February 2000 - GiNaC-devel

Version 0.5.3 on the wire
by Richard B. Kreckel 24 Feb '00

24 Feb '00

i.e. on <ftp://ftpthep.physik.uni-mainz.de/pub/GiNaC/>. There has been a bug in the last version where .normal() was returning wrong results for certain rational functions like 1/(5-4*a)*(-5-2*a-(3+2*a)*(2-5*a)/(-2+a)) which is now fixed. Also, this version features an improved and more flexible scheme for registering functions which makes extensions easier and also allows writing remembering functions in a clean way. See <http://www.ginac.de/reference/class_remember_table.html>. GiNaC-cint is making slow progress---we have it running with the current cint 5.14.31 now. Sex and drugs and Unix -rbk. -- Richard Kreckel <Richard.Kreckel(a)Uni-Mainz.DE> <http://wwwthep.physik.uni-mainz.de/~kreckel/>

1 0

Build problems
by Patrick K Notz 22 Feb '00

22 Feb '00

Hi, I'me trying to build ginac and am having problems getting 'configure' to find the readline and cln libraries. I've installed them in directories (non-standard since I'm not root) under my account. I just can't figure out how to tell configure/make to look there. I've tried the suggestion in Section 3.2 of the tutorial but the syntax given doesn't work. I've tried adding "LDFLAGS = $(LDFLAGS) /my/lib/dir" to all of the Makefile.in files but no luck again. I'm not familiar with autoconf so I'm not sure which files I should try tweeking. Can you offer any suggestions? Thanks ~ pat <notz(a)purdue.edu>

2 1

Today's bugs
by Richard B. Kreckel 21 Feb '00

21 Feb '00

Consider this: higgsino:~/projects/GiNaC/ginsh$ ./ginsh ginsh - GiNaC Interactive Shell (GiNaC V0.5.2) __, _______ Copyright (C) 1999-2000 Johannes Gutenberg University Mainz, (__) * | Germany. This is free software with ABSOLUTELY NO WARRANTY. ._) i N a C | You are welcome to redistribute it under certain conditions; <-------------' see the file COPYING for details. Type ?? for a list of help topics. > series(gamma(x),x,0,2); collect(series(gamma(x),x,0,2),x); -EulerGamma+x*(1/12*Pi^2+1/2*EulerGamma^2)+x^(-1)+Order(x^2) -EulerGamma+x*(1/12*Pi^2+1/2*EulerGamma^2)+x^2*Order(1)+x^(-1) The other bug is with the fraction free elimination in ginac/matrix.cpp: If I try to solve the system [[ [[0,2-a]], [[5-4*a,3+2*a]] ]] * [[ [[x]], [[y]] ]] == [[ [[-2+5*a]], [[-5-2*a]] ]] I run into trouble within assert-blocks. I will have a look at the second problem. -rbk. -- Richard Kreckel <Richard.Kreckel(a)Uni-Mainz.DE> <http://wwwthep.physik.uni-mainz.de/~kreckel/>

3 3

normal(), numer(), denom()
by Christian Bauer 10 Feb '00

10 Feb '00

Hi! normal(), numer(), and denom() should work better now. normal() internally maintains a split representation {numerator, denominator} of the normalized expressions and will not divide them until the very end. Bye, Christian -- / Coding on PowerPC and proud of it \/ http://www.uni-mainz.de/~bauec002/

1 0

Version 0.5.0 uploaded
by Richard B. Kreckel 08 Feb '00

08 Feb '00

It's on <ftp://ftpthep.physik.uni-mainz.de/pub/GiNaC/>. Many things are fixed now and some sophisticated bugs were substantially improved. As always, the most mind-boggling new features are listed under <http://www.ginac.de/News.html>. Fun and happiness -rbk. PS: Apologies if you got this mail twice. Please do recycle the electrons. -- Richard Kreckel <Richard.Kreckel(a)Uni-Mainz.DE> <http://wwwthep.physik.uni-mainz.de/~kreckel/>

1 0

"Cannot resolve method ex::dbgprint to any overloaded instance" on gdb 4.18
by Richard B. Kreckel 05 Feb '00

05 Feb '00

Hi, Upon multiple request I had a quick look at the subject-mentioned problem that rendered the not-so-useless method dgbprint [1] unavailable. There is nothing to overload here (its argument being void anyhow) so this was kind of obscure. There seems to be a bug introduced in gdb 4.18 and it may already be fixed in the current CVS tree, since Daniel Berlin claims he fixed it a couple of days ago (on the gdb mailing list): > 2. gdb is screwing up the function ranking. Right in the comments it > says that nargs includes the this pointer in it's count, and nparams > doesn't, but it doesn't take this into account. So i made it take it > into account, and all of the sudden, it could resolve overloads > properly in most cases. > I can succesfully do a print AnSTLVector.begin() and AnSTLVector[0], > and if AnSTLVector was a vector of strings, I could do * > (AnSTLVector.begin()).length(), and get the right answer. In the meantime, you may simply tell your debugger to (gdb) set overload-resolution off and you can again build any ex foo and call foo.dbgprint() to display the contents. Those using DDD 3.2 may also set this in Edit->Preferences and save options. HTH -rbk. [1] The method dbgprint(void) is a small wrapper around print(ostream &, unsigned) used for debugging, where cout may not be present. -- Richard Kreckel <Richard.Kreckel(a)Uni-Mainz.DE> <http://wwwthep.physik.uni-mainz.de/~kreckel/>

1 0