- GiNaC-list - ginac.de

Manipulation with matrice
by Fmant Dolch 08 Mar '15

08 Mar '15

Hi. In tutorial said there is lst_to_matrix method to create matrix from list of lists. If i want to change some rows in matrice, do i need firsst change list of lists then convert to matrix or there is some direct way to do it?

2 1

list differentiation
by soppon 01 Mar '15

01 Mar '15

Hi. Here two programs. dif1 return 0 for list1 and working well for list2, dif2 return the rigth answer, but if you change table argument x to y on line 12, then get segmentation fault. What should i do?

2 1

list input
by soppon 01 Mar '15

01 Mar '15

Hi. I'm trying to use list as input. List as output prints {expr1,expr2,..}, list.nops() is varying. When I write list as: parser reader; somelist=reader(cin); where input to cin: {expr1,expr2,..} at the end shift-D It returns somelist={expr1,expr2,..}, somelist.nops()=1. When accessing the 2nd and higher element of list, program gets segmentation fault.

3 7

how to match symmetric function
by Zhao Li 14 Feb '15

14 Feb '15

We have a self defined function “LMT” by DECLARE_FUNCTION_2P( LMT ); REGISTER_FUNCTION( LMT, set_symmetry(sy_symm(0,1)) ); Its two arguments are symmetric to each other. And we want to make the substitution in the expressions, LMT(mu1,mu2)*LMT(mu2,mu3) -> LMT(mu1,mu3). Therefore we made an exmap by exmap replacement; replacement.insert(ex2( LMT(wild(1),wild(2))*LMT(wild(2),wild(3)), LMT(wild(1),wild(3)) )); However, it sometimes does not work for ex testexpr = LMT(rho1,rho2)*LMT(rho2,rho3); cout << “before subs, testexpr = “ << testexpr << endl; testexpr = testexpr.subs( replacement, subs_options::algebraic ); cout << “after subs, testexpr = “ << testexpr << endl; The subs here mostly fails. Then we realised GiNaC will change to the canonical form for LMT, so we insert all cases of combinations by exmap replacement; replacement.insert(ex2( LMT(wild(1),wild(2))*LMT(wild(2),wild(3)), LMT(wild(1),wild(3)) )); replacement.insert(ex2( LMT(wild(2),wild(1))*LMT(wild(2),wild(3)), LMT(wild(1),wild(3)) )); replacement.insert(ex2( LMT(wild(1),wild(2))*LMT(wild(3),wild(2)), LMT(wild(1),wild(3)) )); replacement.insert(ex2( LMT(wild(2),wild(1))*LMT(wild(3),wild(2)), LMT(wild(1),wild(3)) )); But this still does not work sometimes. Then we got that GiNaC may also change the form of matching expression of LMT(wild(1),wild(2)). What we can have now is only turn off symmetry property for LMT function, but we worry it may generate difficulty in the future. So is there some option or algorithm for the symmetric function matching? Thanks. Best Regards, Zhao

1 0

Performance of simplify_indexed on long chains of Dirac matrices
by Vladyslav Shtabovenko 26 Jan '15

26 Jan '15

Dear all, I'm wondering about the performance of simplify_indexed applied to a long chain of Dirac matrices in D dimensions. For example, simplifying g^i1 g^i2 g^i3 g^i4 g^i5 g^i6 g^i7 g^i8 g^i9 g^i1 g^i2 g^i3 g^i4 g^i5 g^i6 g^i7 g^i8 g^i9 requires about 15 seconds on my machine. Is this something one would normally expect with GiNaC, or am I lacking some essential optimizations? I'm using Fedora 20 64-bit with newest GiNaC from the Git repository. The machine is an i5-3570 with 16GB RAM. The source code --------------------------------------------------------------------- #include <iostream> #include <ginac/ginac.h> using namespace std; using namespace GiNaC; int main() { symbol D("D"); varidx i1(symbol("i1"), D), i2(symbol("i2"), D), i3(symbol("i3"), D), i4(symbol("i4"), D), i5(symbol("i5"), D), i6(symbol("i6"), D), i7(symbol("i7"), D), i8(symbol("i8"), D), i9(symbol("i9"), D); cout << (simplify_indexed(dirac_gamma(i1)* dirac_gamma(i2)* dirac_gamma(i3)* dirac_gamma(i4)* dirac_gamma(i5)* dirac_gamma(i6)* dirac_gamma(i7)* dirac_gamma(i8)* dirac_gamma(i9)* dirac_gamma(i1.toggle_variance())* dirac_gamma(i2.toggle_variance())* dirac_gamma(i3.toggle_variance())* dirac_gamma(i4.toggle_variance())* dirac_gamma(i5.toggle_variance())* dirac_gamma(i6.toggle_variance())* dirac_gamma(i7.toggle_variance())* dirac_gamma(i8.toggle_variance())* dirac_gamma(i9.toggle_variance()) )).expand() << endl; return 0; } --------------------------------------------------------------------- is compiled and run via c++ test.cpp -o test -Ofast -lcln -lginac -Wl,-rpath,/usr/local/lib time ./test Cheers, Vladyslav

2 6

Release 1.6.3
by Richard B. Kreckel 28 Nov '14

28 Nov '14

Hi everybody, Alexei and I have prepared a new release and are happy to announce that GiNaC 1.6.3 is out and available. These are the changes: * New expand_options::expand_transcendental. GiNaC will apply additional rules when expanding expressions if this option is specified, i.e. log(a*b) -> log(a) + log(b) exp(a + b) -> exp(a)*exp(b) * Functions can define a custom expand() and info() now. * Automatically rewrite (x^{-1})^c -> x^{-c} for positive c. * Transform abs(x)^n -> x^n if x is real and n is even. * Improved power::info(): power with a real base and even exponent reports that it's non-negative. * Fixed spurious match failure due to expairseq::match() bug: > match(sin(y)*exp(b)+sin(x)*exp(a), sin($0)*exp(a)+exp(b)*sin($1)) FAIL * Fixed power::is_polynomial() and mul::is_polynomial() computation. * factor_univariate correctly handles polynomials over rationals (no segfaults any more). * Fixed an infinite loop in factor_multivariate due to incorrect handling of polynomials having a leading coefficient being a sum with a non-trivial integer content. * Fixed numeric evaluation of generalized multiple polylogarithms for real and negative argument. * Avoid calling log(1-x) in numeric evaluation of classical polylogarithm. * Fixed numeric evaluation of multiple polylogarithms (correctly set the signs of the imaginary parts in the intermediate expression). * Fixed numeric evaluation of Nielsen polylogs for arguments close to the 6th root of unity or its conjugate. * Avoid infinite recursion in series expansion of logarithm. * numeric::info(info_flags::nonnegative) correctly handles complex numbers with non-zero imaginary part. * Fixed several build issues: - libreadline and dlopen() misdetection. - multiple definition of symbol on MinGW. - auto* tools and texinfo being unhappy in various ways. * GiNaC can be built with CMake now. Regards, -richy. -- Richard B. Kreckel <http://in.terlu.de/~kreckel/>

1 0

How to put variables to the second argument of ex::ex()
by Shi 16 Nov '14

16 Nov '14

Hello, I'd like to convert std::string to ex constructor. But my code do not work now. How to put variables to the second argument of ex(const std::string &s, const ex &l). ----Code: #include <iostream> #include <ginac/ginac.h> using namespace std; using namespace GiNaC; int main(){ string st("2*a+3*sqrt(4)*b"); symbol a("a"), b("b"); ex sample(st, ex(a)); cout << sample <<endl; return 0; } ----Excuted: terminate called after throwing an instance of 'std::invalid_argument' what(): find_or_insert_symbol: symbol "a" not found IOT trap

1 0

For more information about software patents in Europe, visit this website: swpat.ffii.org.
by Gustavo de Lama 10 Sep '14

10 Sep '14

For more information about software patents in Europe, visit this website: swpat.ffii.org. That is what appears in the main page (http://www.ginac.de/) but that address redirects you to http://a2e.de/oas/ !!! I suspected http://ffii.org/ had been kidnapped. Alas, no! It seems to be "only" that page. In the hope somebody will be able to fix it. Gustavo de Lama

2 1

[BUG] commit 97ef604e breaks evaluation of some G-polylogs
by Valery Yundin 10 Sep '14

10 Sep '14

Hi, The following commit 97ef604e "G_numeric: fix numeric evaluation with trailing zeros and y != 1." breaks evaluation of G-polylogs with some arguments/ For instance: evalf( G({-1,1,-1,0},1) ); evalf( G({-1,-1,1,-1,0},1) ); evalf( G({-1,0,1,-1,0},1) ); evalf( G({-1,1,-1,-1,0},1) ); evalf( G({-1,1,-1,0,0},1) ); evalf( G({-1,1,-1,1,0},1) ); evalf( G({-1,1,0,-1,0},1) ); evalf( G({-1,1,1,-1,0},1) ); evalf( G({0,-1,1,-1,0},1) ); evalf( G({0,1,1,-1,0},1) ); results in exception: log_eval(): log(0) instead of the correct numerical value. With best regards, Valery Yundin.

1 0

Fwd: GiNaC Bug Report
by Mark Round 08 Sep '14

08 Sep '14

Dear all, The GiNaC feature is_polynomial() returns true in a scenario I feel is false. Namely, 1/(1-x) When running the attached code I obtain, -(-1+x)^(-1).is_polynomial(x) 1 x-(-1+x)^(-1).is_polynomial(x) 0 as output. I observed the result on two computers, MacBook Pro Darwin Kernel version 13.1.0 using GiNaC 1.6.2 64-bit DESY blade server scientific linux 2.6.18-371.1.2.el5 using GiNaC 1.6.2 Additionally, in preparing this report the following variations occurred, 1/(1-pow(x,2)) -> is polynomial 1/(1-pow(x,n)) -> is not polynomial (n being a symbol) 1/(1+x) -> is not polynomial on an SL6 installation running kernel version 2.6.32-431.20.3.el6.x86_64, gcc 4.4.7) using GiNaC 1.6.2 Yours, Mark

2 1