Boost logo

Boost-Build :

From: David Abrahams (david.abrahams_at_[hidden])
Date: 2002-07-10 11:10:46

I can't agree. For example, the maximum arity supported by Boost.Python can
be adjusted by changing a #define. In theory this creates ODR violations,
but in practice each individual user of the library can set this #define
differenty for his own use. How would we support that?

Answering my own question: one possibility might be that users write
specific command-line option handling code in Jamfiles:



----- Original Message -----
From: "Vladimir Prus" <ghost_at_[hidden]>
To: <jamboost_at_[hidden]>
Sent: Wednesday, July 10, 2002 11:59 AM
Subject: [jamboost] Disallow free property in build request

> I think that we should disallow free but non-propagated properties in
> request. The reason is that for example:
> exe a.exe : a.cpp biglibrary : <define>FOO ;
> Here, <define>FOO is not propagated to biglibrary, i.e. not used in its
> building. Likewise, even if <define>BAZ is present in build request for
> a.exe, it won't affect biglibrary. So, free properties in build request
> either
> - affect the directly build targets
> - affect nothing
> Both semantics are odd. Let's disallow free properties in build request
> altogether.


Boost-Build list run by bdawes at, david.abrahams at, gregod at, cpdaniel at, john at