For anyone looking to do something similar (i.e. using exception control flow for handling request errors & validation), I found the slingshot<https://github.com/scgilardi/slingshot> library invaluable for enhanced try and throw leveraging Clojure. Going with 'native' try/throw was becoming nightmarish very quickly.
Thank you all for your input on this. Julien On Thursday, March 21, 2013 3:27:07 AM UTC-4, Marko Topolnik wrote: > > On Wednesday, March 20, 2013 11:29:55 PM UTC+1, James Reeves wrote: > >> On 20 March 2013 16:41, Marko Topolnik <marko.t...@gmail.com> wrote: >> >>> On Wednesday, March 20, 2013 4:34:32 PM UTC+1, James Reeves wrote: >>> >>>> >>>> If validation happens "all around", that implies there is no one >>>> function that can test whether a value of data is valid for a given data >>>> store. This strikes me as a somewhat shaky foundation for a system. >>>> >>> >>> The idea is that all validation functions share the same contract to >>> call the appropriate *add-failure* function that registers the >>> validation result. >>> >> >> I don't see why that would be necessary. Why not put all the validation >> logic in one place? >> > > Indeed, why not. If the requirements allow it, it is definitely preferred. > > >> >> >>> There may be instances where it makes sense to use exceptions as a >>>> control flow mechanism, but I wonder whether it wouldn't be better to use >>>> something like CPS in those instances. >>>> >>> >>> I can't picture how such a mechanism would work, and what benefit it >>> would bring over the exceptions mechanism. CPS in Clojure means >>> trampolining, which is quite an unwieldy, and I'd say "cheap" tack-on. A >>> validating function would then be supposed to return a common, >>> globally-defined "continuation", in fact just a simple function, that would >>> redirect the flow towards the validation failure-handling case. >>> >> >> I wasn't thinking of validation when I suggested CPS, because I don't see >> a need for any unusual control flow when validating data. >> > > That's enterely use case-dependent. In some project validation was > specified as fail-fast, and it happened in the middle of business logic > because the validation status couldn't be determined without computing some > interim results. Throwing an exception from a function three-four levels > down was the perfect choice in that case. > > -marko > -- -- 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.
