Re: [boost] uBLAS formal review

From: Joerg Walter (jhr.walter_at_[hidden])
Date: 2002-06-28 00:00:56

----- Original Message -----
From: "Fernando Cacciola" <fcacciola_at_[hidden]>
To: <boost_at_[hidden]>
Sent: Friday, June 28, 2002 12:44 AM
Subject: Re: [boost] uBLAS formal review

> > [snipped]
> >
> > > The RingWithIdentity concept is a model of AdditiveAbelianGroup, but
it
> is
> > > also a model of MultiplicativeAbelianGroup even though this isn't not
> > > reflected by the concept constraints().
> >
> > Not exactly, some problems with the inverse.
>
> I see. But then, is MultiplicativeAbelianGroup ever used?

No.

> > [snipped]
> >
>  > I notice that some headers and some jamfiles contain explicit
directives
> > to
> > > supress warnings. IMHO, this is kind of cheating, actually. I would
> prefer
> > > if warnings were removed with appropriate coding instead, whenever
> > possible
> > > at least.
> >
> > The disabled compiler warnings usually are really dumb (like BCC telling
> us
> > 'can't inline this function because it uses ...'). I'm rather sure,
we're
> > not cheating here.
> >
> > > Tests:
> > >
> > > I tried to compile test1 with bcc5.5, but I get this error while
parsing
> > > test12.cpp:
> > >
> > > matrix_et(559): E2034, Cannot convert 'const std::complex<float>' to
> > 'int'.
> >
> > Sorry for breaking the Borland build. I've just rechecked that. Please
> > change the offending lines from something like
> >
> >     return const_iterator1 (*this, it1, it2, it2 != it2_end ? *it2 : 0);
> >
> > to something like
> >
> >     return const_iterator1 (*this, it1, it2, it2 != it2_end ? *it2 :
> > value_type ());
> >
> Great! I'll do that.
>
> BTW: I'm not entirely sure that value_type() initializes to '0' for
built-in
> types. I think this is specified by the std but I seem to recall that some
> compilers don't get it right. It might be safer to use value_type(0)

It's one of the few things we currently trust in ;-), that value_type()
works on every platform. If we'd get a bug report, we'd change.

Regards

Joerg