Hrvoje> sperber(a)informatik.uni-tuebingen.de (Michael Sperber [Mr. Preprocessor]) writes: Hrvoje> Isn't the whole idea with weak lists that they have special-purpose Hrvoje> code in the GC? Not if it's reducible to a simpler data structure. Right now, there's a plethora of weak data structures in the XEmacs core. We're trying to cut it down. This particular way of dealing with weak lists (which is like no other such mechanism I've come across) is especially hard to deal with, and has some hideous semantic issues. Like, what's this kind of thing going to do: (defvar x (make-weak-list)) (set-weak-list-list x (cons y z)) ... where z is not a list? Or this: (set-weak-list-list x [1 2]) ... all of which XEmacs currently accepts happily. -- Cheers =8-} Mike Friede, Völkerverständigung und überhaupt blabla