comp.lang.ada
 help / color / mirror / Atom feed
From: aek@vib.usr.pu.ru (Alexander Kopilovitch)
Subject: Re: "Ravenscar-like" profile for C/C++
Date: 14 May 2004 19:27:53 -0700
Date: 2004-05-14T19:27:53-07:00	[thread overview]
Message-ID: <e2e5731a.0405141827.7cc723fc@posting.google.com> (raw)
In-Reply-To: 2fvai0F2fcp5U1@uni-berlin.de

Peter Amey wrote:

> > if they had plans for a C++ version and they told they had spent
> > plenty time thinking about how one could achieve something
> > worthwhile for C++ and similar to SPARK but they just couldn't
> > think of anything!
>
> Your memory serves you well!  I don't think things have changed that 
> much.  My personal view is that any such subset, if it did exist, would 
> be so restrictive and unnatural to typical C++ users that they would 
> find it unacceptable.  Persuading a potential Ada user of the merits of 
> SPARK is a much easier proposition because they have already taken 
> several steps down the early error detection route.

How about SPARK-classes for C++? I mean regular C++ classes, but with attribute
SPARK (in GCC you can relatively easily define such additional attributes for
classes, it will look something like __SPARK__), which tells that the class
conforms with SPARK-imposed restrictions. In other words, SPARK in C++ can be
applied for individual classes, which can be mixed in a program (and even in
an individual source file) with other (non-SPARK) classes.

Perhaps many C++ programmers will find this approach acceptable, they may
perceive it as reasonable and useful compromise. (And there will be nothing
heretic in that - even in SPARK applications for Ada it may happen that some
packages are for SPARK examination, while others bypass SPARK for some reasons).



Alexander Kopilovich                      aek@vib.usr.pu.ru
Saint-Petersburg
Russia



  parent reply	other threads:[~2004-05-15  2:27 UTC|newest]

Thread overview: 16+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2004-04-25 13:23 "Ravenscar-like" profile for C/C++ Marc Le Roy
2004-04-25 19:43 ` Marc Le Roy
2004-04-25 20:30 ` Jack Klein
     [not found] ` <c6gked$1ha4$1@ulysses.noc.ntua.gr>
2004-04-25 20:31   ` Jack Klein
2004-04-26  1:14     ` Ioannis Vranos
2004-04-26  5:48       ` Martin Krischik
2004-04-26 11:06         ` Michiel Salters
2004-04-26 11:08           ` Vinzent 'Gadget' Hoefler
2004-04-26 11:13             ` Vinzent 'Gadget' Hoefler
     [not found]             ` <fcaee77e.0405050140.6d3a5b7b@posting.google.com>
     [not found]               ` <p8ih90tob4d617h6tjev9d0jmj20h716lu@jellix.jlfencey.com>
2004-05-05 17:44                 ` Martin Dowie
2004-05-06 17:22                   ` Peter Amey
2004-05-06 21:06                     ` Martin Dowie
2004-05-15  2:27                     ` Alexander Kopilovitch [this message]
     [not found] ` <c6gkip$1hhv$1@ulysses.noc.ntua.gr>
     [not found]   ` <408c0ce4$0$15674$626a14ce@news.free.fr>
2004-04-25 20:37     ` Jack Klein
2004-04-26  5:40       ` Martin Krischik
2004-05-05  6:22       ` Craig Carey
replies disabled

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox