On Wed, 13 Aug 2003 11:02:27 -0400, Raymond Toy <[EMAIL PROTECTED]> wrote:

> Sorry for the slow response.  The fact that no one has responded
> probably means that they're away, are busy hacking on something else
> like dynamic-extents, or, like myself, don't know the answer.  :-)

Hehe. No, apologies for me making noise. Of course I'm aware of the
fact that I cannot expect anyone to hurry to answer my questions. I
just saw the amount of traffic on cmucl-imp and though "damn, none of
them seems to read cmucl-help"... :)

Gerd Moellmann just pointed out to me that 18e on x86 (which I'm
using) doesn't have weak hash tables at all. I could have checked that
myself but I was just looking at the docstring and thought it was
there. So, again, sorry for the noise. I'll be waiting for CMUCL 19...

> I don't know if what you say will work or not.  As another option,
> how about using finalizers?  When the object goes away, have the
> finalizer remove the entry from the other hash table.

Good idea, I'll try that.

Thanks,
Edi.

Reply via email to