From mboxrd@z Thu Jan  1 00:00:00 1970
X-Spam-Checker-Version: SpamAssassin 3.4.4 (2020-01-24) on polar.synack.me
X-Spam-Level: 
X-Spam-Status: No, score=-1.3 required=5.0 tests=BAYES_00,INVALID_MSGID
	autolearn=no autolearn_force=no version=3.4.4
X-Google-Language: ENGLISH,ASCII-7-bit
X-Google-Thread: 103376,7b93c6665052aea2
X-Google-Attributes: gid103376,public
From: dewar@merv.cs.nyu.edu (Robert Dewar)
Subject: Re: How to pass access parameters to/from C safely/portably?
Date: 1998/06/06
Message-ID: <dewar.897137805@merv>#1/1
X-Deja-AN: 360120776
References: <3572CB7D.FD6D8084@praxis-cs.co.uk>
 <3577D39C.3CECE1C0@praxis-cs.co.uk>
X-Complaints-To: usenet@news.nyu.edu
X-Trace: news.nyu.edu 897137919 5857 (None) 128.122.140.58
Organization: New York University
Newsgroups: comp.lang.ada
Date: 1998-06-06T00:00:00+00:00
List-Id: <comp.lang.ada>


<<For example, with ObjectAda, I find that applying pragma Export
forces an anonymous access parameter to not have accessibility
checks applied, and the actual parameter can be a simple C
pointer-to-something - this seems to be the obvious implementation,
but it's worth checking...anyone confirm if GNAT does the same?
>>

Yes, of course it does. However, note that access parameters are NOT
an appropriate mapping for pointer types in C calls in general, since
Ada does not permit a null value to be passed as an access parameter.
It may work to suppress access checking (for the call, or for the
procedure), but this is getting quite non-portable. The only "correct"
Ada way of representing a pointer parameter in a C routine is as
a (named) access type on the Ada side. Yes, this is a nuisance, it
may even be considered to be a real oversight in the language design.
It has caused endless trouble in bindings!