From j.l.schonfelder@liverpool.ac.uk Fri May 28 18:38:30 1999 Received: from mailhub1.liv.ac.uk (mailhub1.liv.ac.uk [138.253.100.94]) by dkuug.dk (8.9.2/8.9.2) with ESMTP id SAA32378 for ; Fri, 28 May 1999 18:38:29 +0200 (CEST) (envelope-from j.l.schonfelder@liverpool.ac.uk) Received: from pcmail2.liv.ac.uk ([138.253.252.15]) by mailhub1.liv.ac.uk with esmtp (Exim 1.92 #1) for sc22wg5@dkuug.dk id 10nPeW-0004a1-00; Fri, 28 May 1999 17:38:36 +0100 Received: from [138.253.102.118] (helo=pc102118.liv.ac.uk) by pcmail2.liv.ac.uk with smtp (Exim 1.92 #1) for sc22wg5@dkuug.dk id 10nPeW-0003Gp-00; Fri, 28 May 1999 17:38:36 +0100 From: Lawrie Schonfelder Reply-To: j.l.schonfelder@liverpool.ac.uk To: "WG5 members (list)" Subject: F95, String Module and problems due to Language deficiencies In-Reply-To: <199905281150.NAA31612@dkuug.dk> Message-ID: Date: Fri, 28 May 1999 17:38:34 +0100 (British Summer Time) Priority: NORMAL X-Mailer: Simeon for Win32 Version 4.1.5 Build (43) X-Authentication: IMSP MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Just a comment prompted to some extent by the German comments on the string module ballot, with which I almost totally agree. We created a partial data-abstraction facility in F90. F95 has improved it but only marginally. We still have too many frequently gratuitous restrictions that actually preclude any realsistic data-abstraction becoming what I term a "semantic extension". By this I mean the ability for the user to define a data type and operations on it so that to all intents and purposes the user of the facility can treat the new datatype as if it were defined intrinsically. F95 falls fundamentally short of this in numerous ways. One that I keep falling over is the inability for the user of a datatype containing pointer components to declare an object of this datatype that is constant (PARAMETER). In the case of the string type this is not too serious in practice since a PARAMETER can be declared as a CHARACTER and the complete set of overloads allows this to be combined with strings in most contexts, but it is nevertheless a serious irregularity in general. For a datatype like a variable precision number this becomes a real pain. Unless I am wrong, given a type like type NUMBER INTEGER :: exp INTEGER, POINTER :: sig(:) ENDTYPE NUMBER there is no way of declaring something that would be equivalent to type(NUMBER),PARAMETER :: zero = 0 no matter what circumlocutions one goes through. -- Lawrie Schonfelder Director, Computing Services Dept. The University of Liverpool, UK, L69 7ZF Phone: 44(151)794 3716, Fax: 44(151)794 3759