Re: [boost] [mpl][docs]why use list?

Subject: Re: [boost] [mpl][docs]why use list?
From: Larry Evans (cppljevans_at_[hidden])
Date: 2009-03-23 08:30:44

On 03/23/09 06:11, Dmitry Goncharov wrote:
> 
> 
> Larry Evans wrote:
>>
>> Unfortunately, following the directions on:
>>
>> http://www.boost.org/doc/libs/1_38_0/libs/mpl/doc/refmanual/limit-vector-size.html 
>>
[snip]
>>
>> Does anyone know what I might be doing wrong?
>>
>>
> You can find the answer here
> http://lists.boost.org/boost-users/2004/08/7647.php.

Thank you, Dmitry.

AFAICT, this is a documentation bug.  Although limit-vector-size.html
contains:

   [Note: Overriding will take effect only if the library is configured 
not to use preprocessed headers. See 
BOOST_MPL_CFG_NO_PREPROCESSED_HEADERS for more information. — end note]

looking at:

http://www.boost.org/doc/libs/1_38_0/libs/mpl/doc/refmanual/cfg-no-preprocessed.html

which contains:

   When defined, it instructs the MPL to discard the pre-generated 
headers found in boost/mpl/aux_/preprocessed directory and use 
preprocessor metaprogramming techniques to generate the necessary 
versions of the library components on the fly.

suggests that mpl will do the metaprogramming when, in fact,
the user must either do the metaprogramming:

   http://lists.boost.org/boost-users/2004/08/7647.php

or the user must first invoke a python program to generate
some headers before doing the compile:

   http://lists.boost.org/boost-users/2004/08/7621.php

If nobody disagrees (within the next day or so), I'll
create a trac-ticket for this doc bug.

-regards,
Larry