Re: [jamboost] Autolinking and Boost.Build

From: Vladimir Prus (ghost_at_[hidden])
Date: 2005-08-02 04:46:42

On Tuesday 02 August 2005 13:32, John Maddock wrote:
> > They don't. When library targets are generate to variants directories,
> > using
> > that name mangling scheme does not seem necessary -- it's all in the
> > target
> > path already.
>
> Oh shucks, so the question then is: if you did add <define>BOOST_ALL_NO_LIB
> as a requirement to all bbv2 targets, how on earth do we verify that the
> installed mangled names are correct? I repeat what I said before: this is
> a common cause of bugs, so eyeball inspection of the names is not a
> solution. If you can come up with some other kind of test (that can be run
> as part of our regression tests) that would validate that the autolink and
> installed names match then I'll be happy.

Don't regression test build just a single variant? No debug/release 
single-threaded/multi-threaded static-runtime/dynamic-runtime combos. If 
there are no all such combos, then the testing of autolink is not so 
comprehensive.

Answering your question -- it might be possible to add BOOST_ALL_NO_LIB 
conditionally.

- Volodya

-- 
Vladimir Prus
http://vladimir_prus.blogspot.com
Boost.Build V2: http://boost.org/boost-build2