I used to think the same, until I wrote the current incarnation of the go macro in core.async. A few months back I ripped out all global vars (and bindings). The go "compiler" is now functionally pure, except for a single atom on the edge that counts locals. More and more I'm convinced that dynamic vars, while pragmatic, are used way more often than they should be.
A classic example, is this OP use case, why are we storing a url in a public dynamic var? Timothy On Thu, Sep 5, 2013 at 10:08 AM, David Nolen <dnolen.li...@gmail.com> wrote: > Your example doesn't accomplish what dynamic binding does. You don't need > to look much further than Clojure / ClojureScript compilers to see cases > where trying to propagate information through an environment like you are > with ctx is absolutely too tedious, the aspect oriented nature of binding > is really what you want. > > David > > > On Thu, Sep 5, 2013 at 1:03 PM, Timothy Baldridge <tbaldri...@gmail.com>wrote: > >> That can easily be done thusly: >> >> (defn do-stuff [ctx c] >> (go (loop [] >> (let [v (<! c)] >> (swap! ctx process v) >> (recur))) >> (go (loop [] >> (let [v (<! c)] >> (swap! ctx process v) >> (recur))) >> (go (loop [] >> (let [v (<! c)] >> (swap! ctx process v) >> (recur)))))) >> >> What am I missing here? Every time I've thought I needed binding, I've >> been able to find a cleaner, more functionally pure way of doing it that >> didn't require binding. >> >> Timothy >> >> >> On Thu, Sep 5, 2013 at 9:01 AM, David Nolen <dnolen.li...@gmail.com>wrote: >> >>> I actually disagree a here as core.async brings a pretty nice >>> concurrency model into play - I suspect there are instances where you might >>> want to construct a series of go blocks with some shared context that you'd >>> rather not put into every go loop. >>> >>> In anycase improved binding support is something I'd like to see in >>> ClojureScript and I don't see the downside of keeping the core.async >>> behavior the same between Clojure and ClojureScript. >>> >>> >>> On Thu, Sep 5, 2013 at 11:30 AM, Timothy Baldridge <tbaldri...@gmail.com >>> > wrote: >>> >>>> As David said, binding preservation works with Clojure, not with >>>> ClojureScript. That being said, while I can see a use for bindings in >>>> Clojure (thread-local redef), I fail to see a use-case for binding in >>>> ClojureScript. Seems like code could/should be structured in a better way. >>>> >>>> >>>> On Thu, Sep 5, 2013 at 7:54 AM, Jacek Lach <jacek.k.l...@googlemail.com >>>> > wrote: >>>> >>>>> That is not what happens in lein repl: >>>>> >>>>> user=> (use 'clojure.core.async) >>>>> nil >>>>> user=> (def pingc (chan)) >>>>> #'user/pingc >>>>> user=> (def ^:dynamic *text* "OUPS BAD ASYNC") >>>>> #'user/*text* >>>>> user=> (binding [*text* "good boy"] >>>>> #_=> (go (while true >>>>> #_=> (<! pingc) >>>>> #_=> (println *text*)))) >>>>> #<ManyToManyChannel >>>>> clojure.core.async.impl.channels.ManyToManyChannel@79d236c2> >>>>> user=> (defn ping [] (put! pingc :ping)) >>>>> #'user/ping >>>>> user=> (ping) >>>>> good boy >>>>> nil >>>>> >>>>> It's of course up to core.async to preserve bindings within its body. >>>>> But since it tries to pretend to run your `go` code 'as if it was >>>>> synchronous', that's probably desirable? >>>>> >>>>> >>>>> On Thursday, September 5, 2013 3:48:10 PM UTC+1, Meikel Brandmeyer >>>>> (kotarak) wrote: >>>>>> >>>>>> Hi, >>>>>> >>>>>> Am Donnerstag, 5. September 2013 16:36:37 UTC+2 schrieb Mathias >>>>>> Picker: >>>>>>> >>>>>>> I just tried this test case (from Anderkent on IRC): >>>>>>> >>>>>>> (def pingc (chan)) >>>>>>> >>>>>>> (def ^:dynamic *text* "OUPS BAD ASYNC") >>>>>>> >>>>>>> (binding [*text* "good boy"] >>>>>>> (go (while true >>>>>>> (<! pingc) >>>>>>> (js/alert *text*)))) >>>>>>> >>>>>>> (defn ping [] >>>>>>> (put! pingc :ping)) >>>>>>> >>>>>>> (ping) >>>>>>> >>>>>>> and I get "OUPS BAD ASYNC". >>>>>>> >>>>>>> So, is this bug in cljs core.async? >>>>>>> >>>>>>> >>>>>> This is to be expected, because the binding is reverted after the go >>>>>> call returns. So the code in the go block (executed later on) sees the >>>>>> old, >>>>>> initial binding. >>>>>> >>>>>> Meikel >>>>>> >>>>>> >>>>> -- >>>>> -- >>>>> 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/groups/opt_out. >>>>> >>>> >>>> >>>> >>>> -- >>>> “One of the main causes of the fall of the Roman Empire was >>>> that–lacking zero–they had no way to indicate successful termination of >>>> their C programs.” >>>> (Robert Firth) >>>> >>>> -- >>>> -- >>>> 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/groups/opt_out. >>>> >>> >>> -- >>> -- >>> 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/groups/opt_out. >>> >> >> >> >> -- >> “One of the main causes of the fall of the Roman Empire was that–lacking >> zero–they had no way to indicate successful termination of their C >> programs.” >> (Robert Firth) >> >> -- >> -- >> 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/groups/opt_out. >> > > -- > -- > 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/groups/opt_out. > -- “One of the main causes of the fall of the Roman Empire was that–lacking zero–they had no way to indicate successful termination of their C programs.” (Robert Firth) -- -- 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/groups/opt_out.
