Re: [boost] Recent PP changes

From: David Abrahams (david.abrahams_at_[hidden])
Date: 2002-06-18 08:25:02

From: "Paul Mensonides" <pmenso57_at_[hidden]>

> --I don't know how to use the CVS anyway.  I just downloaded cygwin the
day
> before yesterday, and all that I know how to do is login, update -or-
checkout,
> and logoff.  I don't even know how to get the MPL files other than
downloading
> them from the CVS shell website. :(

from your boost root:

    cvs update -rmpl-development boost/mpl

> So, if you want to tell me what I need to do
> or where to go for instructions, go ahead.

Let me know if you need to know anything else.

> But right now, here is a *partial*
> solution that will fix the problem for now--at least the ones I know
about
> specifically.
>
> Okay.  Dave this is what you need to change:
>
> file:  preprocessor/logical/bool.hpp
>
>     1.  remove the inclusion of <boost/preprocessor/detail/cat.hpp>
>     2.  redefine BOOST_PP_BOOL as follows:
>
>         #define BOOST_PP_BOOL(X) BOOST_PP_BOOL_DELAY(X)
>         #define BOOST_PP_BOOL_DELAY(X) BOOST_PP_BOOL##X
>
> file:  preprocessor/if.hpp
>
>     1.  remove the inclusion of <boost/preprocessor/detail/cat.hpp>
>     2.  redefine BOOST_PP_IF as follows:
>
>         #define BOOST_PP_IF(COND,THEN,ELSE) \
>             BOOST_PP_IF_BOOL(BOOST_PP_BOOL(COND),THEN,ELSE)
>         #define BOOST_PP_IF_BOOL(COND,THEN,ELSE) \
>             BOOST_PP_IF_BOOL_DELAY(COND,THEN,ELSE)
>         #define BOOST_PP_IF_BOOL_DELAY(COND,THEN,ELSE) \
>             BOOST_PP_IF##COND(ELSE,THEN)
>         //--------------------^ note these are backwards
>
> Also, if you like you can fix the ordering of ELSE and THEN if you fix
the IF0
> and IF1 macros appropriately.  That is a relic from when Vesa was using
> TUPLE_ELEM to get the elements.  So, if you switch those two arguments in
the
> last 'call' above, you also have to change these right below:
>
>     #define BOOST_PP_IF0(T,E) E
>     #define BOOST_PP_IF1(T,E) T
>
> That should fix all the problems in 'apply.hpp' and 'count_if_not.hpp'
(and
> possibly others).  However, I swear this is going to come up again.

See, this is just getting complicated. Since you understand what needs
doing, I'd rather just give you CVS access so you can make and test the
fixes yourself. Every time I bring the PP sources up-to-date and try some
new patch that doesn't work, I end up having to recompile everything for
all of the compilers I'm testing my code with. Why don't you send me your
SourceForge username (get one if you don't have one).

-Dave