Thank so much for your responses. I will test out the current functionality and let you know my impression. Please let me know if you need any more on implementing Coxter groups.
On Mon, Jul 25, 2011 at 4:15 AM, Nicolas Borie <nicolas.bo...@math.u-psud.fr> wrote: > Le dimanche 24 juillet 2011 à 15:38 -0700, Dima Pasechnik a écrit : >> On Sunday, 24 July 2011 22:37:05 UTC+1, Rafael T wrote: >> >> The gap-system already has the ability to work with Coxeter >> groups: >> http://www.gap-system.org/Gap3/Manual3/C075S005.htm Maybe I >> could use >> some of their code? > >> GAP 3 (not the current GAP 4) has it, as an interface to a package >> called Chevie (written mostly in Maple, if I recall right), but it's >> rather old unmaintained stuff. > > I can't manage to use it now... So, it is perhaps broken these days (the > sage-combinat queue of patches is relatively experimental...). > > To get the implementation of Coxeter groups in Sage, there is a lot of > things to do yourself. > > **************************** > See : > > print gap3._install_hints() > > Your attempt to start GAP3 failed, either because you do not have > have GAP3 installed, or because it is not configured correctly. > > - If you do not have GAP3 installed, then you must download and > install it yourself because it is not distrubuted with Sage. > Here are some ways to obtain GAP3: > > - There is an optional Sage package providing GAP3 pre-packaged > with several GAP3 packages: > http://trac.sagemath.org/sage_trac/ticket/8906 > > - Frank Luebeck maintains a GAP3 Linux executable, optimized > for i686 and statically linked for jobs of 2 GByte or more: > http://www.math.rwth-aachen.de/~Frank.Luebeck/gap/GAP3 > > - Jean Michel maintains a version of GAP3 pre-packaged with > CHEVIE and VKCURVE. It can be obtained here: > http://people.math.jussieu.fr/~jmichel/chevie/chevie.html > > - Finally, you can download GAP3 from the GAP website below. > Since > GAP3 is no longer an officially supported distribution of GAP, > it > may not be easy to install this version. > http://www.gap-system.org/Gap3/Download3/download.html > > - If you have GAP3 installed, then perhaps it is not configured > correctly. Sage assumes that you can start GAP3 with the command > /opt/gap3r4p4/bin/gap.sh. Alternatively, you can use the following > command > to point Sage to the correct command for your system. > > gap3 = Gap3(command='/usr/local/bin/gap3') > > ******************************** > > Once you will have gap3 on your computer, it is not still ended. You > will have to make Sage know there Gap3 and chevie available. > > On my computer, gap3 is at /opt/gap3r4p4/ . I tell it to sage like : > > gap3 = Gap3(command='/opt/gap3r4p4/bin/gap.sh') > > after that, i can do : > sage: gap3.console() > > ######## Lehrstuhl D fuer > Mathematik > ### #### RWTH > Aachen > ## > ## > ## # ####### > ######### > ## # ## ## # > ## > ## # # ## # > ## > #### ## ## # # > ## > ##### ### ## ## ## > ## > ######### # ######### > ####### > # > # > ## Version 3 > # > ### Release 4.4 > # > ## # 18 Apr 97 > # > ## > # > ## # Alice Niemeyer, Werner Nickel, Martin > Schoenert > ## # Johannes Meier, Alex Wegner, Thomas > Bischops > ## # Frank Celler, Juergen Mnich, Udo > Polis > ### ## Thomas Breuer, Goetz Pfeiffer, Hans U. > Besche > ###### Volkmar Felsch, Heiko Theissen, Alexander > Hulpke > Ansgar Kaup, Akos Seress, Erzsebet > Horvath > Bettina > Eick > For help > enter: ?<return> > gap> RequirePackage("chevie"); > > Welcome to the CHEVIE package, Version 4.devel (Feb 2010) > http://www.math.rwth-aachen.de/~CHEVIE for the stable version 3.1 > http://www.math.jussieu.fr/~jmichel/chevie for this version > > Meinolf Geck, Frank Luebeck, Gerhard Hiss, > Gunter Malle, Jean Michel, Goetz Pfeiffer > Lehrstuhl D fuer Mathematik, RWTH Aachen > Universit'e Paris VII > AG Computational Mathematics Universit"at Kassel > Galway University > > This replaces the former weyl package. For first help type > ?CHEVIE Version 4 -- a short introduction > > gap> quit;; > sage: > > ***************************** > > This is a good prerequisite before getting that you want. Even you got > that, I think you will still have to wait because something doesn't work > in loading chevie via the Gap3 interface. On my computer, the error is > there : > > sage: W = CoxeterGroup(["A",2]) > --------------------------------------------------------------------------- > AssertionError Traceback (most recent call > last) > > /opt/sage/devel/sage-combinat/<ipython console> in <module>() > > /opt/sage/local/lib/python2.6/site-packages/sage/combinat/root_system/coxeter_group.pyc > in CoxeterGroup(*args, **kwds) > 252 return > sage.libs.coxeter.coxeter_group.CoxeterGroup(cartan_type) > 253 > --> 254 assert is_chevie_available() > 255 gap3.load_package("chevie") > 256 > > AssertionError: > sage: > sage: gap3.load_package("chevie") > --------------------------------------------------------------------------- > UnboundLocalError Traceback (most recent call > last) > > /opt/sage/devel/sage-combinat/<ipython console> in <module>() > > ... > > UnboundLocalError: local variable 'f' referenced before assignment > sage: > sage: from sage.combinat.root_system.complex_reflection_group import > is_chevie_available > sage: is_chevie_available() > False > *************************** > > A lot of people from combinat worked on that. To give somes manes, I > think about Saliola, Hivert, Thiery, ... But a lot of them are probably > not available these days. > > Also, you will meet faster an expert by posting on sage-combinat-devel. > I thus send it to this groups. > > Cheers, > Nicolas B. > > -- > To post to this group, send an email to sage-devel@googlegroups.com > To unsubscribe from this group, send an email to > sage-devel+unsubscr...@googlegroups.com > For more options, visit this group at > http://groups.google.com/group/sage-devel > URL: http://www.sagemath.org > -- To post to this group, send an email to sage-devel@googlegroups.com To unsubscribe from this group, send an email to sage-devel+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/sage-devel URL: http://www.sagemath.org
