On Tue, 21 Jul 2009 13:03:48 +0100 Brian Gough <[email protected]> wrote: > At Mon, 20 Jul 2009 13:06:13 +0400, > Evgeny Kurbatov wrote: > > If the idea of gsl_const_cgs.h was to contain the constants from > > base CGS system only, we have to remove electron charge from it, > > then create gsl_const_cgse.h for right CGSE values. Actually we > > have to add constants of vacuum permittivity and permeability for > > completeness. If nobody dealing with it, I can because I'm > > interested in it. > > The original situation was that I generated the 'cgs' values from > another source (GNU Emacs Calc) without checking what conversion it > used, simply assuming that it was correct. This resulted in some of > the electromagnetic values having inconsistent units. > > So we migrated to gsl_const_cgsm.h to get away from that. > > Your idea of creating gsl_const_cgse.h and removing old > electromagnetic constants from gsl_const_cgs.h sounds like a good one. > I have not attempted it due to lack of familiarity with the different > CGS systems, so I would be happy for anyone to have a go. > > One issue is that the whole set of files is generated by Emacs Calc > (which has a multiple-precision unit conversion system), so it > requires some knowledge of Emacs Lisp and Calc to add new sets of > units. The system is capable of handling arbitrary unit systems, but > unfortunately somewhat complicated.
Ok, I'm ready to go. :) Could you give me an existing scripts for this transformation? My plan is: (i) clean-up gsl_const_cgs.h (remove non-basic constants from it) and gsl_const_cgsm.h (remove basic constants which will be included from gsl_const_cgs.h) (ii) make gsl_const_cgse.h and add into it and into gsl_const_cgsm.h some more electromagnetic constants. Best witches, Evgeny _______________________________________________ Bug-gsl mailing list [email protected] http://lists.gnu.org/mailman/listinfo/bug-gsl
