This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
[Bug ada/58128] Problem using NAME (STANDARD_INPUT) in gcc-4.7.2
- From: "ExtraLeveLInSoftware at ntlworld dot com" <gcc-bugzilla at gcc dot gnu dot org>
- To: gcc-bugs at gcc dot gnu dot org
- Date: Thu, 15 Aug 2013 19:13:25 +0000
- Subject: [Bug ada/58128] Problem using NAME (STANDARD_INPUT) in gcc-4.7.2
- Auto-submitted: auto-generated
- References: <bug-58128-4 at http dot gcc dot gnu dot org/bugzilla/>
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58128
--- Comment #4 from Ellis N. Thomas <ExtraLeveLInSoftware at ntlworld dot com> ---
The decision to close this without any action seems unjustified. The problem
is being rejected on the basis that the handling of the exception is faulty in
the older Darwin. However, the fundamental problem is that this exception
should not have been raised by OPEN (as in the cited sections of the Ada RM).
Some fix to the relevant IO software is needed.
The reply also states that what worked for I/O on VMS decades ago is
essentially irrelevant - but regardless of whether OPEN is going to support
another file being opened to Standard Input, if it doesn't it must say so by
raising Name_Error or Use_Error as appropriate.
In fact from the software developer's point of view it is clearly an advantage
to be able to support interactive tools (like interpreters) written in Ada, and
the ability to manipulate Standard Input is part of this. However, this
facility itself is not the issue here.
ENT/15-Aug-2013