- GiNaC-list - ginac.de

[patch] fix template template declarations [GCC 4.2 compilation fix]
by varg＠theor.jinr.ru 11 May '06

11 May '06

Hello! The attached patch corrects template template declarations, so GiNaC can be compiled with G++ 4.2. Without this patch compilation fails, see http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=362220 for details. Note that `make check' still fails with G++ 4.2. Basically, ex e("-1 + 2*I", symbol()); triggers an exception. My patch *does not* fix this error. Best regards, Alexei. -- All science is either physics or stamp collecting.

2 4

Announcement: SyFi 0.3.0
by Kent-Andre Mardal 19 Apr '06

19 Apr '06

Dear GiNaC users, I would like to announce the 0.3.0 version of SyFi (syfi.sf.net) which is a finite element library based on GiNaC. At present, we have implemented several finite elements, like e.g., Lagrangian, Raviart-Thomas, and Nedelec elements of arbitrary order in both 2D and 3D. The latest release adds several examples with code generation which demonstrate the use of libraries like e.g. the Dolfin (www.fenics.org) PyCC (heim.ifi.uio.no/~skavhaug/heart_simulations.html) and Trilinos (software.sandia.gov/trilinos) The name SyFi stands for Symbolic Finite Elements. The package provides polygonal domains, polynomial spaces, and degrees of freedom as symbolic expressions that are easily manipulated. This makes it easy to define finite elements. SyFi is open source. Thanks for GiNaC!! Regards, Kent-Andre Mardal Simula Research Laboratory

1 0

GiNaC 1.3.4 released
by Jens Vollinga 13 Apr '06

13 Apr '06

Hi everybody, GiNaC 1.3.4 is out and available. The changes are: - More general index dimensions are now possible. - Improved algebraic substitutions. - Fixed wrong matching in .has(). - Fixed bug in differentiation of a power. - collect_common_factors also works with negative powers of common factors now. - Code clean-up and preparations for upcoming cln 1.2. As always, this release can be downloaded from http://www.ginac.de/Download.html Best wishes, Jens

1 0

Rot, div, grad operator will be implemented in feelfem FEM code generator
by Fujio Hidehiro 10 Apr '06

10 Apr '06

Dear GiNaC developers, This is Hidehiro Fujio of NEC Corp. Japan. At first, I would like to say thank you very much for your significant contribution to the formula manipulation area by GiNaC library in the world, and I also have to say that thank you very much for your link to my small document in the Related Software page. (My system is feelfem, an FEM code generator.) Yesterday, I met my supervisor, and got his advice. I wll expand my feelfem system to deal with 3D Maxwell equation, and introduce the differential operator 'rot' and 'div' and 'grad' (rotation, divergence, gradient, respectively), and my professors suggestion is it may be possible to expand these differential operators into just a set of basic differential operators (like dx,dy,dz) inside GiNaC. I would like to investigate your documentation agaian, and I would like to make a reort, how to make an FEM code generator by GiNaC. Again, thank you very much for GiNaC. And I'm afraid it is not an appropriate mail............ Best regards, Hidehiro Fujio # mail to both ( h-fujio(a)ab.jp.nec.com , oimo(a)world.email.ne.jp ,if needed)

2 1

[patch] small improvement of collect_common_factors
by varg＠theor.jinr.ru 04 Apr '06

04 Apr '06

Hello! This ginsh code snippet: collect_common_factors(x/y + z/y^2); returns the original expression unchanged: z*y^(-2)+x*y^(-1) The patch below makes it return (x+z*y^(-1))*y^(-1) Index: normal.cpp =================================================================== RCS file: /home/cvs/GiNaC/ginac/normal.cpp,v retrieving revision 1.94.2.8 diff -u -r1.94.2.8 normal.cpp --- normal.cpp 30 Jun 2005 14:03:16 -0000 1.94.2.8 +++ normal.cpp 10 Mar 2006 13:40:26 -0000 @@ -2388,6 +2388,8 @@ { if (exponent.info(info_flags::posint)) return power(basis.to_rational(repl), exponent); + else if (exponent.info(info_flags::negint)) + return power(replace_with_symbol(power(basis, _ex_1), repl), -exponent); else return replace_with_symbol(*this, repl); } Best regards, Alexei. -- All science is either physics or stamp collecting.

2 1

[[1]]!=1 ??
by Javier Ros Ganuza 18 Mar '06

18 Mar '06

I suppose that it is a feature... (matrix(3,1,lst(0,0,1)).transpose()*matrix(3,1,lst(0,0,1))).evalm() gives a 1,1 matrix [[1]]. To convert it to a number or to a expression not containing a matrix, I have to take directly the 0,0 element of the matrix. Is there a more direct method? If not I'm facing errors like (in ginsh notation) > a=[[0,0,1]]*[[0],[0],[1]]; > a=evalm(a); [[1]] > 1+cos(a); add::eval(): sum of non-commutative objects has non-zero numeric term Thank you Javier Ros

2 1

(no subject)
by Javier Ros Ganuza 09 Mar '06

09 Mar '06

Hello! :) When performing matrix multiplications expressions get expanded like in the following Example. Is there a way to avoid expansion, that is to leave the parenthesis Thanks Javier Ros Example: ----------------- matrix aux1 [[-d331*sin(q3)*dq3-d332*cos(q3)*dq3],[d331*cos(q3)*dq3-d332*sin(q3)*dq3],[0]] matrix R23 [[cos(q3),-sin(q3),0],[sin(q3),cos(q3),0],[0,0,1]] matrix R12.mul(aux1) [[d332*sin(q2)*sin(q3)*dq3-d331*cos(q2)*sin(q3)*dq3-d332*cos(q3)*cos(q2)*dq3-d331*sin(q2)*cos(q3)*dq3],[-d332*cos(q2)*sin(q3)*dq3-d331*sin(q2)*sin(q3)*dq3+d331*cos(q3)*cos(q2)*dq3-d332*sin(q2)*cos(q3)*dq3],[0]] ----------------- What I want is to obtain something like cos(q3)*(-d331*sin(q3)*dq3-d332*cos(q3)*dq3)+.... and so on for the first (0,0) matrix element.

3 4

problem with function pow
by Saurav Pathak 08 Feb '06

08 Feb '06

hi, i am having trouble turning a string with function "pow" into an expression. an example of the code is given below along with the output. i cannot figure out what i am doing wrong. thanks for your help in advance. -------- #include <iostream> #include <string> #include <ginac/ginac.h> #include <ginac/power.h> using namespace std; using namespace GiNaC; int main() { lst all; all.append(symbol("x")); // no pow string str2 = "x*x"; ex e2(str2, all); cout << "nopow : " << e2 << endl; // with pow string str = "pow(x, 2)"; ex e(str, all); cout << "withpow: " << e << endl; return 0; } the output is: nopow : x^2 terminate called after throwing an instance of 'std::runtime_error' what(): no function 'pow' with 2 parameters defined Aborted -------------------- -- saurav

2 4

Re: [GiNaC-list] problem with function pow
by Saurav Pathak 31 Jan '06

31 Jan '06

Anish Muttreja [Mon, Jan 30, 2006 at 07:56:32PM -0500]: + Isn't the function called "power"? + http://www.ginac.de/reference/classGiNaC_1_1power.html hi ashish, there is a wrapper function called "pow". i have been able to use the following without any problem: ex e3 = pow(x,2); it seems if "pow" is in a string, it does not work. saurav + On Mon, Jan 30, 2006 at 07:31:20PM -0500, Saurav Pathak wrote: + >hi, + > + >i am having trouble turning a string with function "pow" into an + >expression. an example of the code is given below along with the + >output. i cannot figure out what i am doing wrong. + > + >thanks for your help in advance. + >-------- + >#include <iostream> + >#include <string> + >#include <ginac/ginac.h> + >#include <ginac/power.h> + >using namespace std; + >using namespace GiNaC; + > + >int main() + >{ + > lst all; + > all.append(symbol("x")); + > + > // no pow + > string str2 = "x*x"; + > ex e2(str2, all); + > cout << "nopow : " << e2 << endl; + > + > // with pow + > string str = "pow(x, 2)"; + > ex e(str, all); + > cout << "withpow: " << e << endl; + > + > return 0; + >} + > + >the output is: + >nopow : x^2 + >terminate called after throwing an instance of 'std::runtime_error' + > what(): no function 'pow' with 2 parameters defined + >Aborted + >-------------------- + >-- + >saurav + >_______________________________________________ + >GiNaC-list mailing list + >GiNaC-list(a)ginac.de + >https://www.cebix.net/mailman/listinfo/ginac-list -- saurav

1 0

let_op(size_t ) problem
by Kerson Rocha Jr. 20 Jan '06

20 Jan '06

The following code: symbol x("x"); symbol y("y"); ex test1 = x + y; for ( size_t i = 0 ; i != test1.nops() ; i++ ) { try { cout << test1.let_op(i) << " " ; } catch(std::range_error & ba) { cout << "i = " << i << endl; cout << ba.what() << endl; } } This results: i = 0 basic::let_op(): add has no operands i = 1 basic::let_op(): add has no operands Using op() runs ok but I want change subexpressions. I am using GiNaC 1.3.3 on Mandriva 2006.0. Thanks.

2 1