Currying != partial application.
And while currying is useful in curried languages (to convert back from uncurried to curried), and partial application is useful period, the question is whether currying is useful, in general, in an uncurried language.
I don't think it is:
* You usually want to perform partial application, currying is an unnecessary intermediate step.
* Currying conflicts with rich "imperative" parameters e.g. overloading, default parameters, keyword parameters.
* Uncurried languages are usually imperative and effectful, that you have filled all the parameter spots does not mean you want to invoke the function.
Having a curried language is neat, and in curried languages converting back from uncurried to curried functions is tremendously useful. But it's not so for uncurried languages (which is most of them).
> which is why the partial application and pipeline operator proposals are so intertwined.
Again, partial application != currying. And pipeline operators can perform partial application (implicitly or explicitely) on their own. For a flagrant example, just see Clojure:
(->> users
(map :email)
(sendEmail)
(mapError (constantly "Error sending emails"))In particular I don't know what's meant by "curried" and "uncurried" languages
add :: Int -> Int -> Int
This is a function that takes one integer and returns a function which itself takes another integer and then finally returns an integer. Syntax sugar allows you to define it and call it like a single function.An uncurried function in these languages is a function that takes multiple arguments as a tuple.
add :: (Int, Int) -> Int
You don't see a lot of functions defined this way in Haskell because there isn't really any advantage to doing so in a lazy, pure language. It's more common in Standard ML from what I've seen, because SML is strict and impure.Currying is the conversion of an n-ary function into a chain of 1-ary function e.g. `fn(a, b, c) -> d` becomes `fn(a) -> fn(b) -> fn(c) -> d`.
Partial application is what it says on the tin, it only applies k parameters to the function returning an (n-k) parameters function, without actually running the function.
Currying is one method allowing partial application, but not necessarily the only one, or usually the best fit as it tends to not interact well with features like overloading, default parameters, keyword parameters, …
> In particular I don't know what's meant by "curried" and "uncurried" languages
Languages like Haskell, OCaml, or Elm are "curried" by default: when you define an n-ary function it's really just sugar for a chain of 1-ary functions; and the application of multiple arguments is similarly the implicit application of single arguments multiple times e.g.
map : (a -> b) -> List a -> List b
map f l =
case l of
[] -> []
(h::t) -> (f h) :: (map f t)
can be called as map fn [1, 2, 3]
or (map fn) [1, 2, 3]
the semantics are identical, which is not the case in an uncurried language like Javascript or Python: map(fn, [1, 2, 3])
and map(fn)([1, 2, 3])
will behave very differently unless the implementer has taken special steps to handle this case e.g. dynamic behaviour based on an optional second parameter, overloading, etc…What would be the use of this? Just delaying the execution of the function's side-effects?
I'll tip my hand: I'm working on a JS-like (and JS-targeted) language that I want to support partial-application/currying by default. Under the hood everything is represented in a curried way, but syntactically you can both define and invoke/partially-invoke functions with the more familiar syntax (commas). But this is fully superficial right now; it all happens at the parser level. Is there a downside to doing things this way?
For defaults/optionals, given "fn f(x, y, z=10)", if "f(1, 2)" evaluates to "f(1)(2)", does that return a function or the result of "f(1)(2)(10)"? What if you have "fn f(x, y = 10, z)"? All of this is a lot easier if you can evaluate the full argument list at the time of function invocation.
As for keyword arguments, they just don't play well with any kind of enforced ordering.
I wonder if that ambiguity could be resolved using contextual type info? I.e. if one of the two interpretations fits the expected type, use that. Otherwise, type-error.
Maybe that would be confusing, not sure.
I read your post yesterday. Good luck with your language!
To be honest this comments section has me seriously reconsidering the "everything is partially-applicable by default" feature, haha.
This doesn't contradict anything you said, but I think Idris is a language that curries everything and has default parameters.
https://docs.idris-lang.org/en/latest/tutorial/miscellany.ht...
errors = list()
for user in users:
if not sendEmail(user.email)
errors.append("Error sending emails")Because I say "what" do to do without going into details of "how" I feel like I have spent less mental capacity reading this portion of code.
However, it does require your transformations to be simple and composable, shove a giant multi-line lambda in there it yes, the loop will be better.
For what it's worth, your example doesn't map 1-to-1 to the intention of what I wrote. Doing it in a more imperative style, I'd have to write
emails = list()
for user in users:
emails.push(user.email)
res = sendEmail emails
if res.IsError:
Error("Error sending emails")
else:
res
(excuse my pseudo Python) try:
return sendEmail(user.email for user in users)
except IsError:
Error("Error sending emails") # if it does something
I agree FP pipelines are nice, but not everything needs it. users.stream()
.map(User::getEmail)
.map(Email::send)
.map(res -> "Error sending email: " + res)
.collect(toList())https://docs.oracle.com/javase/8/docs/api/java/util/stream/p...
with elixir (the previous example) you'd likely use simple hashmaps or lists like {email: "mail"} or [:error, "reason for failure"].
That doesn't sound like its amazing, but actually is in practice, because the whole language is written with that in mind (pattern matching to effectively do function/method overloading depending on the value of each argument for example)
its very concise while still being very explicit. The same would be possible with java, but you'll need to create a lot of classes/interfaces/enums/boilerplate to facilitate it.
i'd still prefer java any time at a dayjob though, because boring and dumb is generally better if you need to write code that's gonna be in use for decades... and likely going to be changed by a lot of people with varying levels of experience.
No, in most cases you don't. The stream interface is mostly designed around simple Lists, Sets and Maps. It also interacts very well with the Collection framework (e.g. myHashmap.entrySet() yielding a set, etc.) which is part of the standard library.
You can extend streams with custom collectors, but rarely if ever you need to define intermediate data structures. You do need to define initial and terminal structures, but I'd argue that's good practice regardless.
that lets you pass the result from each function to the next without explicitly stating what form they have, yes.
You'll still be missing the conciseness/explicitness because the language isn't meant to be used like that and is missing necessary features in order to facilitate it such as pattern matching by the passed in values into function.
to make a simple example, you could theoretically write the following pseudo-code
def sendEmail({email}) do
// send email
end
def sendEmail({id}) do
// get email address
sendEmail({email})
end
sendEmail({id: 244})
or to make sure your code stops executing if an error occurs (positional return values this time) [:ok, msg] = sendEmail(lkajsdf) .collect(Collectors.groupingBy(...))maybe actually learn to program sometimes, then you will figure out what it does.
An error could be a malformed e-mail address, or a timeout from the e-mail server.
What you do not want is to send 10000 e-mail, then have an error like "Error sending emails", then re-run a few time, only to have some people receive 5 e-mail, and some people receive 0 e-mail.
Currying and partial application are really nice when you have a pipeline operator, which is why the partial application and pipeline operator proposals are so intertwined.