Shibboleth SP 2.5.3 fails to build with Boost 1.54

[Cantor, Scott]

On 3/16/14, 11:13 PM, "Russ Allbery" <rra at debian.org> wrote:
>
>I'm not sure what "argument 2" could be referring to, though, since the
>arrow seems to be pointing at the vector instantiation, which only has one
>argument.  A bunch of the remaining errors are from not being able to
>create the m_regex object.

There are usually default trait arguments and other things that don't show
up in the code.

>The error about tuple being ambiguous make me think that part of the code
>is picking up the Boost tuple and part is picking up the C++ tuple and
>they're somehow not compatible.

Oh, they aren't. There is no such thing in C++ prior to the latest
standard, and I'm nowhere near even considering C++ 11 right now, I have
too many old platforms to support for that.

If there are namespace problems because I left things defaulted, adding
boost namespaces to the right spots might fix that.

>  When I force everything to std::tuple
>(I'm not sure if that's a meaningful thing to do, but thought I'd try it),
>most of the errors go away, but I get a different error:

I think you probably should try the reverse, but definitely avoid
std::tuple. That would be a red flag for sure and you're right, that
appears to be the problem.

-- Scott