September 1999 - GiNaC-devel

cint
by Alexander Frink 22 Sep '99

22 Sep '99

Hi, I just wanted to keep you in touch with the cint (C/C++ interpreter) news. The latest version was delayed one week because Mr. Goto used a wrong FTP client which messed up the file name :-( Two of the bugs I reported are fixed, one is still around, and after two minutes I found a new one: typedef vector<T> Tvec; typedef vector<Tvec> Tvecvec; // gives an error class TTvec : public Tvec {}; typedef vector<TTvec> TTvecvec; // works Hopefully this is not too hard for Goto to fix. -- Alexander Frink E-Mail: Alexander.Frink(a)Uni-Mainz.DE Institut fuer Physik Phone: +49-6131-393391 Johannes-Gutenberg-Universitaet D-55099 Mainz, Germany

1 0

Theory symbolic computing (fwd)
by Dirk Kreimer 20 Sep '99

20 Sep '99

---------- Forwarded message ---------- Date: Mon, 20 Sep 1999 06:03:41 +0300 From: "Nazir S. Baaklini" <nsbaaklini(a)intracom.net.lb> To: dirk.kreimer(a)uni-mainz.de Subject: Theory symbolic computing Professor Nazir S. Baaklini, Dhour el Choueir, Lebanon Theoretical Physicist and Symbolic Math Phys Programmer ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Dear Dr Kreimer, I had recently developed a new Mathematica package which is quite useful for transparent symbolic manipulations and computations in theoretical physics, particularly classical and quantum field theory, with several applications. I invite you to have a look at anyone of the following equivalent homepages, which would introduce this package and related works. You would also be able to download freely several Mathematica notebooks that would illustrate the various uses of this new program package. http://webs.intracom.net.lb/nsbaaklini http://www.nsbaaklini.homepage.com For additional information and for email: ............................. [ nsbaaklini(a)intracom.net.lb ] ............................. [nsbaaklini(a)usa.net] Yours sincerely, Nazir S. Baaklini

2 2

Tutorial
by Richard B. Kreckel 18 Sep '99

18 Sep '99

The tutorial is now getting into shape. The newest version, including Alexander's chapter about "why on earth?" is now in CVS. Sorry, you still cannot built it yourself. But may I ask to have a quick look at it and comment what is missing or what you would expect from a text that teaches the basic things about GiNaC (and ideally write and contribute it)? For some of you I have put a printed version into the mail boxes, the online version in HTML is sitting at <http://wwwthep.physik.uni-mainz.de/GiNaC/tutorial/>. ¡Hasta la vista! -richy. -- Richard Kreckel <Richard.Kreckel(a)Uni-Mainz.DE> <http://wwwthep.physik.uni-mainz.de/~kreckel/>

1 0

A proposal
by Richard B. Kreckel 17 Sep '99

17 Sep '99

I would like to have some comments from you about a proposal about constants: In order to forbid insane code like this one in GiNaC: EulerGamma = Pi; cout << EulerGamma << endl; // prints 3.14... all objects of class constant need to be declared const (which is the case for built-in constants). While the copy constructor for constants is sane (because when you declare constant EulerGamma(Pi); this is a new constant with local scope) the above case is not. One could override constant::operator= but there are other cases. E.g. I wanted to override constant::subs() to return *this but while this seemlingly worked for the checks it caused basic's virtual table to be overridden with garbage (and then executed, which caused a SEGV beeing raised) by the xloops check. I do not have any clue why this happens. Also, I had to change symbol::subs to be somewhat more flexible. The whole trouble is IMHO caused by the fact that symbol is currently partly caring for its derived class constant. Can't we just eliminate class constant and declare global symbols like Pi, EulerGamma and the like and assign them an evalf()-function similar to how expressions may be assigned to symbols? Then just declare those special symbols const and that's it? I haven't tried it, just wanted to hear your comments and sleep a night before doing something, maybe it's rubbish! Cheers -rbk. -- Richard Kreckel <Richard.Kreckel(a)Uni-Mainz.DE> <http://wwwthep.physik.uni-mainz.de/~kreckel/>

3 8

ginsh - The GiNaC Interactive Shell
by Christian Bauer 16 Sep '99

16 Sep '99

Hi! Due to popular demand I have written a little interactive shell ("ginsh") for GiNaC with flex/bison. The error handling is catastrophic but most of the basic functionality is there (entering and printing expressions, assigning symbols, functions, and expression manipulation). The syntax is mostly like you would write it in C++ (but you use "^" for powering). You can exit ginsh by entering "quit", "exit" or EOF (Ctrl-D). A small example session: ghost:~/project/ginsh$ ./ginsh ginsh - GiNaC Interactive Shell V0.1 Copyright (C) 1999 Johannes Gutenberg Universitaet Mainz, Germany This is free software, and you are welcome to redistribute it under certain conditions; see the file COPYING for details. > a=x+y; x+y > expand(a^3); 3*y^2*x+3*y*x^2+y^3+x^3 > "/a; x+y^(-1)*(3*y^2*x+3*y*x^2+y^3+x^3) [NOTE: there is an output bug here, it should be "(x+y)^(-1)*..."] > normal("); 2*y*x+y^2+x^2 > diff(",x); 2*x+2*y > subs(",x==2); 2*y+4 > quit ghost:~/project/ginsh$ There is no documentation yet, so consult the source if you have questions. :-) To get ginsh, type cvs checkout ginsh in your "projects" directory. Bye, Christian -- / Coding on PowerPC and proud of it \/ http://www.uni-mainz.de/~bauec002/

2 1

Functions...
by Richard B. Kreckel 15 Sep '99

15 Sep '99

Hello, most of the functions are now written in lowercase, i.e. it's log(x), not Log(x) any more, reflecting the new scheme where functions are not classes any more. Just wanted you to keep this in mind... Happy hacking -richy. PS: ob-mapleV-goof-of-the-day: Does anybody know what this is supposed to be? > restart; > exp(matrix(2,2,[[1,0],[0,1]])); [1 0] exp([ ]) [0 1] > evalf("); exp(x) > x; x > log(matrix(2,2,[[exp(1),0],[0,exp(1)]])); [exp(1) 0 ] ln([ ]) [ 0 exp(1)] > evalf("); ln(x) -- Richard Kreckel <Richard.Kreckel(a)Uni-Mainz.DE> <http://wwwthep.physik.uni-mainz.de/~kreckel/>

1 0

problem with normal()
by Alexander Frink 10 Sep '99

10 Sep '99

Hi! It is time to open the discussion in this mailing list, since all developers seem to be subscribed. I found a problem with normal() which did not exist in older versions of GiNaC, but I have not found out yet when exactly it broke: symbol z("z"); symbol a("a"); ex e=-(z+1)/(2*a-1)+2*(z+1)*a/(2*a-1); cout << "e=" << e << endl; //e.printtree(cout); cout << "e.normal()=" << e.normal() << endl; //e.normal().printtree(cout); cout << "e.normal().normal()=" << e.normal().normal() << endl; //e.normal().normal().printtree(cout); gives e=(2*a-1)^(-1)*(-z-1)+(2*a-1)^(-1)*(2*z+2)*a e.normal()=(2*a-1)^(-1)*(2*a*z-z+2*a-1) e.normal().normal()=z+1 The printtree() output looks reasonable. -- Alexander Frink E-Mail: Alexander.Frink(a)Uni-Mainz.DE Institut fuer Physik Phone: +49-6131-393391 Johannes-Gutenberg-Universitaet D-55099 Mainz, Germany

2 2

New CLN released
by Richard B. Kreckel 07 Sep '99

07 Sep '99

So, nearly everybody has succeded subscribing to the two new lists. After having removed some oopses I suppose everything is up and running now. Welcome folks! It seems like last weeks effort in fixing the anoying radical-bug in CLN has shown some effect on Bruno Haible's side. He put up a new version (CLN 1.0.2) which incorporates my fix and some other stuff. I am just downloading it. Here are some excerpts from the mail he sent me, maybe somebody finds something useful. (Sorry, it's in German.) > 1999-03-09: > > Hierfür muß ich hin und wieder abfragen, ob ein solches > > Objekt nun ein Integer ist oder nicht. Dafür gibt es kein vorgesehenes > > Interface. Man kann natürlich mit cl_I_p() herumtricksen. > > cln-1.0.2 wird ein echtes `instanceof' enthalten, z.B. > > cl_N x = ... > if (instanceof(x,cl_RA_ring)) > // x ist rationale Zahl > > 1999-05-31: > > Wir würden gerne eine Hash-Funktion > > haben, um Objekte vom Typ cl_N zu hashen. Ein unsigned würde uns > > ausreichen. Ich frage mich nur, wie ich das anstellen sollte. Die > > Hash-Funktion sollte halt dieselben Werte zurückliefern wenn die > > Zahlen identisch sind > > cln-1.0.2 wird eine solche Funktion `cl_equal_hashcode' enthalten, die > so gebaut ist, dass z.B. 3/4 und 0.75 denselben Hashcode haben. Oops, Alex, what do you think about this? Shouldn't they be different? Can 0.75 be exact at all??? > > Bei rationalen Zahlen wäre auch ein > > Zugriff auf Zähler / Nenner praktisch. > > Wird ebenfalls drin sein. Love and Happiness! -richy. -- Richard Kreckel <Richard.Kreckel(a)Uni-Mainz.DE> <http://wwwthep.physik.uni-mainz.de/~kreckel/>

1 0