Re: code from spec with clojure.spec?
This is a late reply to my own question but maybe it will be helpful to a
future searcher.
http://blog.cognitect.com/blog/2016/10/5/interactive-development-with-clojurespec
has an example that showed me more clearly how to use :fn.
Specifically here's how I rewrote the spec:
(s/fdef mapper
:args (s/cat :t :fake.flowrs-tests/positive-integer?
:b :fake.flowrs-tests/positive-integer?)
:ret map?
:fn (fn [{args :args ret :ret}]
(= (:tacos ret)
(vec (range (:t args
(= (:burritos ret)
(vec (range (:b args))
(s/exercise-fn `mapper) has the same behavior as before -- it exercises the
*implementation* of the function. But (stest/check `mapper) will return
"Specification-based
check failed" etc. when the implementation of mapper doesn't match.
As for the other part of the question: turning the logical test of :fn into
a function that can be run to generate the result, that would require quite
a bit of code rewriting. A partial step is just to extract the :fn part of
the code, and that can be done as follows:
(:fn (apply hash-map (rest (s/form (get (s/registry)
'fake.flowrs/mapper)
;=>
(clojure.core/fn
[{args :args, ret :ret}]
(clojure.core/=
(:tacos ret)
(clojure.core/vec (clojure.core/range (:t args
(clojure.core/=
(:burritos ret)
(clojure.core/vec (clojure.core/range (:b args)
The function so-obtained can be evaluated, but it won't "do" the job of the
mapper function without further code transformation. I'll leave it at that
for now.
--
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.
code from spec with clojure.spec?
As a running example related to the ideas in
http://clojure.org/guides/spec#_spec_ing_functions, please consider the
following function:
(defn mapper [x y]
{:tacos (vec (range x))
:burritos (vec (range y))})
Example input and output:
(mapper 2 4) ;=> {:tacos [0 1], :burritos [0 1 2 3]}
OK, I'll go ahead and write a sensible-seeming spec for this function now,
as follows:
(s/fdef mapper
:args (s/cat :t (s/and integer? #(> % 0))
:b (s/and integer? #(> % 0)))
:ret map?
:fn (s/and #(= (-> % :ret :tacos)
(vec (range (-> % :args :t
#(= (-> % :ret :burritos)
(vec (range (-> % :args :b))
My first question: Is there a way to only write the spec, and have the
function generated automatically? The behavior I have in mind is (after
all) specified in the :fn part of the fdef.
And, another somewhat related question: if I redefine the function to do
nothing, and then exercise it, I get output that doesn't seem to conform to
the :fn condition given above, but no error -- why is that?
(defn mapper [x y])
(s/exercise-fn `mapper)
;=> ([(1 3) nil] [(1 1) nil] [(4 2) nil] [(2 3) nil] [(6 3) nil] [(15 2633)
nil] [(3 1) nil] [(3 4) nil] [(1 1) nil] [(28 4) nil])
Thanks for any help!
--
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.
Re: breaking homoiconicity?
On Saturday, June 20, 2015 at 4:15:30 AM UTC+1, Sean Corfield wrote: (.getTypeName (Class/forName [Ljava.lang.String;)) ;;= java.lang.String[] — that is more readable! Thanks, that's helpful for me. By chance do you know if the class is natively recoverable from the TypeName for Clojure/Java? Class/forName can't roundtrip that string. I could implement a look-up table to translate between the two formats, but that seems rather kludgey. -- 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.
breaking homoiconicity?
This is an interaction with Clojure via CIDER. repl [1] (type (into-array String [Awesome])) [Ljava.lang.String; repl [2] [Ljava.lang.String; RuntimeException EOF while reading, starting at line 1 clojure.lang.Util.runtimeException (Util.java:221) repl [3] (quote [Ljava.lang.String;) CIDER won't let me enter [3] claiming that the input is not complete. What's going on here, apart from things not working? If [Ljava.lang.String; is a valid return value, and can show up in code that I construct programmatically, why is it not valid input? Is this a flaw in CIDER / lein, or are they giving me the right answers here? -- 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.
