This is the mail archive of the
gcc-prs@gcc.gnu.org
mailing list for the GCC project.
Re: libstdc++/3247
- To: gdr at gcc dot gnu dot org
- Subject: Re: libstdc++/3247
- From: Benjamin Kosnik <bkoz at nabi dot net>
- Date: 27 Jun 2001 01:06:00 -0000
- Cc: gcc-prs at gcc dot gnu dot org,
- Reply-To: Benjamin Kosnik <bkoz at nabi dot net>
The following reply was made to PR libstdc++/3247; it has been noted by GNATS.
From: Benjamin Kosnik <bkoz@nabi.net>
To: Peter Schmid <schmid@snake.iap.physik.tu-darmstadt.de>
Cc: <bkoz@gnu.org>, <gcc-gnats@gcc.gnu.org>, <gdr@codesourcery.com>
Subject: Re: libstdc++/3247
Date: Tue, 26 Jun 2001 20:03:12 -0500 (CDT)
> This is a known (IMHO mis-) "feature" of the slice or
> gslice interface of the ISO standard. This is because the C++ standard
> library does not specify that valarray subsets provide the same
> operations as valarrays. Therefore, some ugly casts to the valarray
> type are needed. Please confer page 554 of "The C++ Standard Library"
> by Josuttis. For example, the following code does indeed compile. To get a
> working program, you have to dimension the (val)arrays, though.
Hmm. Thanks for the info. I think Gaby was trying to explain this to me
before, but this is pretty clear.
Gaby, if this stuff is indeed incorrect code, then the bug reports that
I've filed on this can be closed.
I suppose fixing the standard so that this stuff is "fixed" will
eventually get done..
thanks,
benjamin