- GiNaC-list - ginac.de

Happy birthday, GiNaC!
by Christian Bauer 27 Nov '24

27 Nov '24

Greetings! Today marks the 25th anniversary of the first public release of GiNaC. (Normally, we would have thrown a big party near its original birthplace, but that has unfortunately been torn down a while ago 😅: https://gutenberg-netzwerk.uni-mainz.de/aktuelles/abriss-des-inter-1-eine-c…) Many thanks to everyone who has contributed to GiNaC over all those years! Bye, Christian -- / Physics is an algorithm \/ www.cebix.net

3 2

Unwanted digits when performing arithmetic computations
by Dr. Mithun Bairagi 27 Oct '24

27 Oct '24

Dear all ginac lovers, I have got some wrong results when performing some simple computations using ginac. I have compiled and run the following c++ program using ginac-1.8.7, cln-1.3.7 (mingw64 compiler, gcc version 14.2.0, windows 10 os), #include <ginac/ginac.h> using namespace std; using namespace GiNaC; int main() { cout<<(numeric(0.1)+numeric(7))<<endl; cout<<(numeric("0.01")*numeric("9"))<<endl; cout<<(numeric("0.01")*numeric("99"))<<endl; } the above program shows the following result: 7.1000000000000000056 0.089999999999999999996 0.98999999999999999995 I have got some unwanted digits which look very ugly. I think the result should be 7.1 0.09 0.99 Best regards, Mithun Bairagi

3 3

Patch for ginac-1.8.7 (MacOS 14.6,Apple Silicon)
by 足立聡 02 Aug '24

02 Aug '24

Dear Developers of GiNaC, [0] Thank you very much for developing and maintaining GiNaC, which is a very nice system to write C++ programs for symbolic computation. [1] During recent several months, I had a trouble to build ginac-1.8.7 on my MacBookPro (Apple Silicon). The configuration and compilation steps succeeded, but the check step failed at two tests as follows: ./configure --> checking for a BSD-compatible install... /Ext/sw/bin/ginstall -c ... Configuration of GiNaC 1.8.7 done. Now type "make". make --> Making all in ginac ... /Ext/sw/bin/makeinfo --no-split --html ./ginac-examples.texi make check --> Making check in ginac ... PASS: exam_paranoia ... PASS: exam_color FAIL: exam_clifford FAIL: exam_archive PASS: exam_structure ... PASS: time_parser ============================================================================ Testsuite summary for GiNaC 1.8.7 ============================================================================ # TOTAL: 61 # PASS: 59 # SKIP: 0 # XFAIL: 0 # FAIL: 2 # XPASS: 0 # ERROR: 0 ============================================================================ See check/test-suite.log Please report to ginac-list(a)ginac.de ============================================================================ My environment is as follows: Hardware: MacBookPro (Apple Silicon,14inch,2023) OS: MacOS 14.6 Compiler: Xcode 15.4 g++ --version --> Apple clang version 15.0.0 (clang-1500.3.9.4) Target: arm64-apple-darwin23.6.0 Thread model: posix InstalledDir: /Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/bin [2] I have investigated what caused this trouble and found that the phenomenon described in "ginac/hash_seed.h" occurs not only in Win32 but also in my MacOS 14.6. typeid(ex.bp).name() returns always correct C string as value but its address varies from a call to another call. Accordingly, the problem can be fixed as follows: *** ginac/hash_seed.h.org Sun Jan 1 03:47:20 2023 --- ginac/hash_seed.h Tue Jul 30 23:02:36 2024 *************** *** 8,14 **** #include <typeinfo> #include <cstring> #include "utils.h" ! #ifdef _WIN32 #define GINAC_HASH_USE_MANGLED_NAME 1 #endif --- 8,14 ---- #include <typeinfo> #include <cstring> #include "utils.h" ! #if defined(_WIN32) || defined(__APPLE__) #define GINAC_HASH_USE_MANGLED_NAME 1 #endif [3] Now, everything is fine. make distclean ./configure ... make ... make check --> Making check in ginac ... PASS: exam_paranoia PASS: exam_heur_gcd PASS: exam_match PASS: exam_parser PASS: exam_numeric PASS: exam_relational PASS: exam_powerlaws PASS: exam_collect PASS: exam_inifcns PASS: exam_inifcns_nstdsums PASS: exam_inifcns_elliptic PASS: exam_differentiation PASS: exam_polygcd PASS: exam_collect_common_factors PASS: exam_normalization PASS: exam_sqrfree PASS: exam_factor PASS: exam_pseries PASS: exam_matrices PASS: exam_lsolve PASS: exam_indexed PASS: exam_color PASS: exam_clifford PASS: exam_archive PASS: exam_structure PASS: exam_misc PASS: exam_pgcd PASS: exam_mod_gcd PASS: exam_chinrem_gcd PASS: exam_function_exvector PASS: exam_real_imag PASS: check_numeric PASS: check_inifcns PASS: check_matrices PASS: check_lsolve PASS: check_cra PASS: time_dennyfliegner PASS: time_gammaseries PASS: time_vandermonde PASS: time_toeplitz PASS: time_lw_A PASS: time_lw_B PASS: time_lw_C PASS: time_lw_D PASS: time_lw_E PASS: time_lw_F PASS: time_lw_G PASS: time_lw_H PASS: time_lw_IJKL PASS: time_lw_M1 PASS: time_lw_M2 PASS: time_lw_N PASS: time_lw_O PASS: time_lw_P PASS: time_lw_Pprime PASS: time_lw_Q PASS: time_lw_Qprime PASS: time_antipode PASS: time_fateman_expand PASS: time_uvar_gcd PASS: time_parser ============================================================================ Testsuite summary for GiNaC 1.8.7 ============================================================================ # TOTAL: 61 # PASS: 61 # SKIP: 0 # XFAIL: 0 # FAIL: 0 # XPASS: 0 # ERROR: 0 sudo make install ... [4] There may be other people who run into the same problem as I. In that case, please utilize this information. Finally, GiNaC is a really nice system to write programs for symbolic computation. Thank you very much for publishing GiNaC to help us. Sincerely yours, Satoshi Adachi

2 1

I want to Add Integrating by Parts in GiNaC Symbolic Integration
by DS Glanzsche 01 Aug '24

01 Aug '24

Dear all, I just learned about GiNaC, and I want to add integration by parts for a bit of complex integration like integral of *cos nx * x*, for definite and indefinite integral. I think polynomial integration in GiNaC works amazing, but if we can add all other symbolic integration it will be better. Should I look and modify the source code *integral.cpp*? I never really modified open source code before in my life. Thanks.

2 3

Memory leaks
by tstk3268 30 Jul '24

30 Jul '24

HI, GiNaC seems to have memory leaks that show up in a no-op program under Windows/MSVC using the _CrtSetDbgFlag mechanism. This seems to be due to static initializers. Is this a known and intended feature? It makes developing a proper leak free application pretty difficult.

2 1

Access denied when cloning git repository
by Matthias Volk 08 Jul '24

08 Jul '24

Hello, I am trying the clone the GiNaC and CLN git repositories. Both commands git clone git://www.ginac.de/cln.git and git clone git://www.ginac.de/ginac.git fail with the following error message: fatal: remote error: access denied or repository not exported: /ginac.git Am I using a wrong command or are the git repositories currently not available? Best regards, Matthias

2 2

degree function
by Santos Jha 04 Jul '24

04 Jul '24

Greetings, I am new to GiNac. I was trying to use the degree function, but I got a weird result. My initial expression is -4*(-3+x)*(-1+x)+9/2*(-2+x)*(-1+x)+1/2*(-2+x)*(-3+x) After the "expand" function call, it returns 9/2*x^2-5/2*x-4*x^2+16*x-27/2*x+1/2*x^2 My goal is to add/subtract the coefficient of equal degree terms. I could not find any function for that. ( if it is already there, please point me) I wanted to write a function to achieve it. To do so. I am taking each term and try to see if for (size_t i = 0; i != poly.nops(); ++i) { // Here poly is polynomial as above ex subExpres=poly.op(i); // I get individual terms cout << "is polynomial="<< is_polynomial(subExpres,symb) <<endl; // GiNaC::ex pow2=pow(symb,2); cout << subExpres<< " degree=" << subExpres.degree(symb) << endl; } It determines the subexpression as a polynomial but can not determine the degree. E.g is polynomial=1 9/2*x^2 degree=0 // Here degree returned is wrong. Your thoughts will be appreciated. Regards, Santos

2 2

Ginsh Compiler Issues on Mac OS
by Francesca C Fernandes 29 Jun '24

29 Jun '24

Hello, I am trying to install GINAC on Mac OS. ginsh did not install because it could not find readline. I have already installed readline using Homebrew, and homebrew gave me the message appended below. So I used this configure option: ./configure LDFLAGS="-L/usr/local/opt/readline/lib" CPPFLAGS="-I/usr/local/opt/readline/include" PKG_CONFIG_PATH="/usr/local/opt/readline/lib/pkgconfig" Unfortunately, ginsh still does not compile and I get this information from configure: ***** BEGIN MESSAGE FROM CONFIGURE***** == ginsh will not compile, because libreadline could not be found. *****END MESSAGE FROM CONFIGURE***** *Could you please help me with the configure options to get ginsh to compile?* I’ve attached the config.log as well. Thank you! Sincerely, Francesca *****BEGIN MESSAGE FROM HOMEBREW **** For compilers to find readline you may need to set: export LDFLAGS="-L/usr/local/opt/readline/lib" export CPPFLAGS="-I/usr/local/opt/readline/include" For pkg-config to find readline you may need to set: export PKG_CONFIG_PATH="/usr/local/opt/readline/lib/pkgconfig" readline is keg-only, which means it was not symlinked into /usr/local, because macOS provides BSD libedit. For compilers to find readline you may need to set: export LDFLAGS="-L/usr/local/opt/readline/lib" export CPPFLAGS="-I/usr/local/opt/readline/include" For pkg-config to find readline you may need to set: export PKG_CONFIG_PATH="/usr/local/opt/readline/lib/pkgconfig" ******END MESSAGE FROM HOMEBREW******

2 1

bugs in indexed object
by Dr. Mithun Bairagi 14 Apr '24

14 Apr '24

Dear all ginac lovers, I have found simplify_indexed() is not working as expected. I have compiled and run the following c++ program using ginac 1.87 (mingw64 compiler, windows 10 os), #include <ginac/ginac.h> using namespace GiNaC; using namespace std; int main() { symbol A("A"), B("B"), C("C"), i_sym("i"),j_sym("j"); varidx i(i_sym, 3); varidx ic(i_sym, 3,true); varidx j(j_sym, 3); varidx jc(j_sym, 3,true); ex e = indexed(A, {i,i}) * indexed(B, {ic,ic})-indexed(A, {j,j}) * indexed(B, {jc,jc}); cout << (e).simplify_indexed() << endl; return 0; } the above program shows the following result: A.i.i*B~symbol16.symbol16-A.i.i*B~symbol17.symbol17 But I think the result should be zero. Please tell me what is the problem here? Best regards, Mithun Bairagi

2 1

GPL v2 and v3 compatibility issues
by Diddens, Christian (UT-TNW) 13 Apr '24

13 Apr '24

Dear GiNaC team, first of all, thanks so much for your powerful projects GiNaC and CLN! Thanks so much for making them open to us all! Based on your projects (and others), I have developed a nice multi-physics finite element framework, which I would like to share with the world soon. Of course, I will also go for the GPL, but I get conflicts, since I also have another dependency (pygmsh) using GPL v3, which is not compatible with GPL v2 of GiNaC and CLN. I total, I have to satisfy BSD-3-Clause, BSD-2-Clause, MIT, LGPL-2.1-only, LGPL-2.1-or-later, GPL-2.0-or-later, GPL-2.0-only and GPL-3.0-only. This configuration is not compatible under any outbound licence, unless I leave out either the GPL-2.0-only dependencies (GiNaC & CLN) or the GPL-3.0-only dependency (pygmsh). Of course, I have read your statement in the FAQ: ---- I don't like the GPL. May I have another license, please? A: The idea to dual-license GiNaC has been proposed a number of times. It has, however, been squarely rejected by some developers. Asking for a permission to dual-license it is likely going to waste your valuable time. That being said: The GPL is a good license. Why not directly put derived work under the GPL, too? ---- I am not asking for a dual-license, I support the ideas of the GPL myself, but I just wonder whether there are any plans to go for GPL-2.0-or-later at any time? If not, I have to get rid of the GPL-3.0 dependency, i.e. write my own interface to gmsh, before I can publish my work under GPL-2.0-only as well. Thanks for any information, help or suggestions in this issue! Best regards, Christian

1 1