From dnagle@erols.com Thu Mar 30 15:13:59 2000 Received: from smtp03.mrf.mail.rcn.net (smtp03.mrf.mail.rcn.net [207.172.4.62]) by dkuug.dk (8.9.2/8.9.2) with ESMTP id PAA26147 for ; Thu, 30 Mar 2000 15:13:58 +0200 (CEST) (envelope-from dnagle@erols.com) Received: from 216-164-139-221.s475.tnt5.lnhva.md.dialup.rcn.com ([216.164.139.221] helo=satellite) by smtp03.mrf.mail.rcn.net with smtp (Exim 2.12 #3) id 12aelo-00026g-00 for sc22wg5@dkuug.dk; Thu, 30 Mar 2000 08:13:57 -0500 From: Dan Nagle Date: Thu, 30 Mar 2000 13:13:29 GMT Message-ID: <20000330.13132972@mis.configured.host> Subject: Re: (SC22WG5.1755) Semantics of the SIZE intrinsic To: sc22/wg5 members In-Reply-To: <200003301231.OAA25937@dkuug.dk> References: <200003301231.OAA25937@dkuug.dk> X-Mailer: Mozilla/3.0 (compatible; StarOffice/5.1; Win32) X-Priority: 3 (Normal) MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by dkuug.dk id PAA26147 Hello, SIZE of a zero sized array is zero. Should SIZE of a disassociated array be zero or negative (an impossible value)? That is, should SIZE distinguish between a pointer to a zero sized array and a disassociated pointer? Agreed that a POINTER or ALLOCATABLE array should be able to be initialized with a zero sized array. >>>>>>>>>>>>>>>>>> Original Message <<<<<<<<<<<<<<<<<< On 3/30/00, 7:31:35 AM, Lawrie Schonfelder wrote regarding (SC22WG5.1755) Semantics of the SIZE intrinsic: > I have just run into a difficulty with the SIZE intrinsic. > As currently defined it is illegal to call it with a disassociated pointer to > an array. However, there is no way that I know of to initialise such a > pointer to anything other the NULL(). This means that safe procedures cannot > be easily written that depend on local working variables whose size is a > function of such a pointer argument. This is a common requirement. Many > compilers very sensibly extend the definition to allow disassociated pointers > to arrays and deallocated allocatable arrays as arguments and return the > obvious zero result. One at least appears in checking mode to abort the > program and in non-checking mode to return some arbitrary value. > Could we not have a minor extension to the standard and define the SIZE > intrinsic to return zero in such cases. > The alternative would be to provide some way of allowing a pointer to be > initialised to have a zero sized array target and likewise an allocatable > array to a zero sized array instead of deallocated. In fact I would like both > but I would settle for the extended SIZE definition. > -- > Lawrie Schonfelder > Director, Computing Services Dept. > The University of Liverpool, UK, L69 7ZF > Phone: 44(151)794 3716, Fax: 44(151)794 3759 -- Cheers! Dan Nagle, eMail: dnagle@erols.com Purple Sage Computing Solutions, Inc., 12142 Purple Sage Ct., Reston VA 20194