Subject: Re: [Boost-build] Python port development
From: Artyom (artyomtnk_at_[hidden])
Date: 2010-07-09 14:40:36
> > - Normal readable documentation (that Boost.Build just do not have)
> > - Build time checks like check if code sample compiles,
I'm glad it is so.
> This is now supported through the configure module.
> > - Build time checks like sizeof something.
> This isn't possible now, but can be implemented via the
> same mechanisms that are used to check if a target builds.
> > - Build time checks like check if standard library supports function (build
>in like iconv, socket etc)
> > - Build time search for specific function is several optional libraries.
> These are special cases of checking if a target builds.
I understand clearly that if you have something that test that something
you can do the rest. The question is what the effor?
One line? two lines? 100 lines?
Telling that A can be solved when B exists is not much then telling that if you
have Turing complete language you can solve any algorithmic problem.
The question is totally practical.
> > - Creation of config.h like files or results of compile time checks..
> Generating header files is fairly easy. I usually use make/print.
Easy but still. In autotools it is just generated with all checks, in CMake you
have a simple template that is filled automatically.
Once again. I can write everything and anything but the question is the effort.
I'm sure that you can do most of stuff with BB, the question is how
simple is that.
Same way: "I don't need Boost as everything Boost does can be implemented in C++
or even C."
Boost-Build list run by bdawes at acm.org, david.abrahams at rcn.com, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk