See https://groups.google.com/d/topic/clojure/2dHvX7bf7nA/discussion, http://stackoverflow.com/a/22409846/1756702, where the old and new state of an atom is returned using the lower-level compare-and-set! operation.
On Tuesday, April 8, 2014 10:41:50 AM UTC-5, John Hume wrote: > > I sometimes find that after mutating an atom, I want to create some > side-effect that depends on the old and new state as well as the context in > which the change was made. Because of the dependence on context, a watch > doesn't work (unless there's something I'm not thinking of). So I add > things to the new atom state (returned by swap!) purely to tell the calling > code what side-effect to have (or give it the data it needs to decide what > side-effect to have). That additional state isn't used anywhere other than > the fn that called swap!. > > One gotcha to this approach is that one must be careful not to leave some > old side-effect causing state in place to cause another side-effect based > on stale data. > > Is there a name for this pattern? A standard way of implementing it? A > better alternative? > > One alternative I'm aware of is using mutable locals (provided by > https://github.com/ztellman/proteus) as a side-channel of communication > from swap!. Both approaches strike me as messy, though a let-mutable > probably makes it more obvious that something funny is going on, and it > doesn't pollute the atom. > > Thanks. > -hume. > -- 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.
