From: Daryle Walker (darylew_at_[hidden])
Date: 2001-07-04 13:30:17
on 7/3/01 2:31 PM, Helmut at helmut.zeisel_at_[hidden] wrote:
> Sorry for posting several times - it seems that previous trials did not work.
> --- In boost_at_y..., darylew_at_m... wrote:
>> --- In boost_at_y..., <boost_at_y...> wrote:
>>> You can access this file at the URL
> I tried to compile the file under Linux, GCC 2.95:
> In file included from bigbitint_test.cpp:16:
> ../../../boost/math/bigbitint.hpp:13: ios: Datei oder Verzeichnis nicht
> ../../../boost/math/bigbitint.hpp:14: istream: Datei oder Verzeichnis nicht
> ../../../boost/math/bigbitint.hpp:15: ostream: Datei oder Verzeichnis nicht
> (In English: "File or folder not found")
I think the problem is that GCC 2.x is Standard in a lot of things, except
the I/O sublibrary. It uses the pre-standard everything-in-<iostream>,
only-char-and-wchar_t model. If you can't switch to GCC 3.x, we would have
to add a bit of #conditionalizing.
> and under NT / VC++ 6.0:
> c:\temp\bignum\boost\math\bigbitint.hpp(841) : error C2258: illegal pure
> syntax, must be '= 0'
> c:\temp\bignum\boost\math\bigbitint.hpp(841) : error C2252: 'is_specialized' :
> pure specifier can only be specified for functions
> c:\temp\bignum\boost\math\bigbitint.hpp(846) : error C2252: 'digits' : pure
> specifier can only be specified for functions
I guess this compiler doesn't support inline class-static integral
constants. Maybe I could use BOOST_STATIC_CONSTANT or #conditionalize the
specializations of std::numeric_limits for only when the constant form is
supported. (I have to do similar actions when I make a numeric class
template and try to partially specialize std::numeric_limits.)
> As I understand, your program is right and the compilers are wrong.
> Nevertheless, would it be possible for you to convert the code so that it can
> be compiled using these compilers?
-- Daryle Walker Mac, Internet, and Video Game Junkie darylew AT mac DOT com
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk