Re: [Boost-users] [Review] Phoenix V3: Mini-review starts February 20th

Subject: Re: [Boost-users] [Review] Phoenix V3: Mini-review starts February 20th
From: Larry Evans (cppljevans_at_[hidden])
Date: 2011-02-26 17:08:41

On 02/26/11 15:55, Larry Evans wrote:
> On 02/26/11 15:25, Michael Caisse wrote:
>> On 02/26/2011 01:08 PM, Larry Evans wrote:
>>>
>>> followed by several other error messages.
>>>
>>> Any ideas what's wrong?
>>>
>>> -Larry
>>>
>>>
>>>    
>>
>> Hi Larry -
>>
>> FWIW, I think most of us test libraries under review by placing them
>> within a boost source tree. Is there a reason that this approach is not
>> acceptable?
>>
> 
> Well, I just hadn't thought of that.
> IOW, I'd move:
> 
>   /home/evansl/prog_dev/boost-svn/ro/trunk/sandbox/ro/SOC
>     /2010/phoenix3/libs/phoenix
> 
> to:
> 
>   /home/evansl/prog_dev/boost-svn/ro/trunk/libs/phoenix
> 
> and:
> 
>   /home/evansl/prog_dev/boost-svn/ro/trunk/sandbox/ro/SOC
>     /2010/phoenix3/boost/phoenix
> 
> to:
> 
>   /home/evansl/prog_dev/boost-svn/ro/trunk/boost/phoenix
> 
> and then run the tests?
> 

However, I'm not sure that would avoid the error messages I last
reported:

gcc.compile.c++
bin/container_tests1a.test/gcc-4.4.3/debug/container/container_tests1a.o
In file included from
/home/evansl/prog_dev/boost-svn/ro/trunk/boost/proto/matches.hpp:33,
                 from ../../../boost/phoenix/core/meta_grammar.hpp:15,
                 from ../../../boost/phoenix/core/is_nullary.hpp:15,
                 from ../../../boost/phoenix/core/actor.hpp:18,
                 from ../../../boost/phoenix/core.hpp:12,
                 from container/container_tests.hpp:11,
                 from container/container_tests1a.cpp:7:
/home/evansl/prog_dev/boost-svn/ro/trunk/boost/mpl/aux_/template_arity.hpp:86:1:
error: macro "BOOST_PP_TUPLE_ELEM_2_0" requires 2 arguments, but only 1
given

The above message does not report any missing .hpp file, it just
reports a macro given the wrong number of arguments.  How would
changing the location of the phoenix files avoid that kind of
error?

-Larry