I guess this is a parallel discussion, right? https://github.com/pure-data/pure-data/pull/604
we've been discussing how to add an inlet to [receive] so it behaves like [iem_receive] Em qua., 25 de mar. de 2020 às 18:50, Christof Ressi <[email protected]> escreveu: > Haha, what I wanted to say is that if you're worried about the danger of > using [iem_receive], then dynamically destroying [r] is not a solution > because it is just as dangerous because the destructor of [r] will > unbind the symbol. > > An easy way to avoid this problem on the user side is to always use a > clock for the message that is going to unbind the symbol. E.g. the > following is always safe: > > ... -> [del 0] -> [set foo( -> [iem_receive] > > The reason is that [del 0] breaks the message passing chain and executes > the messages downstream at the beginning of the next scheduler tick. > Same thing for dynamically deleting objects. > > Don't worry too much. If you've been using [iem_receive] and dynamically > deleted objects and didn't experience a crash, it probably means you're > using it in more or less safe ways. It's just something to keep in mind > for your next patch :-) > > Christof > > > On 25.03.2020 22:16, oliver wrote: > > Christof Ressi wrote: > >> Check out the following discussion on GitHub: > >> https://github.com/pure-data/pure-data/pull/604 > >> > >> TL;DR: if you unbind a symbol from a receiver while sending to the > >> symbol, Pd can crash because you modify the bind list while iterating > >> over it. > >> > >> Personally, I've been using [iem_receive] in some projects and didn't > >> run into problems, but only because I avoid the case described above. > > > > Thanks Christof for the explanation ! > > > > @IOhannes: i think this should be noted in the helpfile. > > > > > >> Dynamically destroying objects is even more dangerous ;-) > > > > Bummer ... another thing i do quite often ;-) > > > > Since i am not a programmer: how would i decode that twinkle emoji in > > your sentence ? > > > > does it mean: > > > > a.) be happy that you survived so far, because you're really on mined > > territory here ... > > > > b.) i know it's not encouraged, but as long as you don't tell anybody ... > > > > > > > > sorry ... getting a little anxious these days ... ;-) > > > > best > > > > oliver > > > > > > > > _______________________________________________ > > [email protected] mailing list > > UNSUBSCRIBE and account-management -> > > https://lists.puredata.info/listinfo/pd-list > > > > _______________________________________________ > [email protected] mailing list > UNSUBSCRIBE and account-management -> > https://lists.puredata.info/listinfo/pd-list >
_______________________________________________ [email protected] mailing list UNSUBSCRIBE and account-management -> https://lists.puredata.info/listinfo/pd-list
