The predicate parameter introduces an additional predicate check on a
per-step basis which can be avoided under most circumstances.
If check-auth or check-balance return just their input value where that is
valid, threading has no use here.
They could be modified to only return something if there is an error,
otherwise nil. For clarity, they could then be called auth-error and
balance-error. They could be composed with some-fn and be used in a handler
to serve the error as early as possible.
E. g.
(def my-handler-error (some-fn auth-error balance-error))
(defn my-handler
[request]
(or (my-handler-error request)
;; serve response
))
On Monday, July 21, 2014 6:20:50 PM UTC+2, Max Countryman wrote:
>
> Hi,
>
> Recently I found myself wanting a macro similar to some-> or some->> but
> one where I could specify an arbitrary predicate instead of nil?. For
> example, I have a Ring request map which I wish to pass through a number of
> functions. Any of these functions might return an error response and if so
> I do not want to evaluate any remaining forms—much like how some-> will not
> evaluate after a nil response. To do this, I wrote a macro that is
> essentially some-> but which takes a predicate.
>
> Here’s the macro I ended up with:
>
> (defmacro until-pred->
> "Like some-> but evalutes via -> until or if a predicate is true."
> [pred expr & forms]
> (let [g (gensym)
> pstep (fn [step]
> `(if (~pred ~g)
> ~g
> (-> ~g ~step)))]
> `(let [~g ~expr
> ~@(interleave (repeat g) (map pstep forms))]
> ~g)))
>
> An example of how I might use this:
>
> (defn my-handler
> [request]
> (until-pred-> error-response? request
> check-auth
> check-balance
> …))
>
> That said, I’m wondering if there is a more idiomatic way of passing Ring
> request maps through a series of functions, each of which might be a
> terminal step? I’m also curious if there’s a particular reason some->
> wasn’t implemented in a more general way; perhaps the fact that I can so
> easily write a macro that achieves this myself is a good enough reason?
>
> Thanks,
>
>
> Max
--
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.
