In general the problem of whether a local can be cleared is
undecidable. However a very advanced compiler might be able to find
out that the closure of a future will only be called once, and clear
the reference to the closure. It is much simpler to just special case
futures. One problem that might
Yes. Concerned about making future work, less concerned about hypothetical
examples. :-) The "I know I don't need what I closed over ever again" case can
be solved on a per-occurrence basis where it matters.
Stu
> On Wed, Aug 11, 2010 at 3:06 PM, Stuart Halloway
> wrote:
>> Chouser,
>>
>> The
On Wed, Aug 11, 2010 at 3:06 PM, Stuart Halloway
wrote:
> Chouser,
>
> There is now a ticket and roadmap for fixing this: See
> https://www.assembla.com/spaces/clojure/tickets/423-make-sure-future-clears-closed-overs.
Did you see that my examples didn't use future at all?
--Chouser
http://joyof
Chouser,
There is now a ticket and roadmap for fixing this: See
https://www.assembla.com/spaces/clojure/tickets/423-make-sure-future-clears-closed-overs.
Stu
> On Wed, Aug 11, 2010 at 1:25 AM, timcharper wrote:
>>
>> I've distilled the issue down to a series of two tests: in one case,
>> the
Hi,
Am 11.08.2010 um 07:25 schrieb timcharper:
> On Aug 7, 6:42 pm, Ben Mabey wrote:
>
>> (deftype Foo [])
>> ; The entire lazy-seq is held in memory resulting in an OutOfMemoryError.
>> (let [foos (take 1000 (repeatedly #(Foo.)))]
>> @(future (doseq [foo foos] foo)))
It looks like a non
On Wed, Aug 11, 2010 at 1:25 AM, timcharper wrote:
>
> I've distilled the issue down to a series of two tests: in one case,
> the head is properly released, and in the other, the head is retained
> even though it seems it shouldn't be.
>
> http://gist.github.com/510601
Those are nice tests. Plea
Have you considered taking a function that will produce the
collection?
(map-queue odd? #(range 1 10) :queue-size 5)
and in your function of course, (doseq [value (coll)]
So that there's no intermediary holding on to the head. Making your
function a macro would perhaps clean it up a bit so clien
On Aug 7, 6:42 pm, Ben Mabey wrote:
> Hi all,
> I've run into an issue with a lazy-seq either being prematurely realized
> or having the head unwittingly retained. Reading chapter 5 in The Joy
> of Clojure I realize that I am breaking one of the rules (page 150 in my
> MEAP version): avoid bind
