == '%' )
+		OperatorPercentage();
+
+
 	if( *pstring == 0 )
 		return 1;
 	else
@@ -1949,7 +2026,7 @@ int ReadOperator(Item & result)
 		++pstring;
 	}
 	else
-	if( *pstring == ';' )
+	if( *pstring == ';' || (param_sep!=0 && *pstring==param_sep) )
 	{
 		result.type = Item::semicolon;
 		++pstring;
@@ -1979,7 +2056,9 @@ return 0;
 void MakeStandardMathematicOperation(ValueType & value1, typename MatOperator::Type mat_operator,
 									const ValueType & value2)
 {
-int res;
+uint res;
+
+	calculated = true;
 
 	switch( mat_operator )
 	{
@@ -2042,7 +2121,6 @@ int res;
 
 		break;
 
-
 	default:
 		/*
 			on the stack left an unknown operator but we had to recognize its before
@@ -2423,12 +2501,16 @@ public:
 Parser(): default_stack_size(100)
 {
 	pstop_calculating = 0;
-	puser_variables = 0;
+	puser_variables   = 0;
-	puser_functions = 0;
+	puser_functions   = 0;
 	pfunction_local_variables = 0;
-	base = 10;
+	base              = 10;
-	deg_rad_grad = 1;
+	deg_rad_grad      = 1;
-	error = err_ok;
+	error             = err_ok;
+	group             = 0;
+	comma             = '.';
+	comma2            = ',';
+	param_sep         = 0;
 
 	CreateFunctionsTable();
 	CreateVariablesTable();
@@ -2447,7 +2529,11 @@ Parser<ValueType> & operator=(const Parser<ValueType> & p)
 	pfunction_local_variables = 0;
 	base              = p.base;
 	deg_rad_grad      = p.deg_rad_grad;
-	error             = err_ok;
+	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.
@@ -2474,7 +2560,8 @@ 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)
 {
@@ -2529,16 +2616,50 @@ void SetFunctions(const Objects * pf)
 }
 
 
+/*!
+	setting the group character
+	default zero (not used)
+*/
+void SetGroup(int g)
+{
+	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;
+}
+
 
 /*!
 	the main method using for parsing string
 */
-ErrorCode Parse(const tt_char * str)
+ErrorCode Parse(const char * str)
 {
 	stack_index  = 0;
 	pstring      = str;
 	error        = err_ok;
-	
+	calculated   = false;
+
 	stack.resize( default_stack_size );
 
 	try
@@ -2548,6 +2669,7 @@ ErrorCode Parse(const tt_char * str)
 	catch(ErrorCode c)
 	{
 		error = c;
+		calculated = false;
 	}
 
 	NormalizeStack();
@@ -2556,9 +2678,52 @@ 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. 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

@@ -65,7 +65,7 @@
 #define TTMATH_MAJOR_VER		0
 #define TTMATH_MINOR_VER		9
 #define TTMATH_REVISION_VER		0
-#define TTMATH_PRERELEASE_VER	1
+#define TTMATH_PRERELEASE_VER	0
 
 
 /*!
@@ -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
 
 	/*!
@@ -157,13 +166,26 @@ 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;
+	#ifdef _MSC_VER
-	typedef signed   long sint;
+		/* 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
@@ -197,59 +219,30 @@ 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(UNICODE) || defined(_UNICODE)
+#if defined(TTMATH_MULTITHREADS) && !defined(TTMATH_MULTITHREADS_NOSYNC)
-#define TTMATH_USE_WCHAR
+	#if !defined(TTMATH_POSIX_THREADS) && !defined(TTMATH_WIN32_THREADS)
+
+		#if defined(_WIN32)
+			#define TTMATH_WIN32_THREADS
+		#elif defined(unix) || defined(__unix__) || defined(__unix)
+			#define TTMATH_POSIX_THREADS
+		#endif
+
+	#endif
 #endif
 
 
-#ifdef TTMATH_USE_WCHAR
-
-	typedef	wchar_t					tt_char;
-	typedef	std::wstring			tt_string;
-	typedef std::wostringstream		tt_ostringstream;
-	typedef std::wostream			tt_ostream;
-	typedef std::wistream			tt_istream;
-	#define TTMATH_TEXT_HELPER(txt)	L##txt
-
-#else
-
-	typedef	char					tt_char;
-	typedef	std::string				tt_string;
-	typedef std::ostringstream		tt_ostringstream;
-	typedef std::ostream			tt_ostream;
-	typedef std::istream			tt_istream;
-	#define TTMATH_TEXT_HELPER(txt)	txt
-
-#endif
-
-#define TTMATH_TEXT(txt) 	TTMATH_TEXT_HELPER(txt)
-
-
-
-
-
-/*!
-	characters which represent the comma operator
-
-	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 ','
-
-
 
 /*!
 	this variable defines how many iterations are performed
@@ -272,14 +265,18 @@ namespace ttmath
 	if the size of a vector is smaller than TTMATH_USE_KARATSUBA_MULTIPLICATION_FROM_SIZE
 	the Karatsuba algorithm will use standard schoolbook multiplication
 */
-#ifdef __GNUC__
+#ifdef TTMATH_DEBUG_LOG
-#define TTMATH_USE_KARATSUBA_MULTIPLICATION_FROM_SIZE 3
+	// if TTMATH_DEBUG_LOG is defined then we should use the same size regardless of the compiler
+	#define TTMATH_USE_KARATSUBA_MULTIPLICATION_FROM_SIZE 3
 #else
-#define TTMATH_USE_KARATSUBA_MULTIPLICATION_FROM_SIZE 5
+	#ifdef __GNUC__
+		#define TTMATH_USE_KARATSUBA_MULTIPLICATION_FROM_SIZE 3
+	#else
+		#define TTMATH_USE_KARATSUBA_MULTIPLICATION_FROM_SIZE 5
+	#endif
 #endif
 
 
-
 /*!
 	this is a special value used when calculating the Gamma(x) function
 	if x is greater than this value then the Gamma(x) will be calculated using
@@ -292,9 +289,30 @@ namespace ttmath
 
 
 
+
 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
 	*/
@@ -325,10 +343,132 @@ namespace ttmath
 		err_object_exists,
 		err_unknown_object,
 		err_still_calculating,
-		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)
@@ -355,20 +495,20 @@ namespace ttmath
 	*/
 	class ExceptionInfo
 	{
-	const tt_char * file;
+	const char * file;
 	int line;
 
 	public:
 		ExceptionInfo() : file(0), line(0) {}
-		ExceptionInfo(const tt_char * f, int l) : file(f), line(l) {}
+		ExceptionInfo(const char * f, int l) : file(f), line(l) {}
 
-		tt_string Where() const
+		std::string Where() const
 		{
 			if( !file )
-				return TTMATH_TEXT("unknown");
+				return "unknown";
 
-			tt_ostringstream result;
+			std::ostringstream result;
-			result << file << TTMATH_TEXT(":") << line;
+			result << file << ":" << line;
 
 		return result.str();
 		}
@@ -382,7 +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 (or std::wstring) 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)
 
@@ -410,12 +550,12 @@ namespace ttmath
 		{
 		}
 
-		ReferenceError(const tt_char * f, int l) :
+		ReferenceError(const char * f, int l) :
 							std::logic_error("reference error"), ExceptionInfo(f,l)
 		{
 		}
 
-		tt_string Where() const
+		std::string Where() const
 		{
 			return ExceptionInfo::Where();
 		}
@@ -430,7 +570,7 @@ namespace ttmath
 		of this type
 
 		if you compile with gcc you can get a small benefit 
-		from using method Where() (it returns std::string (or std::wstring) 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)
 	*/
@@ -442,12 +582,12 @@ namespace ttmath
 		{
 		}
 
-		RuntimeError(const tt_char * f, int l) :
+		RuntimeError(const char * f, int l) :
 						std::runtime_error("internal error"), ExceptionInfo(f,l)
 		{
 		}
 
-		tt_string Where() const
+		std::string Where() const
 		{
 			return ExceptionInfo::Where();
 		}
@@ -462,19 +602,11 @@ namespace ttmath
 
 		#if defined(__FILE__) && defined(__LINE__)
 
-			#ifdef TTMATH_USE_WCHAR
-				#define TTMATH_FILE_HELPER2(arg)  L##arg
-				#define TTMATH_FILE_HELPER(x)     TTMATH_FILE_HELPER2(x)
-				#define TTMATH_FILE               TTMATH_FILE_HELPER(__FILE__)
-			#else
-				#define TTMATH_FILE               __FILE__
-			#endif
-
 			#define TTMATH_REFERENCE_ASSERT(expression) \
-				if( &(expression) == this ) throw ttmath::ReferenceError(TTMATH_FILE, __LINE__);
+				if( &(expression) == this ) throw ttmath::ReferenceError(__FILE__, __LINE__);
 
 			#define TTMATH_ASSERT(expression) \
-				if( !(expression) ) throw ttmath::RuntimeError(TTMATH_FILE, __LINE__);
+				if( !(expression) ) throw ttmath::RuntimeError(__FILE__, __LINE__);
 
 		#else
 
@@ -493,24 +625,19 @@ namespace ttmath
 
 
 	#ifdef TTMATH_DEBUG_LOG
-
+		#define TTMATH_LOG(msg)                             PrintLog(msg, std::cout);
-		#ifdef TTMATH_USE_WCHAR
+		#define TTMATH_LOGC(msg, carry)                     PrintLog(msg, carry, std::cout);
-			#define TTMATH_LOG_HELPER(msg) \
+		#define TTMATH_VECTOR_LOG(msg, vector, len)         PrintVectorLog(msg, std::cout, vector, len);
-				PrintLog(L##msg, std::wcout);
+		#define TTMATH_VECTOR_LOGC(msg, carry, vector, len) PrintVectorLog(msg, carry, std::cout, vector, len);
-		#else
-			#define TTMATH_LOG_HELPER(msg) \
-				PrintLog(msg, std::cout);
-		#endif
-
-		#define TTMATH_LOG(msg) TTMATH_LOG_HELPER(msg)
-
 	#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
 

ttmath/ttmathuint_noasm.h

@@ -51,6 +51,56 @@
 
 namespace ttmath
 {
+
+	/*!
+		returning the string represents the currect type of the library
+		we have following types:
+		  asm_vc_32   - with asm code designed for Microsoft Visual C++ (32 bits)
+		  asm_gcc_32  - with asm code designed for GCC (32 bits)
+		  asm_vc_64   - with asm for VC (64 bit)
+		  asm_gcc_64  - with asm for GCC (64 bit)
+		  no_asm_32   - pure C++ version (32 bit) - without any asm code
+		  no_asm_64   - pure C++ version (64 bit) - without any asm code
+	*/
+	template<uint value_size>
+	const char * UInt<value_size>::LibTypeStr()
+	{
+		#ifdef TTMATH_PLATFORM32
+			static const char info[] = "no_asm_32";
+		#endif		
+
+		#ifdef TTMATH_PLATFORM64
+			static const char info[] = "no_asm_64";
+		#endif
+
+	return info;
+	}
+
+	
+	/*!
+		returning the currect type of the library
+	*/
+	template<uint value_size>
+	LibTypeCode UInt<value_size>::LibType()
+	{
+		#ifdef TTMATH_PLATFORM32
+			LibTypeCode info = no_asm_32;
+		#endif		
+
+		#ifdef TTMATH_PLATFORM64
+			LibTypeCode info = no_asm_64;
+		#endif
+
+	return info;
+	}
+
+
+	/*!
+		this method adds two words together
+		returns carry
+
+		this method is created only when TTMATH_NOASM macro is defined
+	*/
 	template<uint value_size>
 	uint UInt<value_size>::AddTwoWords(uint a, uint b, uint carry, uint * result)
 	{
@@ -95,7 +145,7 @@ namespace ttmath
 		for(i=0 ; i<value_size ; ++i)
 			c = AddTwoWords(table[i], ss2.table[i], c, &table[i]);
 
-		TTMATH_LOG("UInt::Add")
+		TTMATH_LOGC("UInt::Add", c)
 	
 	return c;
 	}
@@ -131,7 +181,7 @@ namespace ttmath
 		for(i=index+1 ; i<value_size && c ; ++i)
 			c = AddTwoWords(table[i], 0, c, &table[i]);
 
-		TTMATH_LOG("UInt::AddInt")
+		TTMATH_LOGC("UInt::AddInt", c)
 	
 	return c;
 	}
@@ -184,7 +234,7 @@ namespace ttmath
 		for(i=index+2 ; i<value_size && c ; ++i)
 			c = AddTwoWords(table[i], 0, c, &table[i]);
 
-		TTMATH_LOG("UInt::AddTwoInts")
+		TTMATH_LOGC("UInt::AddTwoInts", c)
 	
 	return c;
 	}
@@ -224,7 +274,7 @@ namespace ttmath
 		for( ; i<ss1_size ; ++i)
 			c = AddTwoWords(ss1[i], 0, c, &result[i]);
 
-		TTMATH_LOG("UInt::AddVector")
+		TTMATH_VECTOR_LOGC("UInt::AddVector", c, result, ss1_size)
 
 	return c;
 	}
@@ -232,6 +282,12 @@ namespace ttmath
 
 
 
+	/*!
+		this method subtractes one word from the other
+		returns carry
+
+		this method is created only when TTMATH_NOASM macro is defined
+	*/
 	template<uint value_size>
 	uint UInt<value_size>::SubTwoWords(uint a, uint b, uint carry, uint * result)
 	{
@@ -273,7 +329,7 @@ namespace ttmath
 		for(i=0 ; i<value_size ; ++i)
 			c = SubTwoWords(table[i], ss2.table[i], c, &table[i]);
 
-		TTMATH_LOG("UInt::Sub")
+		TTMATH_LOGC("UInt::Sub", c)
 
 	return c;
 	}
@@ -311,7 +367,7 @@ namespace ttmath
 		for(i=index+1 ; i<value_size && c ; ++i)
 			c = SubTwoWords(table[i], 0, c, &table[i]);
 
-		TTMATH_LOG("UInt::SubInt")
+		TTMATH_LOGC("UInt::SubInt", c)
 	
 	return c;
 	}
@@ -351,7 +407,7 @@ namespace ttmath
 		for( ; i<ss1_size ; ++i)
 			c = SubTwoWords(ss1[i], 0, c, &result[i]);
 
-		TTMATH_LOG("UInt::SubVector")
+		TTMATH_VECTOR_LOGC("UInt::SubVector", c, result, ss1_size)
 
 	return c;
 	}
@@ -385,7 +441,7 @@ namespace ttmath
 			c        = new_c;
 		}
 
-		TTMATH_LOG("UInt::Rcl2_one")
+		TTMATH_LOGC("UInt::Rcl2_one", c)
 
 	return c;
 	}
@@ -424,7 +480,7 @@ namespace ttmath
 			c        = new_c;
 		}
 
-		TTMATH_LOG("UInt::Rcr2_one")
+		TTMATH_LOGC("UInt::Rcr2_one", c)
 
 	return c;
 	}
@@ -462,7 +518,7 @@ namespace ttmath
 			c        = new_c;
 		}
 
-		TTMATH_LOG("UInt::Rcl2")
+		TTMATH_LOGC("UInt::Rcl2", c)
 
 	return (c & 1);
 	}
@@ -501,7 +557,7 @@ namespace ttmath
 			c        = new_c;
 		}
 
-		TTMATH_LOG("UInt::Rcr2")
+		TTMATH_LOGC("UInt::Rcr2", c)
 
 	return (c & TTMATH_UINT_HIGHEST_BIT) ? 1 : 0;
 	}
@@ -509,10 +565,9 @@ namespace ttmath
 
 
 
-	/*
+	/*!
 		this method returns the number of the highest set bit in x
 		if the 'x' is zero this method returns '-1'
-
 	*/
 	template<uint value_size>
 	sint UInt<value_size>::FindLeadingBitInWord(uint x)
@@ -539,7 +594,7 @@ namespace ttmath
 		this method sets a special bit in the 'value'
 		and returns the last state of the bit (zero or one)
 
-		bit is from <0,63>
+		bit is from <0,TTMATH_BITS_PER_UINT-1>
 
 		e.g.
 		 uint x = 100;
@@ -553,7 +608,7 @@ namespace ttmath
 
 		uint mask = 1;
 
-		if( bit > 1 )
+		if( bit > 0 )
 			mask = mask << bit;
 
 		uint last = value & mask;
@@ -741,11 +796,11 @@ namespace ttmath
 			temp1.u_.high = a_.u_.low;
 			temp1.u_.low  = b_.u_.high;
 
-			res_.u_.high  = temp1.u / c;
+			res_.u_.high  = (unsigned int)(temp1.u / c);
-			temp2.u_.high = temp1.u % c;
+			temp2.u_.high = (unsigned int)(temp1.u % c);
 			temp2.u_.low  = b_.u_.low;
 			
-			res_.u_.low  = temp2.u / c;
+			res_.u_.low  = (unsigned int)(temp2.u / c);
 			*rest        = temp2.u % c;
 
 			*r = res_.u;

ttmath/ttmathuint_x86.h

@@ -60,6 +60,50 @@
 namespace ttmath
 {
 
+	/*!
+		returning the string represents the currect type of the library
+		we have following types:
+		  asm_vc_32   - with asm code designed for Microsoft Visual C++ (32 bits)
+		  asm_gcc_32  - with asm code designed for GCC (32 bits)
+		  asm_vc_64   - with asm for VC (64 bit)
+		  asm_gcc_64  - with asm for GCC (64 bit)
+		  no_asm_32   - pure C++ version (32 bit) - without any asm code
+		  no_asm_64   - pure C++ version (64 bit) - without any asm code
+	*/
+	template<uint value_size>
+	const char * UInt<value_size>::LibTypeStr()
+	{
+		#ifndef __GNUC__
+			static const char info[] = "asm_vc_32";
+		#endif		
+
+		#ifdef __GNUC__
+			static const char info[] = "asm_gcc_32";
+		#endif
+
+	return info;
+	}
+
+
+	/*!
+		returning the currect type of the library
+	*/
+	template<uint value_size>
+	LibTypeCode UInt<value_size>::LibType()
+	{
+		#ifndef __GNUC__
+			LibTypeCode info = asm_vc_32;
+		#endif		
+
+		#ifdef __GNUC__
+			LibTypeCode info = asm_gcc_32;
+		#endif
+
+	return info;
+	}
+
+
+
 	/*!
 	*
 	*	basic mathematic functions
@@ -152,7 +196,7 @@ namespace ttmath
 				: "cc", "memory" );
 		#endif
 
-		TTMATH_LOG("UInt::Add")
+		TTMATH_LOGC("UInt::Add", c)
 
 	return c;
 	}
@@ -253,7 +297,7 @@ namespace ttmath
 
 		#endif
 	
-		TTMATH_LOG("UInt::AddInt")
+		TTMATH_LOGC("UInt::AddInt", c)
 
 	return c;
 	}
@@ -374,7 +418,7 @@ namespace ttmath
 
 		#endif
 
-		TTMATH_LOG("UInt::AddTwoInts")
+		TTMATH_LOGC("UInt::AddTwoInts", c)
 	
 	return c;
 	}
@@ -506,6 +550,8 @@ namespace ttmath
 
 		#endif
 
+		TTMATH_VECTOR_LOGC("UInt::AddVector", c, result, ss1_size)
+
 	return c;
 	}
 
@@ -592,7 +638,7 @@ namespace ttmath
 
 		#endif
 
-		TTMATH_LOG("UInt::Sub")
+		TTMATH_LOGC("UInt::Sub", c)
 
 	return c;
 	}
@@ -694,7 +740,7 @@ namespace ttmath
 
 		#endif
 		
-		TTMATH_LOG("UInt::SubInt")
+		TTMATH_LOGC("UInt::SubInt", c)
 	
 	return c;
 	}
@@ -832,6 +878,8 @@ namespace ttmath
 
 		#endif
 
+		TTMATH_VECTOR_LOGC("UInt::SubVector", c, result, ss1_size)
+
 	return c;
 	}
 
@@ -908,7 +956,7 @@ namespace ttmath
 
 		#endif
 
-		TTMATH_LOG("UInt::Rcl2_one")
+		TTMATH_LOGC("UInt::Rcl2_one", c)
 
 	return c;
 	}
@@ -980,7 +1028,7 @@ namespace ttmath
 
 		#endif
 
-		TTMATH_LOG("UInt::Rcr2_one")
+		TTMATH_LOGC("UInt::Rcr2_one", c)
 
 	return c;
 	}
@@ -1114,7 +1162,7 @@ namespace ttmath
 
 		#endif
 
-		TTMATH_LOG("UInt::Rcl2")
+		TTMATH_LOGC("UInt::Rcl2", c)
 
 	return c;
 	}
@@ -1248,7 +1296,7 @@ namespace ttmath
 
 		#endif
 
-		TTMATH_LOG("UInt::Rcr2")
+		TTMATH_LOGC("UInt::Rcr2", c)
 
 	return c;
 	}

ttmath/ttmathuint_x86_64.h

@@ -51,10 +51,77 @@
 	this file is included at the end of ttmathuint.h
 */
 
+#ifdef _MSC_VER
+#include <intrin.h>
+#endif
+
 
 namespace ttmath
 {
 
+	#ifdef _MSC_VER
+
+		extern "C"
+			{
+			uint __fastcall ttmath_adc_x64(uint* p1, const uint* p2, uint nSize, uint c);
+			uint __fastcall ttmath_addindexed_x64(uint* p1, uint nSize, uint nPos, uint nValue);
+			uint __fastcall ttmath_addindexed2_x64(uint* p1, uint nSize, uint nPos, uint nValue1, uint nValue2);
+			uint __fastcall ttmath_addvector_x64(const uint * ss1, const uint * ss2, uint ss1_size, uint ss2_size, uint * result);
+			uint __fastcall ttmath_sbb_x64(uint* p1, const uint* p2, uint nSize, uint c);
+			uint __fastcall ttmath_subindexed_x64(uint* p1, uint nSize, uint nPos, uint nValue);
+			uint __fastcall ttmath_subvector_x64(const uint * ss1, const uint * ss2, uint ss1_size, uint ss2_size, uint * result);
+			uint __fastcall ttmath_rcl_x64(uint* p1, uint nSize, uint nLowestBit);
+			uint __fastcall ttmath_rcr_x64(uint* p1, uint nSize, uint nLowestBit);
+			uint __fastcall ttmath_div_x64(uint* pnValHi, uint* pnValLo, uint nDiv);
+			uint __fastcall ttmath_rcl2_x64(uint* p1, uint nSize, uint nBits, uint c);
+			uint __fastcall ttmath_rcr2_x64(uint* p1, uint nSize, uint nBits, uint c);
+			};
+	#endif
+
+
+	/*!
+		returning the string represents the currect type of the library
+		we have following types:
+		  asm_vc_32   - with asm code designed for Microsoft Visual C++ (32 bits)
+		  asm_gcc_32  - with asm code designed for GCC (32 bits)
+		  asm_vc_64   - with asm for VC (64 bit)
+		  asm_gcc_64  - with asm for GCC (64 bit)
+		  no_asm_32   - pure C++ version (32 bit) - without any asm code
+		  no_asm_64   - pure C++ version (64 bit) - without any asm code
+	*/
+	template<uint value_size>
+	const char * UInt<value_size>::LibTypeStr()
+	{
+		#ifdef _MSC_VER
+			static const char info[] = "asm_vc_64";
+		#endif		
+
+		#ifdef __GNUC__
+			static const char info[] = "asm_gcc_64";
+		#endif
+
+	return info;
+	}
+
+
+	/*!
+		returning the currect type of the library
+	*/
+	template<uint value_size>
+	LibTypeCode UInt<value_size>::LibType()
+	{
+		#ifdef _MSC_VER
+			LibTypeCode info = asm_vc_64;
+		#endif		
+
+		#ifdef __GNUC__
+			LibTypeCode info = asm_gcc_64;
+		#endif
+
+	return info;
+	}
+
+
 	/*!
 	*
 	*	basic mathematic functions
@@ -82,8 +149,12 @@ namespace ttmath
 		// we don't have to use TTMATH_REFERENCE_ASSERT here
 		// this algorithm doesn't require it
 
-		#ifndef __GNUC__
+		#if !defined(__GNUC__) && !defined(_MSC_VER)
-			#error "another compiler than GCC is currently not supported in 64bit mode"
+			#error "another compiler than GCC or Microsoft VC is currently not supported in 64bit mode, you can compile with TTMATH_NOASM macro"
+		#endif
+
+		#ifdef _MSC_VER
+			c = ttmath_adc_x64(p1,p2,b,c);
 		#endif
 
 		#ifdef __GNUC__
@@ -113,7 +184,7 @@ namespace ttmath
 
 		#endif
 
-		TTMATH_LOG("UInt::Add")
+		TTMATH_LOGC("UInt::Add", c)
 	
 	return c;
 	}
@@ -149,10 +220,16 @@ namespace ttmath
 
 		TTMATH_ASSERT( index < value_size )
 
-		#ifndef __GNUC__
+		#if !defined(__GNUC__) && !defined(_MSC_VER)
-			#error "another compiler than GCC is currently not supported in 64bit mode"
+			#error "another compiler than GCC or Microsoft VC is currently not supported in 64bit mode, you can compile with TTMATH_NOASM macro"
 		#endif
 
+
+		#ifdef _MSC_VER
+			c = ttmath_addindexed_x64(p1,b,index,value);
+		#endif
+
+
 		#ifdef __GNUC__
 		uint dummy, dummy2;
 	
@@ -179,7 +256,7 @@ namespace ttmath
 
 		#endif
 
-		TTMATH_LOG("UInt::AddInt")
+		TTMATH_LOGC("UInt::AddInt", c)
 	
 	return c;
 	}
@@ -227,10 +304,16 @@ namespace ttmath
 
 		TTMATH_ASSERT( index < value_size - 1 )
 
-		#ifndef __GNUC__
+		#if !defined(__GNUC__) && !defined(_MSC_VER)
-			#error "another compiler than GCC is currently not supported in 64bit mode"
+			#error "another compiler than GCC or Microsoft VC is currently not supported in 64bit mode, you can compile with TTMATH_NOASM macro"
 		#endif
 
+
+		#ifdef _MSC_VER
+			c = ttmath_addindexed2_x64(p1,b,index,x1,x2);
+		#endif
+
+
 		#ifdef __GNUC__
 		uint dummy, dummy2;
 
@@ -261,7 +344,7 @@ namespace ttmath
 
 		#endif
 
-		TTMATH_LOG("UInt::AddTwoInts")
+		TTMATH_LOGC("UInt::AddTwoInts", c)
 
 	return c;
 	}
@@ -293,15 +376,21 @@ namespace ttmath
 	{
 		TTMATH_ASSERT( ss1_size >= ss2_size )
 
-		uint rest = ss1_size - ss2_size;
 		uint c;
 
-		#ifndef __GNUC__
+		#if !defined(__GNUC__) && !defined(_MSC_VER)
-			#error "another compiler than GCC is currently not supported in 64bit mode"
+			#error "another compiler than GCC or Microsoft VC is currently not supported in 64bit mode, you can compile with TTMATH_NOASM macro"
 		#endif
 
+
+		#ifdef _MSC_VER
+			 c = ttmath_addvector_x64(ss1, ss2, ss1_size, ss2_size, result);
+		#endif
+
+
 		#ifdef __GNUC__
 		uint dummy1, dummy2, dummy3;	
+		uint rest = ss1_size - ss2_size;
 			
 			//	this part should be compiled with gcc
 		
@@ -343,15 +432,13 @@ namespace ttmath
 
 		#endif
 
-		TTMATH_LOG("UInt::AddVector")
+		TTMATH_VECTOR_LOGC("UInt::AddVector", c, result, ss1_size)
 
 	return c;
 	}
 
 
 
-
-
 	/*!
 		this method's subtracting ss2 from the 'this' and subtracting
 		carry if it has been defined
@@ -373,10 +460,16 @@ namespace ttmath
 		// we don't have to use TTMATH_REFERENCE_ASSERT here
 		// this algorithm doesn't require it
 
-		#ifndef __GNUC__
+		#if !defined(__GNUC__) && !defined(_MSC_VER)
-			#error "another compiler than GCC is currently not supported in 64bit mode"
+			#error "another compiler than GCC or Microsoft VC is currently not supported in 64bit mode, you can compile with TTMATH_NOASM macro"
 		#endif
 
+
+		#ifdef _MSC_VER
+			c = ttmath_sbb_x64(p1,p2,b,c);
+		#endif
+
+
 		#ifdef __GNUC__
 		uint dummy, dummy2;
 
@@ -401,12 +494,13 @@ namespace ttmath
 
 		#endif
 
-		TTMATH_LOG("UInt::Sub")
+		TTMATH_LOGC("UInt::Sub", c)
 
 	return c;
 	}
 
 
+
 	/*!
 		this method subtracts one word (at a specific position)
 		and returns a carry (if it was)
@@ -432,15 +526,22 @@ namespace ttmath
 	uint b = value_size;
 	uint * p1 = table;
 	uint c;
-	uint dummy, dummy2;
 
 		TTMATH_ASSERT( index < value_size )
 
-		#ifndef __GNUC__
+		#if !defined(__GNUC__) && !defined(_MSC_VER)
-			#error "another compiler than GCC is currently not supported in 64bit mode"
+			#error "another compiler than GCC or Microsoft VC is currently not supported in 64bit mode, you can compile with TTMATH_NOASM macro"
 		#endif
 
+
+		#ifdef _MSC_VER
+			c = ttmath_subindexed_x64(p1,b,index,value);
+		#endif
+
+
 		#ifdef __GNUC__
+			uint dummy, dummy2;
+
 			__asm__ __volatile__(
 			
 				"subq %%rdx, %%rcx 				\n"
@@ -464,14 +565,12 @@ namespace ttmath
 
 		#endif
 
-		TTMATH_LOG("UInt::SubInt")
+		TTMATH_LOGC("UInt::SubInt", c)
 
 	return c;
 	}
 
 
-
-
 	/*!
 		this static method subtractes one vector from the other
 		'ss1' is larger in size or equal to 'ss2'
@@ -498,20 +597,25 @@ namespace ttmath
 	{
 		TTMATH_ASSERT( ss1_size >= ss2_size )
 
-		uint rest = ss1_size - ss2_size;
 		uint c;
 
-		#ifndef __GNUC__
+		#if !defined(__GNUC__) && !defined(_MSC_VER)
-			#error "another compiler than GCC is currently not supported in 64bit mode"
+			#error "another compiler than GCC or Microsoft VC is currently not supported in 64bit mode, you can compile with TTMATH_NOASM macro"
 		#endif
 
-		#ifdef __GNUC__
 
-		/*
+		#ifdef _MSC_VER
-			the asm code is nearly the same as in AddVector
+			c = ttmath_subvector_x64(ss1, ss2, ss1_size, ss2_size, result);
-			only two instructions 'adc' are changed to 'sbb'
+		#endif
-		*/
+
+
+		#ifdef __GNUC__
+		
+		//	the asm code is nearly the same as in AddVector
+		//	only two instructions 'adc' are changed to 'sbb'
+		
 		uint dummy1, dummy2, dummy3;
+		uint rest = ss1_size - ss2_size;
 
 			__asm__ __volatile__(
 				"mov %%rdx, %%r8					\n"
@@ -551,13 +655,12 @@ namespace ttmath
 
 		#endif
 
-		TTMATH_LOG("UInt::SubVector")
+		TTMATH_VECTOR_LOGC("UInt::SubVector", c, result, ss1_size)
 
 	return c;
 	}
 
 
-
 	/*!
 		this method moves all bits into the left hand side
 		return value <- this <- c
@@ -579,10 +682,16 @@ namespace ttmath
 	uint * p1 = table;
 	
 
-		#ifndef __GNUC__
+		#if !defined(__GNUC__) && !defined(_MSC_VER)
-			#error "another compiler than GCC is currently not supported in 64bit mode"
+			#error "another compiler than GCC or Microsoft VC is currently not supported in 64bit mode, you can compile with TTMATH_NOASM macro"
 		#endif
 
+
+		#ifdef _MSC_VER
+			c = ttmath_rcl_x64(p1,b,c);
+		#endif
+
+
 		#ifdef __GNUC__
 		uint dummy, dummy2;
 
@@ -606,7 +715,7 @@ namespace ttmath
 	
 		#endif
 
-		TTMATH_LOG("UInt::Rcl2_one")
+		TTMATH_LOGC("UInt::Rcl2_one", c)
 
 	return c;
 	}
@@ -633,10 +742,16 @@ namespace ttmath
 	uint * p1 = table;
 	
 
-		#ifndef __GNUC__
+		#if !defined(__GNUC__) && !defined(_MSC_VER)
-			#error "another compiler than GCC is currently not supported in 64bit mode"
+			#error "another compiler than GCC or Microsoft VC is currently not supported in 64bit mode, you can compile with TTMATH_NOASM macro"
 		#endif
 
+
+		#ifdef _MSC_VER
+			c = ttmath_rcr_x64(p1,b,c);
+		#endif
+
+
 		#ifdef __GNUC__
 		uint dummy;
 
@@ -658,7 +773,7 @@ namespace ttmath
 
 		#endif
 
-		TTMATH_LOG("UInt::Rcr2_one")
+		TTMATH_LOGC("UInt::Rcr2_one", c)
 
 	return c;
 	}
@@ -688,10 +803,16 @@ namespace ttmath
 	uint * p1 = table;
 
 
-		#ifndef __GNUC__
+		#if !defined(__GNUC__) && !defined(_MSC_VER)
-			#error "another compiler than GCC is currently not supported in 64bit mode"
+			#error "another compiler than GCC or Microsoft VC is currently not supported in 64bit mode, you can compile with TTMATH_NOASM macro"
 		#endif
 
+
+		#ifdef _MSC_VER
+			c = ttmath_rcl2_x64(p1,b,bits,c);
+		#endif
+
+
 		#ifdef __GNUC__
 		uint dummy, dummy2, dummy3;
 
@@ -731,7 +852,7 @@ namespace ttmath
 
 		#endif
 
-		TTMATH_LOG("UInt::Rcl2")
+		TTMATH_LOGC("UInt::Rcl2", c)
 
 	return c;
 	}
@@ -758,14 +879,20 @@ namespace ttmath
 
 	sint b = value_size;
 	uint * p1 = table;
-	uint dummy, dummy2, dummy3;
 
-		#ifndef __GNUC__
+		#if !defined(__GNUC__) && !defined(_MSC_VER)
-			#error "another compiler than GCC is currently not supported in 64bit mode"
+			#error "another compiler than GCC or Microsoft VC is currently not supported in 64bit mode, you can compile with TTMATH_NOASM macro"
+		#endif
+
+
+		#ifdef _MSC_VER
+			c = ttmath_rcr2_x64(p1,b,bits,c);
 		#endif
 
 
 		#ifdef __GNUC__
+			uint dummy, dummy2, dummy3;
+
 			__asm__  __volatile__(
 
 			"movq %%rcx, %%rsi				\n"
@@ -805,7 +932,7 @@ namespace ttmath
 
 		#endif
 
-		TTMATH_LOG("UInt::Rcr2")
+		TTMATH_LOGC("UInt::Rcr2", c)
 
 	return c;
 	}
@@ -823,10 +950,23 @@ namespace ttmath
 	sint result;
 
 
-		#ifndef __GNUC__
+		#if !defined(__GNUC__) && !defined(_MSC_VER)
-			#error "another compiler than GCC is currently not supported in 64bit mode"
+			#error "another compiler than GCC or Microsoft VC is currently not supported in 64bit mode, you can compile with TTMATH_NOASM macro"
 		#endif
 
+		
+		#ifdef _MSC_VER
+
+			unsigned long nIndex = 0;
+
+			if( _BitScanReverse64(&nIndex,x) == 0 )
+				result = -1;
+			else
+				result = nIndex;
+
+		#endif
+
+
 		#ifdef __GNUC__
 		uint dummy;
 
@@ -868,11 +1008,16 @@ namespace ttmath
 		uint old_bit;
 		uint v = value;
 
-
+		#if !defined(__GNUC__) && !defined(_MSC_VER)
-		#ifndef __GNUC__
+			#error "another compiler than GCC or Microsoft VC is currently not supported in 64bit mode, you can compile with TTMATH_NOASM macro"
-			#error "another compiler than GCC is currently not supported in 64bit mode"
 		#endif
 
+
+		#ifdef _MSC_VER
+			old_bit = _bittestandset64((__int64*)&value,bit) != 0;
+		#endif
+
+
 		#ifdef __GNUC__
 
 			__asm__ (
@@ -924,10 +1069,16 @@ namespace ttmath
 	uint result1_;
 	uint result2_;
 
-		#ifndef __GNUC__
+		#if !defined(__GNUC__) && !defined(_MSC_VER)
-			#error "another compiler than GCC is currently not supported in 64bit mode"
+			#error "another compiler than GCC or Microsoft VC is currently not supported in 64bit mode, you can compile with TTMATH_NOASM macro"
 		#endif
 
+
+		#ifdef _MSC_VER
+			result1_ = _umul128(a,b,&result2_);
+		#endif
+
+
 		#ifdef __GNUC__
 
 		__asm__ (
@@ -981,10 +1132,20 @@ namespace ttmath
 
 		TTMATH_ASSERT( c != 0 )
 
-		#ifndef __GNUC__
+		#if !defined(__GNUC__) && !defined(_MSC_VER)
-			#error "another compiler than GCC is currently not supported in 64bit mode"
+			#error "another compiler than GCC or Microsoft VC is currently not supported in 64bit mode, you can compile with TTMATH_NOASM macro"
 		#endif
 
+
+		#ifdef _MSC_VER
+
+			ttmath_div_x64(&a,&b,c);
+			r_    = a;
+			rest_ = b;
+			
+		#endif
+
+
 		#ifdef __GNUC__
 		
 			__asm__ (
@@ -1002,7 +1163,6 @@ namespace ttmath
 		*rest = rest_;
 	}
 
-
 } //namespace
 
 

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