From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mail1-relais-roc.national.inria.fr (mail1-relais-roc.national.inria.fr [192.134.164.82]) by walapai.inria.fr (8.13.6/8.13.6) with ESMTP id p04GN95H009687 for ; Tue, 4 Jan 2011 17:23:09 +0100 X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: Aq4AADPYIk2AcIgmmWdsb2JhbACkOhUBAQEBAQgLCgcRJLRohl6FSgSEaIk8 X-IronPort-AV: E=Sophos;i="4.60,273,1291590000"; d="scan'208";a="93785909" Received: from bluebox.cs.princeton.edu ([128.112.136.38]) by mail1-smtp-roc.national.inria.fr with ESMTP/TLS/DHE-RSA-AES256-SHA; 04 Jan 2011 17:23:04 +0100 Received: from [192.168.1.101] (ool-4355b7a4.dyn.optonline.net [67.85.183.164]) (authenticated bits=0) by bluebox.CS.Princeton.EDU (8.13.8/8.13.8) with ESMTP id p04GMtZO023390 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=NOT); Tue, 4 Jan 2011 11:22:58 -0500 Mime-Version: 1.0 (Apple Message framework v1082) Content-Type: text/plain; charset=iso-8859-1 From: Yitzhak Mandelbaum In-Reply-To: Date: Tue, 4 Jan 2011 11:22:55 -0500 Cc: Jean Krivine , caml-list@inria.fr Message-Id: References: <4D23353C.8020803@gmail.com> To: =?iso-8859-1?Q?Daniel_B=FCnzli?= X-Mailer: Apple Mail (2.1082) X-Proofpoint-Virus-Version: vendor=nai engine=5400 definitions=6215 signatures=655433 X-Proofpoint-Spam-Reason: safe Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by walapai.inria.fr id p04GN95H009687 Subject: Re: [Caml-list] Array.make exception and parser For one perspective on this issue, you might want to take a look here: http://lambda-the-ultimate.org/node/4150 A short answer (among many) to your particular question: it often requires a lot off effort just to discover whether " a recursive descent parser with some combinators fits perfectly," and many, many grammars do not fit perfectly. On Jan 4, 2011, at 10:31 AM, Daniel Bünzli wrote: > P.S. > > I don't know if that's your case but so many languages to parse are > LL(k) for some k. I don't really understand why people insist on using > yacc like parser generators where a recursive descent parser with some > combinators fits perfectly, seems nearly as efficient and allow you to > give more precise syntax errors to your users. ----------------------------- Yitzhak Mandelbaum