Mailing list for all users of the OCaml language and system.
 help / color / mirror / Atom feed
From: Patrick M Doane <patrick@watson.org>
To: Pierre Weis <Pierre.Weis@inria.fr>
Cc: Markus Mottl <mottl@miss.wu-wien.ac.at>, caml-list@inria.fr
Subject: Re: Typing of patterns
Date: Mon, 5 Jun 2000 21:10:45 -0400 (EDT)	[thread overview]
Message-ID: <Pine.BSF.3.96.1000605204225.15847A-100000@fledge.watson.org> (raw)
In-Reply-To: <200006051356.PAA17512@pauillac.inria.fr>

On Mon, 5 Jun 2000, Pierre Weis wrote:

> However, we get a strange difference in typing since, as you
> mentioned, the explicit ``as clause'' does not set up a typing
> connection between input and output :
> 
> let (ensure_nil : 'a pliste -> 'b pliste) = function
>   | `Cons (x, l) -> failwith "Not nil"
>   | `Nil as l -> l;;
> val ensure_nil : 'a pliste -> 'b pliste = <fun>
> 
> Jacques may explain us if the above suggested generalization scheme is
> used for identifiers bound in as clauses of patterns (and if not,
> which scheme is used ?)...

This discussion seems related to a recent typing difficulty I have had
with polymorphic variants.

I want to change the data associated to a particular constructor of a
variant and leave the others unmodified:

     # let f = function `A -> `A () | _ as t -> t;;
     Characters 41-42:
     This expression has type [< `A | ..] but is here used with type
       [> `A of unit]

Notice that the type for '_' includes `A when it obviously cannot. One
could include every possible expected type in place of '_' but that
removes one of the advantages of polymorphic variants: that the type can 
be refined and extended easily.


While further exploring this issue I managed to get an uncaught exception
from the type checker:

     # let f = function `A true -> () | `A 5 -> ();; 
     Uncaught exception: File "typing/parmmatch.ml", ...

Taking the conjuction of 'unit' and 'int' does seem to work however,

     # let f = function `A -> () | `A 5 -> ();;
     val f : [< `A of & int] -> unit = <fun>

but it prints the type without mentioning 'unit'.

Patrick Doane




  parent reply	other threads:[~2000-06-06  7:33 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2000-06-01 13:23 Markus Mottl
2000-06-05 13:56 ` Pierre Weis
2000-06-05 15:29   ` Markus Mottl
2000-06-06  0:55   ` Jacques Garrigue
2000-06-06 15:32     ` Pierre Weis
2000-06-06  1:10   ` Patrick M Doane [this message]
2000-06-06  8:55     ` Jacques Garrigue
2000-06-06  7:24 Pierre Weis

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=Pine.BSF.3.96.1000605204225.15847A-100000@fledge.watson.org \
    --to=patrick@watson.org \
    --cc=Pierre.Weis@inria.fr \
    --cc=caml-list@inria.fr \
    --cc=mottl@miss.wu-wien.ac.at \
    /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