410 lines
24 KiB
Plaintext
410 lines
24 KiB
Plaintext
Version 0.9.0 prerelease (2009.10.25):


* added: support for wide characters (wchar_t, std::wstring)


* added: Big::IsInteger()


returns true if the value is integer (without fraction)


(NaN flag is not checked)


* added: global Gamma() function


* added: gamma() function to the parser


* added: CGamma<ValueType> class


is used with Gamma() and Factorial() in multithreaded environment


* added: multithread support for Big<> class


you should compile with TTMATH_MULTITHREADS


and use TTMATH_MULTITHREADS_HELPER macro somewhere in your *.cpp file


* added: x86_64 asm code for Microsoft Visual compiler


file: ttmathuint_x86_64_msvc.asm


(this file should be compiled first because MS VC doesn't support inline assembler in x86_64 mode)


* added: flag TTMATH_BIG_ZERO to Big<> class


if this flag is set then there is a value zero


Big::IsZero() is faster now


* added: Big::ClearInfoBit(unsigned char)


Big::SetInfoBit(unsigned char)


Big::IsInfoBit(unsigned char)


some methods for manipulating the info flags


* added: macro: TTMATH_BITS(min_bits)


which returns the number of machine words


capable to hold min_bits bits


* added: bool Parser::Calculated()


this method returns true is something was calculated


(at least one mathematical operator was used or a function or variable)


* added: to the parser: operator percentage


e.g. 100050%=1000(1000*0,5)=500


* 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


* 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






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 noasm parts  when macro TTMATH_NOASM was defined


* fixed: UInt<value_size>::MulInt(uint ss2)


there was a buffer overflow when value_size was equal 1


* fixed: UInt::AddVector() and UInt::SubVector() didn't want to compile


when macro TTMATH_NOASM was defined


* fixed: Big::operator>> didn't correctly recognize values in scientific mode (with 'e' character)


* fixed: Int::FromString(const tt_string & s, uint b = 10)


didn't use 'b' (always was '10')


* fixed: buffer overflow in Big::ToInt(Int<int_size> & result)


* fixed: powering algorithm in:


UInt::Pow(UInt<value_size> pow)


Big::Pow(UInt<pow_size> pow)


Big::PowUInt(Big<exp, man> pow)


when 'pow' was sufficient large the algorithm returned carry


but the result could have been calculated correctly






Version 0.8.5 (2009.06.16):


* fixed: Big::Mod(x) didn't correctly return a carry


and the result was sometimes very big (even greater than x)


* fixed: global function Mod(x) didn't set an ErrorCode object


* fixed: global function Round() didn't test a carry


now it sets ErrorCode object


* changed: function Sin(x) to Sin(x, ErrorCode * err=0)


when x was very big the function returns zero


now it sets ErrorCode object to err_overflow


and the result has a NaN flag set


the same is to Cos() function


* changed: PrepareSin(x) is using Big::Mod() now when reducing 2PI period


should be a little accurate especially on a very big 'x'


* changed: uint Mul(const UInt<value_size> & ss2, uint algorithm = 100)


void MulBig(const UInt<value_size> & ss2, UInt<value_size*2> & result, uint algorithm = 100)


those methods by default use MulFastest() and MulFastestBig()


* changed: changed a little Mul2Big() to cooperate with Mul3Big()


* added: uint UInt::Mul3(const UInt<value_size> & ss2)


void UInt::Mul3Big(const UInt<value_size> & ss2, UInt<value_size*2> & result)


a new multiplication algorithm: Karatsuba multiplication,


on a vector UInt<100> with all items different from zero this algorithm is faster


about 3 times than Mul2Big(), and on a vector UInt<1000> with all items different from


zero this algorithm is faster more than 5 times than Mul2Big()


(measured on 32bit platform with GCC 4.3.3 with O3 and DTTMATH_RELEASE)


* added: uint MulFastest(const UInt<value_size> & ss2)


void MulFastestBig(const UInt<value_size> & ss2, UInt<value_size*2> & result)


those methods are trying to select the fastest multiplication algorithm


* added: uint AddVector(const uint * ss1, const uint * ss2, uint ss1_size, uint ss2_size, uint * result)


uint SubVector(const uint * ss1, const uint * ss2, uint ss1_size, uint ss2_size, uint * result)


three forms: asm x86, asm x86_64, noasm


those methods are used by the Karatsuba multiplication algorithm


* added: to Big<> class: support for NaN flag (Not a Number)


bool Big::IsNan()  returns true if the NaN flag is set


void Big::SetNan()  sets the NaN flag


The NaN flag is set by default after creating an object:


Big<1, 2> a; // NaN is set (it means the object has not a valid number)


std::cout << a; // cout gives "NaN"


a = 123; // now NaN is not set


std::cout << a; // cout gives "123"


The NaN is set if there was a carry during calculations


a.Mul(very_big_value); // a will have a NaN set


The NaN is set if an argument is NaN too


b.SetNan();


a.Add(b); // a will have NaN because b has NaN too


If you try to do something on a NaN object, the result is a NaN too


a.SetNan();


a.Add(2); // a is still a NaN


The NaN is set if you use incorrect arguments


a.Ln(10); // a will have the NaN flag


The only way to clear the NaN flag is to assign a correct value or other correct object,


supposing 'a' has NaN flag, to remove the flag you can either:


a = 10;


a.FromInt(30);


a.SetOne();


a.FromBig(other_object_without_nan);


etc.






Version 0.8.4 (2009.05.08):


* fixed: UInt::DivInt() didn't check whether the divisor is zero


there was a hardware interruption when the divisor was zero


(now the method returns one)


* fixed: the problem with GCC optimization on x86_64


sometimes when using O2 or O3 GCC doesn't set correctly


the stack pointer (actually the stack is used for other things)


and you can't use instructions like push/pop in assembler code.


All the asm code in x86_64 have been rewritten, now instructions


push/pop are not used, other thing which have access to stack


(like "m" (mask) constraints in Rcl2 and Rcr2) have also gone away,


now the library works well with O2 and O3 and the asm code


is a little faster


* added: UInt::PrintLog(const char * msg, std::ostream & output)


used (for debugging purposes) by macro TTMATH_LOG(msg)


(it is used in nearly all methods in UInt class)


* added: macro TTMATH_DEBUG_LOG: when defined then TTMATH_LOG()


put some debug information (to std::cout)


* added: ttmathuint_x86.h, ttmathuint_x86_64.h, ttmathuint_noasm.h,


all the methods which are using assembler code have been


rewritten to noasm forms, now we have:


1. asm for x86 file: ttmathuint_x86.h


2. asm for x86_64 file: ttmathuint_x86_64.h


3. no asm file: ttmathuint_noasm.h


(it's used when macro TTMATH_NOASM is defined)


The third form can be used on x86 and x86_64 as well and


on other platforms with a little effort.






Version 0.8.3 (2009.04.06):


* fixed: RclMoveAllWords() and RcrMoveAllWords() sometimes didn't return


the proper carry, (when 'bits' was greater than or equal to 'value_size')


this had impact on Rcl() and Rcr(), they also returned the wrong carry


* fixed: UInt::Div() didn't return a correct result when the divisor was equal 1


there was an error in UInt::DivInt()  when the divisor was 1 it returned


zero and the carry was set


* fixed: there was a TTMATH_REREFENCE_ASSERT error in Big::PowUInt() caused by: start.Mul(start)


* fixed: Big::Add incorrectly rounded 'this' when both exponents were equal


it caused that sometimes when adding a zero the result has changed


this had impact among other things on FromString() method


"0,8" had different binary representation from "0,80"


* fixed: template Big::FromBig(const Big<another_exp, another_man> & another)


didn't correctly set the exponent (when the mantisses had different size 


when 'man' was different from 'another_man')


this had impact on operator= too


sample:


Big<2,3> a = 100;


Big<3,5> b;


b = a; // b had a wrong value


* fixed: Big::Pow(const Big<exp, man> & pow)


it's using PowInt() only when pow.exponent is in range (man*TTMATH_BITS_PER_UINT; 0]


previously the powering 'hung' on an input like this: "(1+ 1e10000) ^ 10e100000000"


(there was 10e100000000 iterations in PowInt())


* fixed: in function DegToRad(const ValueType & x, ErrorCode * err = 0) it is better


to make division first and then mutliplication  the result is more


accurate especially when x is: 90,180,270 or 360


* fixed: the parser didn't correctly treat operators for changing the base


(radix)  operators '#' and '&', e.g.:


'#sin(1)' was equal '0'  there was a zero from '#' and then


it was multipied by 'sin(1)'


the parser didn't check whether Big::FromString() has actually


read a proper value  the method Big::FromString() didn't have


something to report such a situation


* fixed: Big::FromString() when the base is 10, the method reads the scientific


part only if such a part it correctly supplied, e.g:


'1234e10', '1234e+10', '1234e5'


previous '1234e' was treated as: '1234e0' (now parsing stops on 'e' and


the 'e' can be parsed by other parsers, e.g. the mathematical


parser  now in the parser would be: '1234e' = '1234 * e' = '3354,3597...' )


* changed: renamed: Big::PowUInt(UInt<pow_size> pow) > Big::Pow(UInt<pow_size> pow)


it returns 2 when there is: 0^0


* changed: renamed: Big::PowInt(Int<pow_size> pow) > Big::Pow(Int<pow_size> pow)


it returns 2 when there is: 0^0 or 0^(something)


* changed: renamed: Big::PowBUInt() > PowUInt(), Big::PowBInt() > Big::PowInt()


they return 2 when the arguments are incorrect (like above)


* changed: UInt::SetBitInWord(uint & value, uint bit) is taking the first argument by a reference now,


the specific bit is set in the 'value' and the method returns the last state of the bit (zero or one)


* changed: UInt::SetBit(uint bit_index)  it's using TTMATH_ASSERT now


* changed: the size of builtin variables (constants) in ttmathbig.h


now they consist of 256 32bit words


macro TTMATH_BUILTIN_VARIABLES_SIZE is equal: 256u on a 32bit platform and 128ul on a 64bit platform


* changed: the asm code in ttmathuint.h and ttmathuint64.h has been completely rewritten


now UInt<> is faster about 1530% than UInt<> from 0.8.2


this has impact on Big<> too  it's faster about 10% now


* changed: in the parser: the form with operators '#' and '&' is as follows:


[+][#&]numeric_value


previous was: [+][#&][+]numeric_value


* changed: in the parser: the short form of multiplication has the same


priority as the normal multiplication, e.g.:


'2x^3' = 2 * (x^3)


previous the priority was greater than powering priority


previous: '2x^3' = (2*x) ^ 3


* added: UInt::GetBit(uint bit_index)  returning the state of the specific bit


* added: Big::operator=(double) and Big::Big(double)


* added: UInt::Pow(UInt<value_size> pow) and Int::Pow(Int<value_size> pow)


* added: global template functions in ttmath.h:


ValueType GradToRad(const ValueType & x, ErrorCode * err = 0)


ValueType RadToGrad(const ValueType & x, ErrorCode * err = 0)


ValueType DegToGrad(const ValueType & x, ErrorCode * err = 0)


ValueType DegToGrad(const ValueType & d, const ValueType & m,


const ValueType & s, ErrorCode * err = 0)


ValueType GradToDeg(const ValueType & x, ErrorCode * err = 0)


* added: Parser::SetDegRadGrad(int angle)  0 deg, 1 rad (default), 2 grad


this affects following functions (in the parser only): sin, cos, tan, cot,


asin, acos, atan, acot


* added: functions to the parser: gradtorad(grad), radtograd(rad), degtograd(deg),


degtograd(d,m,s), gradtodeg(grad)


* added: UInt::FromString, added a parametr 'after_source'


which is pointing at the end of the parsed string


* added: Int::FromString(): parameter 'const char ** after_source = 0'


if exists it's pointing at the end of the parsed string


* added: to UInt::FromString(), Int::FromString(), Big::FromString():


parameter 'bool * value_read = 0'  (if exists) tells


whether something has actually been read (at least one digit)


* added: Objects::IsDefined(const std::string & name)


returning true if such an object is defined


* removed: Big::FromString() this method doesn't longer recognize operators


for changing the base ('#' and '&')






Version 0.8.2 (2008.06.18):


* added: UInt::BitNot2() this method has been proposed by


Arek <kmicicc AnTispam users.sourceforge.net>


* changed: Int::FromInt(const Int<argument_size> & p),


Int::FromInt(sint value) (it returns zero now)


Int::operator=(uint i)


Int::Int(uint i)


* added: Int::FromUInt(const UInt<argument_size> & p),


Int::FromUInt(uint value)


and appropriate constructors and assignment


operators as well


* changed: Big::FromInt(Int<int_size> value),


* added: Big::FromUInt(UInt<int_size> value),


Big::operator=(const UInt<int_size> & value)


Big::Big(const UInt<int_size> & value)


* changed: the parser is allowed to recognize values which


begin with a dot, e.g '.5' is treated as '0.5'


* added: a method Big::FromDouble(double) which converts from


standard double into a Big


* added: uint Big::ToDouble(double&)  converting into double


* added: Big::FromBig() and an operator= and a contructor


for converting from another kind of a Big class


* added: to the parser: avg(), sum()


* added: 'decimal_point' parameter into Big::ToString(...)


* fixed: Big::operator>> didn't use TTMATH_COMMA_CHARACTER_2 macro


* added: a short form of multiplication (without the '*' character)


e.g. '5y', (it's used only if the second parameter


is a variable or function)


* changed: variables and functions are casesensitive now


* added: variables and functions can have underline characters


in their names


* changed: 'max_digit_after_comma' in Big::ToString()


remove the 2 state


* added: 'remove_trailing_zeroes' in Big::ToString()


it's either true or false


* fixed/changed: the way of using Big::SetSign()


the method do not check whether there is a zero or not now


(even if there's a zero the method can set a sign bit)


I changed this due to some prior errors


(errors corrected in revision 17, 49 and 58)






Version 0.8.1 (2007.04.17):


* fixed: Big::PowFrac(..) didn't return a correct error code


(when 'this' was negative)


* added: Root(x; index) (and to the parser as well)


* added: macro: TTMATH_PRERELEASE_VER (can be either zero or one)


* added: UInt::MulInt(int, UInt<int another_size>::&)


* added: Big::MulUInt(uint)


* changed: Big::MulInt(sint)


* added: Big::ToUInt(uint &)


* changed: Big::ToInt(sint&)


* changed: Factorial() it uses Big::MulUInt() at the beginning


(faster now especially more on a 32bit platform)


* added: doxygen.cfg for generating a documentation from the doxygen


* changed: UInt::Rcl(uint c=0) and UInt::Rcr(uint c=0) into


UInt::Rcl2(uint bits, uint c) and UInt::Rcr2(uint bits, uint c)


now they can move more than one bit and they are only private


* fixed: UInt::Rcl(uint bits, uint c) and UInt::Rcr(uint bits, uint c)


didn't correctly return a carry if the 'bits' were equal


to 'value_size*TTMATH_BITS_PER_UINT'


* changed: UInt::Rcl(uint bits, uint c) and UInt::Rcr(uint bits, uint c)


into UInt::Rcl(uint bits, uint c=0) and


UInt::Rcr(uint bits, uint c=0)


they are faster now when the bits is greater than a half of


the TTMATH_BITS_PER_UINT


* changed: UInt::CompensationToLeft() it's faster now


* changed: more small changes where there were UInt::Rcl(uint c=0) and


UInt::Rcr(uint c=0) used


* changed: as the Big type uses UInt::Rcl() and UInt::Rcr() a lot then


it is much faster now (about 525%)


* added: ASinh(), ACosh(), ATanh() /ATgh()/, ACoth() /ACtgh()/


and to the parser as well


* added: UInt::BitAnd(), UInt::BitOr(), UInt::BitXor(), UInt::BitNot(),


Big::BitAnd(), Big::BitOr(), Big::BitXor()


* added: to the parser: bitand(), bitor(), bitxor()


/band(), bor(), bxor()/


* changed: the way of parsing operators in the mathematical parser


(the parser is not too much greedy now)






Version 0.8.0 (2007.03.28):


* added: into the parser: SetFactorialMax()


* added: DegToDeg(deg, min, sec), DegToRad(deg), DegToRad(deg, min, sec),


RadToDeg(rad), Ceil(x), Floor(x), Sqrt(x), Sinh(x), Cosh(x),


Tanh(x) /Tgh(x)/, Coth(x) /Ctgh(x)/


* changed: class Objects in ttmathobjects.h has been completely rewritten,


we can change the names of userdefined variables or functions,


and the names are casesensitive now


* added: class History which is used in functions which take a lot of time


during calculating e.g. Factorial(x)


* added: Tg(x) a wrapper for Tan(x)


* changed: CTan(x) is Cot(x) now


* added: Ctg(x) a wrapper for Cot(x)


* added: ATg(x) a wrapper for ATan(x)


* changed: ACTan(x) is ACot(x) now


* added: ACtg(x) a wrapper for ACot(x)


* added: UInt::PrintTable() (for debugging etc.)


* changed: the methods Big::SetPi() Big::SetE() and Big::SetLn2() have


been rewritten, now they have 128 32bit words (it's about


1232 valid decimal digits)


* fixed: previous values from Big::SetPi() Big::SetE() and


Big::SetLn2() were not too much accurate (last 23 words were wrong)


* added: Big::SetLn10() (128 32bit words as well)


* added: macro TTMATH_BUILTIN_VARIABLES_SIZE which is equal 128u on


32bit platforms and 64ul on 64bit platforms (128/2=64)


* added: macros TTMATH_PLATFORM32 and TTMATH_PLATFORM64


* changed: a small optimisation in UInt::Mul2Big()


* added: at the end of ttmath.h: #include "ttmathparser.h"


this is for convenience for a programmer, he can only use #include


with ttmath.h even if he uses the parser


* added: to samples: big.cpp, parser.cpp


* added/changes/fixed: in copyconstructors and operators= in Int,


Uint and Big (more info in the commit log)


* renamed: Big::SetDotOne() into Big::Set05()


* changes: a few small optimisations in Big


* deleted: the word 'virtual' from destructors: UInt, Int, Big


(types in this library are not projected to be baseclasses for


another ones derived from them)


* and more small changes (look at the commit log)






Version 0.7.2 (2007.03.09):


* added: Big::Mod  the remainder from a division


* added: Big::Sgn  the 'sign' from the value (1,0,1)


* added: global functions Mod and Sgn too


* added: checking whether a user gives a correct value of a variable or function


(userdefined variables/functions in the mathematical parser)


* added: into the parser: logical operators: > < >= <= == != && 


* added: into the parser: logical functions: and() or() not() if()


* added: ErrorCode::err_unknown_operator when the parser couldn't read an operator






Version 0.7.1 (2007.02.27):


* fixed: the error 'overflow during printing' which was caused


by Big::FromInt(Int<int_size> value) (the sign has to be set at the end)


* fixed: many small errors


* added: ATan (arctan), ACTan (arc ctan) functions






Version 0.7.0 (2007.02.24):


* finished: support for 64bit platforms


* added: ASin (arcsin), ACos (arccos) functions






Version 0.6.4 (2007.01.29):


* fixed: the problem with a sign in the mathematical parser /(1) was 1/


* added: UInt::AddInt and UInt::SubInt


* changed: UInt::AddOne and UInt::SubOne (much faster now)


* added: UInt::SetBitInWord


* changed: UInt::SetBit (much faster now)


UInt::AddTwoUints renamed to UInt::AddTwoInts


UInt::FindLeadingBit32 renamed to UInt::FindLeadingBitInWord


UInt::Mul64 renamed to UInt::MulTwoWords


UInt::Div64 renamed to UInt::DivTwoWords


* added: UInt::SetBitInWord


* and more small changes in UInt type


* start adding support for Amd64 (not finished yet) (added ttmathuint64.h)






Version 0.6.3 (2007.01.22):


* changed: position of arguments (x and base) in logarithm functions are swapped


* changed: it's possible to use any multiplication algorithms in the same time


(macros UINT_MUL_VERSION_'X' have gone)


* added: ExceptionInfo, ReferenceError and RuntimeError classes


* changed: the mess in macros has been cleaned up


* added: TTMATH_RELEASE macro






Version 0.6.2 (2007.01.10):


* added: New division algorithm (radix b) where b is 2^32
