[Larceny-users] Weak references

[David Rush]

On 7/24/07, William D Clinger <will at ccs.neu.edu> wrote:
> David Rush wrote:
> > Are there weak references in there that can be exploited by an application?
>
> I don't think the current code base contains any form
> of weak references, unless you count gctwa.

It certainly didn't look like there were. Thanks for the confirmation.

> Weak references would be pretty easy to implement,
> however.  Why don't you point me to your preferred
> API for weak references, so I can respond with an
> implementation of it?

Because I have never actually *used* one before to build this kind of
caching solution. The problem I have is attempting to minimize DB
traffic for a long-running online ETL process into a complex data
structure. Essentially if I knew how to partition the data stream to
keep an optimal data set in memory, I wouldn't need to run the program
:)

I'm pretty sure that both PLT and Gambit have weak-reference and/or
object-finalization systems and I will try to take a look in the next
few days  to express a preference. In the meantime, I guess I will
have to find out just how much memory I'm going to end up using
without fancy cache expiration machinery.  It's always good to measure
first :)

> (I'll respond to your FFI questions later; they're
> harder to answer.)

I thought the 64-bit issues might be kinda tricky, that's why I asked
before crawling down the rabbit hole :)

david rush
-- 
Once you label me, you negate me
    - Soren Kierkegaard