From owner-sc22wg5@dkuug.dk Wed Jul 9 16:45:13 2003 Received: (from majordom@localhost) by dkuug.dk (8.12.8p1/8.9.2) id h69EjDqJ052164 for sc22wg5-domo; Wed, 9 Jul 2003 16:45:13 +0200 (CEST) (envelope-from owner-sc22wg5@dkuug.dk) X-Authentication-Warning: ptah.dkuug.dk: majordom set sender to owner-sc22wg5@dkuug.dk using -f Received: from inf.rl.ac.uk (nfs7.inf.rl.ac.uk [130.246.72.7]) by dkuug.dk (8.12.8p1/8.9.2) with ESMTP id h69Ej7Ec052159 for ; Wed, 9 Jul 2003 16:45:09 +0200 (CEST) (envelope-from j.k.reid@rl.ac.uk) Received: from numerical.cc.rl.ac.uk (numerical [130.246.8.23]) by inf.rl.ac.uk (8.11.6+Sun/8.8.8) with ESMTP id h69EgwD02439 for ; Wed, 9 Jul 2003 15:42:58 +0100 (BST) Received: from rl.ac.uk (jkr.cse.rl.ac.uk [130.246.9.202]) by numerical.cc.rl.ac.uk (8.8.8+Sun/8.8.8) with ESMTP id PAA01906 for ; Wed, 9 Jul 2003 15:54:00 +0100 (BST) Message-ID: <3F0C2BEB.6030605@rl.ac.uk> Date: Wed, 09 Jul 2003 15:51:23 +0100 From: John Reid Reply-To: j.k.reid@rl.ac.uk Organization: Rutherford Appleton Laboratory User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.0.1) Gecko/20020823 Netscape/7.0 X-Accept-Language: en-us, en MIME-Version: 1.0 To: WG5 Subject: Re: (SC22WG5.2854) Straw votes to ponder concerning Modules TR Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Sender: owner-sc22wg5@dkuug.dk Precedence: bulk -------- Original Message -------- From: Aleksandar Donev Organization: Princeton University To: WG5 Subject: Re: (SC22WG5.2854) Straw votes to ponder concerning Modules TR Date: Wed, 9 Jul 2003 10:22:52 -0400 User-Agent: KMail/1.5 References: <200307021755.h62HtGgM006819@dkuug.dk> <200307081808.h68I8hxG046526@dkuug.dk> In-Reply-To: <200307081808.h68I8hxG046526@dkuug.dk> MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Content-Disposition: inline Message-Id: <200307091022.52292.adonev@math.princeton.edu> Hello, I stronly disagree with most of the suggestions Bill made, particularly the one about submodule procedures having external linkeage. If one wants that, use BIND(C). Name-space pollution is not something Fortran needs--it is a *modern* programming language and should have long passed the phase when names are global...Same goes for more then one level of nesting. There is a technical reason (implementation complexity/feasibility) why only 1 level of contained procedures is allowed (the contained procedure needs to get the stack of its parent etc.), but I see no such thing for submodules. And I definitely agree that 3 levels are needed for maximal guarding against recompilation cascades (which I am appreciating more and more now that I have 120 modules in my program...). And trust me, there are widely used Fortran compilers which already make a mess of how they handle USE statements and take hours to compile a program with a 10-level USE chain. Sure, they can do the same for submodules, but this is not neccesary or inherent to the standard (which you should also believe me because another compiler takes 2 minutes for the same USE hierarchy). In general, this discussion has gone more then I am willing (or have time to) discuss. Proposing radically new concepts at such a late time is not very productive. The goal I though was to polish the TR and get it going, not to invent a new one... Best, Aleksandar