How do I programmatically figure out which Vars may affect the results of a function defined in Clojure?
Consider this definition of a Clojure function:
(def ^:dynamic *increment* 3)
(defn f [x]
(+ x *increment*))
This is a function of x, but also of *increment* (and also of clojure.core/+(1); but I'm less concerned with that). When writing tests for this function, I want to make sure that I control all relevant inputs, so I do something like this:
(assert (= (binding [*increment* 3] (f 1)) 4))
(assert (= (binding [*increment* -1] (f 1)) 0))
(Imagine that *increment* is a configuration value that someone might reasonably change; I don't want this function's tests to need changing when this happens.)
My question is: how do I write an assertion that the value of (f 1) can depend on *increment* but not on any other Var? Because I expect that one day someone will refactor some code and cause the function to be
(defn f [x]
(+ x *increment* *additional-increment*))
and neglect to update the test, and I would like to have the test fail even if *additional-increment* is zero.
This is of course a simplified example – in a large system, there can be lots of dynamic Vars, and they can get referenced through a long chain of function calls. The solution needs to work even if f calls g which calls h which references a Var. It would be great if it didn't claim that (with-out-str (prn "foo")) depends on *out*, but this is less important. If the code being analyzed calls eval or uses Java interop, of course all bets are off.
I can think of three categories of solutions:
Get the information from the compiler
I imagine the compiler does scan function definitions for the necessary information, because if I try to refer to a nonexistent Var, it throws:
user=> (defn g [x] (if true x (+ *foobar* x)))
CompilerException java.lang.RuntimeException: Unable to resolve symbol: *foobar* in this context, compiling:(NO_SOURCE_PATH:24)
Note that this happens at compile time, and regardless of whether the offending code will ever be executed. Thus the compiler should know what Vars are potentially referenced by the function, and I would like to have access to that information.
Parse the source code and walk the syntax tree, and record when a Var is referenced
Because code is data and all that. I suppose this means calling macroexpand and handling each Clojure primitive and every kind of syntax they take. This looks so much like a compilation phase that it would be great to be able to call parts of the compiler, or somehow add my own hooks to the compiler.
Instrument the Var mechanism, execute the test and see which Vars get accessed
Not as complete as the other methods (what if a Var is used in a branch of the code that my test fails to exercise?) but this would suffice. I imagine I would need to redefine def to produce something that acts like a Var but records its accesses somehow.
(1) Actually that particular function doesn't change if you rebind +; but in Clojure 1.2 you can bypass that optimization by making it (defn f [x] (+ x 0 *increment*)) and then you can have fun with (binding [+ -] (f 3)). In Clojure 1.3 attempting to rebind + throws an error.
Regarding your first point you could consider using the analyze library. With it you can quite easily figure out which dynamic vars are used in an expression:
user> (def ^:dynamic *increment* 3)
user> (def src '(defn f [x]
(+ x *increment*)))
user> (def env {:ns {:name 'user} :context :eval})
user> (->> (analyze-one env src)
expr-seq
(filter (op= :var))
(map :var)
(filter (comp :dynamic meta))
set)
#{#'user/*increment*}
I know that this doesn't answer your question, but wouldn't it be a lot less work to just provide two versions of a function where one version has no free variables, and the other version calls the first one with the appropriate top-level defines?
For example:
(def ^:dynamic *increment* 3)
(defn f
([x]
(f x *increment*))
([x y]
(+ x y)))
This way you can write all your tests against (f x y), which doesn't rely on any global state.
Related
I find myself writing a lot of clojure in this manner:
(defn my-fun [input]
(let [result1 (some-complicated-procedure input)
result2 (some-other-procedure result1)]
(do-something-with-results result1 result2)))
This let statement seems very... imperative. Which I don't like. In principal, I could be writing the same function like this:
(defn my-fun [input]
(do-something-with-results (some-complicated-procedure input)
(some-other-procedure (some-complicated-procedure input)))))
The problem with this is that it involves recomputation of some-complicated-procedure, which may be arbitrarily expensive. Also you can imagine that some-complicated-procedure is actually a series of nested function calls, and then I either have to write a whole new function, or risk that changes in the first invocation don't get applied to the second:
E.g. this works, but I have to have an extra shallow, top-level function that makes it hard to do a mental stack trace:
(defn some-complicated-procedure [input] (lots (of (nested (operations input)))))
(defn my-fun [input]
(do-something-with-results (some-complicated-procedure input)
(some-other-procedure (some-complicated-procedure input)))))
E.g. this is dangerous because refactoring is hard:
(defn my-fun [input]
(do-something-with-results (lots (of (nested (operations (mistake input))))) ; oops made a change here that wasn't applied to the other nested calls
(some-other-procedure (lots (of (nested (operations input))))))))
Given these tradeoffs, I feel like I don't have any alternatives to writing long, imperative let statements, but when I do, I cant shake the feeling that I'm not writing idiomatic clojure. Is there a way I can address the computation and code cleanliness problems raised above and write idiomatic clojure? Are imperitive-ish let statements idiomatic?
The kind of let statements you describe might remind you of imperative code, but there is nothing imperative about them. Haskell has similar statements for binding names to values within bodies, too.
If your situation really needs a bigger hammer, there are some bigger hammers that you can either use or take for inspiration. The following two libraries offer some kind of binding form (akin to let) with a localized memoization of results, so as to perform only the necessary steps and reuse their results if needed again: Plumatic Plumbing, specifically the Graph part; and Zach Tellman's Manifold, whose let-flow form furthermore orchestrates asynchronous steps to wait for the necessary inputs to become available, and to run in parallel when possible. Even if you decide to maintain your present course, their docs make good reading, and the code of Manifold itself is educational.
I recently had this same question when I looked at this code I wrote
(let [user-symbols (map :symbol states)
duplicates (for [[id freq] (frequencies user-symbols) :when (> freq 1)] id)]
(do-something-with duplicates))
You'll note that map and for are lazy and will not be executed until do-something-with is executed. It's also possible that not all (or even not any) of the states will be mapped or the frequencies calculated. It depends on what do-something-with actually requests of the sequence returned by for. This is very much functional and idiomatic functional programming.
i guess the simplest approach to keep it functional would be to have a pass-through state to accumulate the intermediate results. something like this:
(defn with-state [res-key f state]
(assoc state res-key (f state)))
user> (with-state :res (comp inc :init) {:init 10})
;;=> {:init 10, :res 11}
so you can move on to something like this:
(->> {:init 100}
(with-state :inc'd (comp inc :init))
(with-state :inc-doubled (comp (partial * 2) :inc'd))
(with-state :inc-doubled-squared (comp #(* % %) :inc-doubled))
(with-state :summarized (fn [st] (apply + (vals st)))))
;;=> {:init 100,
;; :inc'd 101,
;; :inc-doubled 202,
;; :inc-doubled-squared 40804,
;; :summarized 41207}
The let form is a perfectly functional construct and can be seen as syntactic sugar for calls to anonymous functions. We can easily write a recursive macro to implement our own version of let:
(defmacro my-let [bindings body]
(if (empty? bindings)
body
`((fn [~(first bindings)]
(my-let ~(rest (rest bindings)) ~body))
~(second bindings))))
Here is an example of calling it:
(my-let [a 3
b (+ a 1)]
(* a b))
;; => 12
And here is a macroexpand-all called on the above expression, that reveal how we implement my-let using anonymous functions:
(clojure.walk/macroexpand-all '(my-let [a 3
b (+ a 1)]
(* a b)))
;; => ((fn* ([a] ((fn* ([b] (* a b))) (+ a 1)))) 3)
Note that the expansion doesn't rely on let and that the bound symbols become parameter names in the anonymous functions.
As others write, let is actually perfectly functional, but at times it can feel imperative. It's better to become fully comfortable with it.
You might, however, want to kick the tires of my little library tl;dr that lets you write code like for example
(compute
(+ a b c)
where
a (f b)
c (+ 100 b))
Suppose I've got a function (remove-bad-nodes g) that returns a sequence like this:
[updated-g bad-nodes]
where updated-g is a graph with its bad nodes removed, and bad-nodes is a collection containing the removed nodes.
As an argument to a function or inside a let, I could destructure it like this:
(let [[g bads] (remove-bad-nodes g)]
...)
but that only defines local variables. How could I do that in the REPL, so that in future commands I can refer to the updated graph as g and the removed nodes as bads? The first thing that comes to mind is this:
(def [g bads] (remove-bad-nodes g)
but that doesn't work, because def needs its first argument to be a Symbol.
Note that I'm not asking why def doesn't have syntax like let; there's already a question about that. I'm wondering what is a convenient, practical way to work in the REPL with functions that return "multiple values". If there's some reason why in normal Clojure practice there's no need to destructure in the REPL, because you do something else instead, explaining that might make a useful answer. I've been running into this a lot lately, which is why I'm asking. Usually, but not always, these functions return an updated version of something along with some other information. In side-effecting code, the function would modify the object and return only one value (the removed nodes, in the example), but obviously that's not the Clojurely way to do it.
I think the way to work with such functions in the repl is just to not def your intermediate results unless they are particularly interesting; for interesting-enough intermediate results it's not a big hassle to either def them to a single name, or to write multiple defs inside a destructuring form.
For example, instead of
(def [x y] (foo))
(def [a b] (bar x y))
you could write
(let [[x y] (foo),
[x' y'] (bar x y)])
(def a x') ; or maybe leave this out if `a` isn't that interesting
(def b y'))
A nice side effect of this is that the code you write while playing around in the repl will look much more similar to the code you will one day add to your source file, where you will surely not be defing things over and over, but rather destructuring them, passing them to functions, and so on. It will be easier to adapt the information you learned at the repl into a real program.
There's nothing unique about destructuring w/r/t the REPL. The answer to your question is essentially the same as this question. I think your options are:
let:
(let [[light burnt just-right] (classify-toasts (make-lots-of-toast))]
(prn light burnt just-right))
def the individual values:
(def result (classify-toasts (make-lots-of-toast)))
(def light (nth result 0))
(def burnt (nth result 1))
(def just-right (nth result 2))
Or write a macro to do that def work for you.
You could also consider a different representation if your function is always returning a 3-tuple/vector e.g. you could alternatively return a map from classify-toasts:
{:light 1, :burnt 2, :just-right 3}
And then when you need one of those values, destructure the map using the keywords wherever you need:
(:light the-map) => 1
Observe:
user=> (def results [1 2 3])
#'user/results
user=> (let [[light burnt just-right] results] (def light light) (def burnt burnt) (def just-right just-right))
#'user/just-right
user=> light
1
user=> burnt
2
user=> just-right
3
Is there a reasonable way to have multiple def statements happen with destructing the same way that let does it? For Example:
(let [[rtgs pcts] (->> (sort-by second row)
(apply map vector))]
.....)
What I want is something like:
(defs [rtgs pcts] (->> (sort-by second row)
(apply map vector)))
This comes up a lot in the REPL, notebooks and when debugging. Seriously feels like a missing feature so I'd like guidance on one of:
This exists already and I'm missing it
This is a bad idea because... (variable capture?, un-idiomatic?, Rich said so?)
It's just un-needed and I must be suffering from withdrawals from an evil language. (same as: don't mess up our language with your macros)
A super short experiment give me something like:
(defmacro def2 [[name1 name2] form]
`(let [[ret1# ret2#] ~form]
(do (def ~name1 ret1#)
(def ~name2 ret2#))))
And this works as in:
(def2 [three five] ((juxt dec inc) 4))
three ;; => 3
five ;; => 5
Of course and "industrial strength" version of that macro might be:
checking that number of names matches the number of inputs. (return from form)
recursive call to handle more names (can I do that in a macro like this?)
While I agree with Josh that you probably shouldn't have this running in production, I don't see any harm in having it as a convenience at the repl (in fact I think I'll copy this into my debug-repl kitchen-sink library).
I enjoy writing macros (although they're usually not needed) so I whipped up an implementation. It accepts any binding form, like in let.
(I wrote this specs-first, but if you're on clojure < 1.9.0-alpha17, you can just remove the spec stuff and it'll work the same.)
(ns macro-fun
(:require
[clojure.spec.alpha :as s]
[clojure.core.specs.alpha :as core-specs]))
(s/fdef syms-in-binding
:args (s/cat :b ::core-specs/binding-form)
:ret (s/coll-of simple-symbol? :kind vector?))
(defn syms-in-binding
"Returns a vector of all symbols in a binding form."
[b]
(letfn [(step [acc coll]
(reduce (fn [acc x]
(cond (coll? x) (step acc x)
(symbol? x) (conj acc x)
:else acc))
acc, coll))]
(if (symbol? b) [b] (step [] b))))
(s/fdef defs
:args (s/cat :binding ::core-specs/binding-form, :body any?))
(defmacro defs
"Like def, but can take a binding form instead of a symbol to
destructure the results of the body.
Doesn't support docstrings or other metadata."
[binding body]
`(let [~binding ~body]
~#(for [sym (syms-in-binding binding)]
`(def ~sym ~sym))))
;; Usage
(defs {:keys [foo bar]} {:foo 42 :bar 36})
foo ;=> 42
bar ;=> 36
(defs [a b [c d]] [1 2 [3 4]])
[a b c d] ;=> [1 2 3 4]
(defs baz 42)
baz ;=> 42
About your REPL-driven development comment:
I don't have any experience with Ipython, but I'll give a brief explanation of my REPL workflow and you can maybe comment about any comparisons/contrasts with Ipython.
I never use my repl like a terminal, inputting a command and waiting for a reply. My editor supports (emacs, but any clojure editor should do) putting the cursor at the end of any s-expression and sending that to the repl, "printing" the result after the cursor.
I usually have a comment block in the file where I start working, just typing whatever and evaluating it. Then, when I'm reasonably happy with a result, I pull it out of the "repl-area" and into the "real-code".
(ns stuff.core)
;; Real code is here.
;; I make sure that this part always basically works,
;; ie. doesn't blow up when I evaluate the whole file
(defn foo-fn [x]
,,,)
(comment
;; Random experiments.
;; I usually delete this when I'm done with a coding session,
;; but I copy some forms into tests.
;; Sometimes I leave it for posterity though,
;; if I think it explains something well.
(def some-data [,,,])
;; Trying out foo-fn, maybe copy this into a test when I'm done.
(foo-fn some-data)
;; Half-finished other stuff.
(defn bar-fn [x] ,,,)
(keys 42) ; I wonder what happens if...
)
You can see an example of this in the clojure core source code.
The number of defs that any piece of clojure will have will vary per project, but I'd say that in general, defs are not often the result of some computation, let alone the result of a computation that needs to be destructured. More often defs are the starting point for some later computation that will depend on this value.
Usually functions are better for computing a value; and if the computation is expensive, then you can memoize the function. If you feel you really need this functionality, then by all means, use your macro -- that's one of the sellings points of clojure, namely, extensibility! But in general, if you feel you need this construct, consider the possibility that you're relying too much on global state.
Just to give some real examples, I just referenced my main project at work, which is probably 2K-3K lines of clojure, in about 20 namespaces. We have about 20 defs, most of which are marked private and among them, none are actually computing anything. We have things like:
(def path-prefix "/some-path")
(def zk-conn (atom nil))
(def success? #{200})
(def compile* (clojure.core.memoize/ttl compiler {} ...)))
(def ^:private nashorn-factory (NashornScriptEngineFactory.))
(def ^:private read-json (comp json/read-str ... ))
Defining functions (using comp and memoize), enumerations, state via atom -- but no real computation.
So I'd say, based on your bullet points above, this falls somewhere between 2 and 3: it's definitely not a common use case that's needed (you're the first person I've ever heard who wants this, so it's uncommon to me anyway); and the reason it's uncommon is because of what I said above, i.e., it may be a code smell that indicates reliance on too much global state, and hence, would not be very idiomatic.
One litmus test I have for much of my code is: if I pull this function out of this namespace and paste it into another, does it still work? Removing dependencies on external vars allows for easier testing and more modular code. Sometimes we need it though, so see what your requirements are and proceed accordingly. Best of luck!
Occasionally when looking at other people's Clojure code, I see a function defined via defn and then called using the var-quote syntax, e.g.:
user> (defn a [] 1)
#'user/a
user> (a) ; This is how you normally call a function
1
user> (#'a) ; This uses the var-quote syntax and produces the same result
1
For the life of me I can't figure out the difference between these two ways of calling a function. I can't find anything in the evaluation documentation to say what happens when the operator of a call is a var that might suggest why the second form would be preferred. They both seem to respond in the same to binding assignments and syntax-quoting.
So, can somebody please provide a code sample that will illustrate the difference between (a) and (#'a) above?
Edit: I know that var-quote can be used to get to a var that's shadowed by a let lexical binding, but that doesn't seem to be the case in the code that I'm looking at.
(#'a) always refers to the var a, while (a) can be shadowed by local bindings:
user> (defn a [] 1)
#'user/a
user> (let [a (fn [] "booh")] [(a) (#'a)])
["booh" 1]
But most actual uses of var-quote / function call are not calling the var-quote expression directly, but instead cache its value so that higher-order constructs refer to the current value of var a instead of its value when passed in:
(defn a [] 1)
(defn my-call [f] (fn [] (+ 1 (f))))
(def one (my-call a))
(def two (my-call #'a))
(defn a [] 2)
user> (one)
2
user> (two)
3
This is mostly useful for interactive development, where you're changing some function that gets wrapped in a bunch of other functions in other packages.
The second form allows you to circumvent the privacy restrictions that clojure puts in place.
So, for instance, if you develop a library with private functions, but want to test them from a separate namespace, you cannot refer to them directly. But you can get to them using the var quote syntax. It's very useful for this.
Privacy is clojure is, in essence, a form of automatic documentation, as opposed to the privacy you see in Java. You can get around it.
user> (defn- a [] 1)
#'user/a
user> (ns user2)
nil
user2> (user/a)
CompilerException java.lang.IllegalStateException: var: #'user/a is not public, compiling:(NO_SOURCE_PATH:1)
user2> (#'user/a)
1
user2>
When using ClojureScript I tried to define a function that is a closure over a variable like this:
(let [x 42]
(defn foo [n] (+ x n)))
That prints the following source at the Rhino REPL:
function foo(n){
return cljs.core._PLUS_.call(null,x__43,n);
}
The function works as I expect but when trying to get at the variable named x__43 I can't get it. Where did it go?
the x variable is defined outside the foo function, in the let binding. you can't "get it" because you're not in the scope of the let binding. that's more or less the whole point of using closures.
conceptually, let bindings are implemented as function calls:
(let [x 2] ...)
is equivalent to
((fn [x] ...) 2)
which is probably similar to let is implemented in ClojureScript - either as a macro transformation to fn or directly to (function(x){...})(2).
(let [x 42]
(defn foo [n] (+ x n)))
is currently compiled to
var x__1311 = 42;
cljs.user.foo = (function foo(n){
return (x__1311 + n);
});
The exact number attached to the x may of course vary from compilation to compilation and cljs.user would be replaced by the appropriate namespace name.
There is no attempt to conceal the generated variable from unrelated code in a JavaScript closure, so in principle it could still be modified if one went out of one's way to do so. Accidental collisions are extremely unlikely and simply will not happen with regular ClojureScript.
To discover things like the above, you can either call the compiler with {:optimizations :simple :pretty-print true} among the options or ask it to emit some JavaScript at the REPL (as provided by script/repl in the ClojureScript source tree or lein repl in a Leiningen project with ClojureScript declared as a dependency):
(require '[cljs.compiler :as comp])
(binding [comp/*cljs-ns* 'cljs.user]
(comp/emit
(comp/analyze {:ns {:name 'cljs.user} :context :statement :locals {}}
'(let [x 42] (defn foo [n] (+ x n))))))