Thanks. I'm not interested in reading the messages as pairs, I just need to make sure nothing gets in between them. I think I'll choose some way to merge the messages, for example just a vector like you suggested, and then have the end consumer recognise and split the vector. I guess splitting the vector into two messages somewhere before the end consumer can't be made deterministic.
keskiviikko 11. lokakuuta 2017 8.02.40 UTC+3 Rangel Spasov kirjoitti: > > I think simply wrapping the two values in a vector is the only thing > that's needed in this case. Simply do a put like (>!! ch [my-val-1 > my-val-2]) and take from the channel from another place. If my-val-1 and > my-val-2 are not immediately available, you can have a separate mechanism > (atoms, (loop [...] (recur ...)) or another set of core.async channels > perhaps) that "collects" them until they are ready to be sent together as > one value. > > partition-all on its own does not really help with the ordering here. For > example: > > (let [ch (chan 42 (partition-all 2))] > (>!! ch :x') > (thread ;in another > thread far, far away > (<!! (timeout (rand-int 1000))) ;simulate busy > for random 0-99ms > ;put value on ch > (>!! ch :y')) > > (<!! (timeout (rand-int 1000))) ;simulate busy > for random 0-99ms > (>!! ch :x'') > > ;take from ch > (<!! ch)) > > > This will non-deterministically output either > => [:x' :x''] > > or > > => [:x' :y'] > > But again as I said, you can employ partition-all separately as a > "collecting" mechanism before doing the (>!! ch [my-val-1 my-val-2]). > Didn't write an example for that but let me know if that's not clear and I > can put together a few lines. > > On Tuesday, October 10, 2017 at 7:50:20 AM UTC-7, Gary Trakhman wrote: >> >> So, at the point where you need 2 things to be consecutive, wrap them up >> in a vector so they're one thing. `(partition-all 2)` will return a >> transducer xform that chunks up your message stream by 2 >> https://clojuredocs.org/clojure.core/partition-all, and if you need to >> undo it you can do so with the `cat` transducer >> https://clojuredocs.org/clojure.core/cat . >> >> On Tue, Oct 10, 2017 at 10:42 AM JokkeB <jahv...@gmail.com> wrote: >> >>> Thanks for the response. That's what I suspected. >>> >>> How does partition help me in this case? I can't see any way of using it. >>> >>> How about onto-chan, would that be deterministic? Or any other function? >>> >>> >>> tiistai 10. lokakuuta 2017 17.33.22 UTC+3 Gary Trakhman kirjoitti: >>> >>>> I think a core assumption is that all writes can yield control at any >>>> time to other worker go processes. I wouldn't rely on the consecutive >>>> behavior even if it were true, nondeterminism is a good assumption with >>>> core.async. If you need that particular guarantee, consider a call to >>>> partition? >>>> >>>> On Tue, Oct 10, 2017 at 10:30 AM JokkeB <jahv...@gmail.com> wrote: >>>> >>> I'm wondering about a case when using async: >>>>> >>>>> I have a channel where I write from multiple threads. One of the >>>>> sources is a go-loop with a timeout, so each second I write something to >>>>> the channel. If I do two consecutive writes inside the go block, can >>>>> another thread write between the two writes? Or can I rely on the two >>>>> messages being consecutive in the channel? >>>>> >>>>> -- >>>>> You received this message because you are subscribed to the Google >>>>> Groups "Clojure" group. >>>>> >>>> To post to this group, send email to clo...@googlegroups.com >>>> >>>> >>>>> Note that posts from new members are moderated - please be patient >>>>> with your first post. >>>>> To unsubscribe from this group, send email to >>>>> >>>> clojure+u...@googlegroups.com >>>> >>>> >>>>> For more options, visit this group at >>>>> http://groups.google.com/group/clojure?hl=en >>>>> --- >>>>> You received this message because you are subscribed to the Google >>>>> Groups "Clojure" group. >>>>> >>>> To unsubscribe from this group and stop receiving emails from it, send >>>>> an email to clojure+u...@googlegroups.com. >>>> >>>> >>>>> For more options, visit https://groups.google.com/d/optout. >>>>> >>>> -- >>> You received this message because you are subscribed to the Google >>> Groups "Clojure" group. >>> To post to this group, send email to clo...@googlegroups.com >>> Note that posts from new members are moderated - please be patient with >>> your first post. >>> To unsubscribe from this group, send email to >>> clojure+u...@googlegroups.com >>> For more options, visit this group at >>> http://groups.google.com/group/clojure?hl=en >>> --- >>> You received this message because you are subscribed to the Google >>> Groups "Clojure" group. >>> To unsubscribe from this group and stop receiving emails from it, send >>> an email to clojure+u...@googlegroups.com. >>> For more options, visit https://groups.google.com/d/optout. >>> >> -- You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com Note that posts from new members are moderated - please be patient with your first post. To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en --- You received this message because you are subscribed to the Google Groups "Clojure" group. To unsubscribe from this group and stop receiving emails from it, send an email to clojure+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
