On May 4, 9:52 pm, cppljev...@gmail.com wrote:
>    template<class T, T... C>
>    struct SEQ_c
>    : SEQ<integral_c<T,C>...>
>    {};
>
> ?  I tried compiling something like this; however, I got
> a compile error:
>
>    error: parameter packs not expanded with '...':

Shouldn't it actually be "integral_c<T, C...>" (i.e. expand the C,
since it's a pack)?


--
[ comp.std.c++ is moderated.  To submit articles, try just posting with ]
[ your news-reader.  If that fails, use mailto:std-c++@netlab.cs.rpi.edu]
[              --- Please see the FAQ before posting. ---               ]
[ FAQ: http://www.comeaucomputing.com/csc/faq.html                      ]

On May 5, 3:09 pm, Pavel Minaev <int...@gmail.com> wrote:
> On May 4, 9:52 pm, cppljev...@gmail.com wrote:
>
> >    template<class T, T... C>
> >    struct SEQ_c
> >    : SEQ<integral_c<T,C>...>
> >    {};
>
> > ?  I tried compiling something like this; however, I got
> > a compile error:
>
> >    error: parameter packs not expanded with '...':
>
> Shouldn't it actually be "integral_c<T, C...>" (i.e. expand the C,
> since it's a pack)?

The integral-sequence-wrapper.html is supposed to convey the
idea that SEQ_C is composed of a sequence composed of:elements:

    integral_c<T,c1>
    integral_c<T,c2>
    ...
    integral_c<T,cN>

So, no I didn't mean for the expansion to be integral_c<T,C...>.
I was hoping that, since T was not a package, the compiler
would treat it like some constant type, e.g. int, instead of
like a template parameter.  The list_c_expansion_wish.cpp
was supposed to make it clearer, but I guess I need to
work a little more on my communication skills.

Sorry for the confusion.

-regards,
Larry



--
[ comp.std.c++ is moderated.  To submit articles, try just posting with ]
[ your news-reader.  If that fails, use mailto:std-c++@netlab.cs.rpi.edu]
[              --- Please see the FAQ before posting. ---               ]
[ FAQ: http://www.comeaucomputing.com/csc/faq.html                      ]

On May 5, 3:33 pm, cppljev...@gmail.com wrote:
> On May 5, 3:09 pm, Pavel Minaev <int...@gmail.com> wrote:
>
> > On May 4, 9:52 pm, cppljev...@gmail.com wrote:
>
> > >    template<class T, T... C>
> > >    struct SEQ_c
> > >    : SEQ<integral_c<T,C>...>
> > >    {};
[snip]
> > Shouldn't it actually be "integral_c<T, C...>" (i.e. expand the C,
> > since it's a pack)?
[snip]

Probably what you're really missing is the concept of "unpacking
patterns"
as described in section 2.4 of:

   http://www.generic-programming.org/~dgregor/cpp/variadic-templates.pdf

There a pattern, AFAICT, can be some expression, E, which contains
a "pack name".  For example, the section 2.4 contains:

   Args&...

The pack name is Args but the expression is Args&.  In the integral_c
case, the pack name is C but the expression is integral_c<T,C>.

Maybe "pack name" is the wrong term, but hopefully that makes
things clearer.


--
[ comp.std.c++ is moderated.  To submit articles, try just posting with ]
[ your news-reader.  If that fails, use mailto:std-c++@netlab.cs.rpi.edu]
[              --- Please see the FAQ before posting. ---               ]
[ FAQ: http://www.comeaucomputing.com/csc/faq.html                      ]

On 05/06/09 11:39, cppljevans@gmail.com wrote:
[snip]
>
> Probably what you're really missing is the concept of "unpacking
> patterns"
> as described in section 2.4 of:
>
>    http://www.generic-programming.org/~dgregor/cpp/variadic-templates.pdf
>
> There a pattern, AFAICT, can be some expression, E, which contains
> a "pack name".  For example, the section 2.4 contains:
>
>    Args&...
>
> The pack name is Args but the expression is Args&.  In the integral_c
> case, the pack name is C but the expression is integral_c<T,C>.
>
> Maybe "pack name" is the wrong term, but hopefully that makes
> things clearer.

A better reference is:

   http://www.open-std.org/jtc1/sc22/wg21/docs/papers/2007/n2242.pdf

where, in section 3.6.1 there terms:

   parameter pack
   pack expansion
   pattern of pack expansion (or pack expansion pattern)

In the list_c_expansion_wish.cpp code mentioned in my previous post,

   `Args` is a parameter pack.
   `integral_c<Integral,Args>...` is a pack expansion.
   `integral_c<Integral,Args>` is the pattern of the above pack
     expansion.

The section 3.6.1 suggested that the code should work; hence, I
simplified it to isolate the reason for its failure.  The code below
compiles OK; hence, there must be something wrong with the original
code. I'm still investigating why.

//-{--demo of gcc4.4 compiler w.r.t. parameter pack expansions--
   enum
A
{ a0
, a1
, a2
, a3
};

   template
   < typename Integral
   , Integral Value
   >
struct integral_c
{};
   template
   < typename... Types
   >
struct seq
{};
   template
   < typename Integral
   , Integral... Values
   >
struct seq_c
: seq
   < integral_c
     < Integral
     , Values //template param pack.
     >// pattern of pattern expansion.
     ...//the pattern expansion
   >
{
};

     typedef
   seq_c
   < A
   , a0
   , a1
   , a2
   >
actual_seq_cA3
;
//-}--demo of gcc4.4 compiler w.r.t. parameter pack expansions--

--
[ comp.std.c++ is moderated.  To submit articles, try just posting with ]
[ your news-reader.  If that fails, use mailto:std-c++@netlab.cs.rpi.edu]
[              --- Please see the FAQ before posting. ---               ]
[ FAQ: http://www.comeaucomputing.com/csc/faq.html                      ]