From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <lpw25@cam.ac.uk>
X-Original-To: caml-list@sympa.inria.fr
Delivered-To: caml-list@sympa.inria.fr
Received: from mail3-relais-sop.national.inria.fr (mail3-relais-sop.national.inria.fr [192.134.164.104])
	by sympa.inria.fr (Postfix) with ESMTPS id DFC287F75C
	for <caml-list@sympa.inria.fr>; Sun, 24 Aug 2014 17:33:46 +0200 (CEST)
Received-SPF: None (mail3-smtp-sop.national.inria.fr: no sender
  authenticity information available from domain of
  lpw25@cam.ac.uk) identity=pra; client-ip=131.111.8.150;
  receiver=mail3-smtp-sop.national.inria.fr;
  envelope-from="lpw25@cam.ac.uk"; x-sender="lpw25@cam.ac.uk";
  x-conformance=sidf_compatible
Received-SPF: None (mail3-smtp-sop.national.inria.fr: no sender
  authenticity information available from domain of
  lpw25@cam.ac.uk) identity=mailfrom; client-ip=131.111.8.150;
  receiver=mail3-smtp-sop.national.inria.fr;
  envelope-from="lpw25@cam.ac.uk"; x-sender="lpw25@cam.ac.uk";
  x-conformance=sidf_compatible
Received-SPF: None (mail3-smtp-sop.national.inria.fr: no sender
  authenticity information available from domain of
  postmaster@ppsw-50.csi.cam.ac.uk) identity=helo;
  client-ip=131.111.8.150;
  receiver=mail3-smtp-sop.national.inria.fr;
  envelope-from="lpw25@cam.ac.uk";
  x-sender="postmaster@ppsw-50.csi.cam.ac.uk";
  x-conformance=sidf_compatible
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: AlkGANQE+lODbwiWf2dsb2JhbABZs1QGpG4BgQgWEAEBCwsKCBQphAQBBAEnUgULCyElDwEESYhNCAS9KYRLF4V8iGhoB4RMBZxJmGuDOgEBAQ
X-IPAS-Result: AlkGANQE+lODbwiWf2dsb2JhbABZs1QGpG4BgQgWEAEBCwsKCBQphAQBBAEnUgULCyElDwEESYhNCAS9KYRLF4V8iGhoB4RMBZxJmGuDOgEBAQ
X-IronPort-AV: E=Sophos;i="5.04,390,1406584800"; 
   d="scan'208";a="76383993"
Received: from ppsw-50.csi.cam.ac.uk ([131.111.8.150])
  by mail3-smtp-sop.national.inria.fr with ESMTP/TLS/DHE-RSA-AES256-SHA; 24 Aug 2014 17:33:46 +0200
X-Cam-AntiVirus: no malware found
X-Cam-ScannerInfo: http://www.cam.ac.uk/cs/email/scanner/
Received: from host86-144-7-20.range86-144.btcentralplus.com ([86.144.7.20]:38016 helo=study.localdomain)
	by ppsw-50.csi.cam.ac.uk (smtp.hermes.cam.ac.uk [131.111.8.158]:587)
	with esmtpsa (PLAIN:lpw25) (TLSv1.2:AES128-GCM-SHA256:128)
	id 1XLZnc-0001Ch-sk (Exim 4.82_3-c0e5623)
	(return-path <lpw25@cam.ac.uk>); Sun, 24 Aug 2014 16:33:45 +0100
From: Leo White <lpw25@cam.ac.uk>
To: Christoph =?utf-8?Q?H=C3=B6ger?= <christoph.hoeger@tu-berlin.de>
Cc: caml users <caml-list@inria.fr>
References: <53F49BAE.40901@tu-berlin.de>
X-Face: "XWxb[u_Z\PA_Y?9@|IA!!+jTb(/290-*ea/Un$I0B98.$n%eL.;2w<l)q>*,z]WR#T:,p[
 NBd++M7l]#7zj7!{<NUTKJA{}Ox<uO[wb-s<\6jzp.r]-AE[W%E+ckCQ'2WZS@Zu/PJU$~~oy|>~iw
 $W-"/=|dVjhT[D{4~gE}gK<2`.6fs!;uqqud]vs2N/3^m7{aS1V,
Date: Sun, 24 Aug 2014 16:48:33 +0100
In-Reply-To: <53F49BAE.40901@tu-berlin.de> ("Christoph \=\?utf-8\?Q\?H\=C3\=B6ge\?\=
 \=\?utf-8\?Q\?r\=22's\?\= message of
	"Wed, 20 Aug 2014 14:59:26 +0200")
Message-ID: <87k35x6g8e.fsf@study.localdomain>
User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/24.3 (gnu/linux)
MIME-Version: 1.0
Content-Type: text/plain
Subject: Re: [Caml-list] recursive values of abstract types, aka Batteries
 ParserCo

> I wanted to parse a small lambda-calculus in OCaml and had batteries
> already as a dependency. Batteries comes with ParserCo, a small parser
> combinator library. Since I did not want to complicate things further,
> I decided to stick with it and avoid any parser generator. Using a
> monadic combinator library for a scannerless parser should be a
> relatively simple exercise, right?
>
> Unfortunately, it seems that ParserCo only exposes an abstract type:
>
> type ('a, 'b, 'c) t
>
> which is, of course, a function:
>
> type ('a, 'b, 'c) t = ('a, 'c) Source.t -> ('a, 'b, 'c) result
>
> This prevents me from defining any recursive parsers, since OCaml
> won't let me let-rec a value of an abstract function type. I cannot do
> eta-expansion, because the type is abstract, and I also do not have a
> type constructor at hand.
>
> How does one handle such cases in OCaml? Does this really mean, you
> cannot have recursive values of abstract function types?

Hi Christoph,

I haven't used ParserCo (or even looked at the documentation), but the
'lazy' keyword has a relaxed notion of what can be defined recursively,
so creating a recusive `('a, 'b, 'c) t Lazy.t` and then forcing it may
well solve your problem.

Regards,

Leo