Help needed... Martin Rubey <[EMAIL PROTECTED]> writes:
> Martin Rubey <[EMAIL PROTECTED]> writes: > > > Dear Waldek, > > > > thanks a million! > > > > > Basically, one must find all places where we use symbols which may end up > > > as arguments to SPADCALL. Actually, one may also replace symbols by > > > functions in other funcall-s -- Juergen Weiss did this. > > > > I only missed the ones in i-eval and i-coerce... Great - that worked! > > I just ran the testsuite. There are surprisingly few errors, most (all?) of > them being due to the fact that anonymous functions do not seem to work > anymore. Unfortunately (stupid me) I deleted the non-optimized version, so I > do not know whether it's due to optimization or because somewhere a coercion > to FUNCTION is missing... I recompiled without optimisation. As I suspected, the trouble is that anonymous functions as in i +-> i map(i +-> i, [1,2,3]) are not converted into a lisp function object, when I apply only the patches Waldek sent. However, I cannot find the place where to convert these functions, usually named |*1;anonymousFunction;0;initial;internal|. Apart from that, turning optimization on and eliminating the need for SPADCALL to check whether it get's a function or not makes a huge difference in speed. I really think we should pursue that further. Help would be greatly appreciated. Martin ------------------------------------------------------------------------- Check out the new SourceForge.net Marketplace. It's the best place to buy or sell services for just about anything Open Source. http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace _______________________________________________ open-axiom-devel mailing list open-axiom-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/open-axiom-devel
