From: "Lukasz Stafiniak" <lukstafi@gmail.com>
To: "caml users" <caml-list@inria.fr>
Subject: [Caml-list] Instruction selection in the OCaml compiler: Modulesor classes?
Date: Sat, 24 Feb 2007 14:33:12 +0100 [thread overview]
Message-ID: <4a708d20702240533r573a1d1w218d5ae4bde8c814@mail.gmail.com> (raw)
In-Reply-To: <4a708d20702240518l2c430b06r18fe64cabe5cbe9@mail.gmail.com>
[I'm sorry, I forgot to send to the list!]
On 2/24/07, David Baelde <david.baelde@gmail.com> wrote:
> On 2/24/07, skaller <skaller@users.sourceforge.net> wrote:
> > It seems like a module functor allows both anonymous
> > signatures (structural) and also anonymous argument
> > modules (structural), yet you cannot have
> > anonymous functor applications: you have to bind the application to
> > a module name. If we really had structural typing that name would
> > simply be an alias. Why can't we eliminate that name? ***
>
> One worthy remark here might be that side-effects can occur at functor
> instantiation. Then, forcing the user to think about when the
> instantiation occurs is useful...
>
Yet OCaml already has trouble with some nasty patterns of effects at
functor instantiation.
Taken from one of the slides at http://www.cs.cmu.edu/~rwh/courses/modules/
This is good (Skaller: does it solve your problem?):
module type DICT = functor (Key : Set.OrderedType) -> sig
type 'a dict
val domain : 'a dict -> Set.Make(Key).t
end;;
module type PRIO_QUEUE = functor (Elt : Set.OrderedType) -> sig
type queue
val contents : queue -> Set.Make(Elt).t
end;;
module QueuedDict (Data : Set.OrderedType)
(MyDict : DICT) (MyPrioQueue : PRIO_QUEUE) =
struct
module Dict = MyDict (Data)
module PQ = MyPrioQueue (Data)
module DataSet = Set.Make (Data)
let queued_domain dict pq =
DataSet.equal (Dict.domain dict) (PQ.contents pq)
end;;
This is bad:
let moonFull = let v = ref false in fun () -> v := not !v; !v
;;
module type S = sig
type t
val x : t
val f : t -> t
val g : t -> t -> bool
end;;
module Weird (Top : sig end) = (struct
type t = int
let x = if moonFull () then 1 else 2
let f x = x + 2
let g x y = (3*x + 2*y) / (x - y + 1) = 7
end : S);;
module Gen (X : functor (Top : sig end) -> S) = X (struct end);;
module W1 = Gen (Weird);; (* moon full now *)
module W2 = Gen (Weird);; (* moon no longer full now *)
W1.g W1.x W2.x;;
next prev parent reply other threads:[~2007-02-24 13:33 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2007-02-22 14:38 Instruction selection in the OCaml compiler: Modules or classes? Joel Reymont
2007-02-22 17:21 ` [Caml-list] " Tom
2007-02-22 17:38 ` Xavier Leroy
2007-02-22 19:55 ` Chris King
2007-02-22 19:59 ` Markus Mottl
2007-02-23 16:13 ` brogoff
2007-02-23 18:14 ` Tom
2007-02-23 19:28 ` [Caml-list] Instruction selection in the OCaml compiler: Modulesor classes? Andreas Rossberg
2007-02-24 2:51 ` skaller
2007-02-24 11:48 ` David Baelde
[not found] ` <4a708d20702240518l2c430b06r18fe64cabe5cbe9@mail.gmail.com>
2007-02-24 13:33 ` Lukasz Stafiniak [this message]
2007-02-24 14:58 ` [Caml-list] Instruction selection in the OCaml compiler:Modulesor classes? Andreas Rossberg
2007-02-24 17:39 ` skaller
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=4a708d20702240533r573a1d1w218d5ae4bde8c814@mail.gmail.com \
--to=lukstafi@gmail.com \
--cc=caml-list@inria.fr \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox