Gabriel Dos Reis <[EMAIL PROTECTED]> writes: > On Sat, 22 Dec 2007, Martin Rubey wrote: > > | > ... but in the immediate future this is not something I'm working on -- I > | > have pressing regressions to fix first. > | > | Too bad. I was hoping I could get you hooked. > > It is difficult to hook someone already hooked. > > FWIW, I'm not bothered by not having every runtime stuff at compile > time, but I appreciate there are opposite opinions.
I'm not sure I understand. All I'm saying is that *this* restriction is severe, as without it, it's impossible to have Complex R export Field only when this is the case - and this breaks factorization. Its a shame. I'm sure there are other situations where this applies, too, but I do not know them from the top of my head. All the best, Martin NB: In fact, Ralf and I were having a similar "dual" problem with aldor - only on a different level: we wanted to be able to create domains given a (possibly recursive) specification in the form of a string. For example, such a specification for binary trees would be "B=Plus(Atom, Times(B,B))". To create the domain of binary trees at runtime (or in the axiom interpreter), we want to be able to say BinTree ==> Interpret(parse "Plus(Atom, Times(B,B))") and then, for example, structures([1,2,3,4])$BinTree to get the BinaryTree s with 4 leaves. In the end I managed to find a workaround, although, it seems that it is not legal Aldor. Well, it works with all revisions of the Aldor Compiler so far, so I'm more or less happy. So, sometimes it would be good to have the compiler at runtime... ------------------------------------------------------------------------- This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ _______________________________________________ open-axiom-devel mailing list open-axiom-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/open-axiom-devel
