Re: [Boost-build] Future development

From: Vladimir Prus (ghost_at_[hidden])
Date: 2007-10-04 01:15:49

Benjamin Kosnik wrote:

> If you attempt to qualify usage of build tools in general on linux (say
> with debian popularity contest, link:)
> 
> http://popcon.debian.org/main/devel/by_recent
> 
> You'll see something like (rank/build tool)
> 
> 2     make (gnu)
> 53    ant
> 71    automake1.9
> 75    automake
> 97    automake1.4
> 115   automake1.7
> 182   cmake
> 241   automake1.8
> 354   bjam
> 368   boost build

Am I right that the number is position in the list, not just
any metric of installation. If so, all you data proves that
that cmake now is more used than boost.build -- which is
something I agree to in another email. So, what's your point?

>> > I think the answer is "bjam". As you said above, the Jam language is
>> > essentially undocumented. It's also a rather odd language, and its
>> > eccentricities are exaggerated by the object-oriented system BBv2
>> > has built on top of Jam. We picked Jam several years ago, when
>> > Perforce was still working on it and it looked like there would be
>> > an active community around Jam. That hasn't happened, and we're
>> > somewhat isolated in our use of this language and build tool,
>> > forced to customize and maintain "bjam" on our own.
>> 
>> You surely understand that with core in Python, the amount of
>> maintenance and customization necessary for bjam will be greatly
>> reduced, almost to the point of being zero.
> 
> Famous last words.
> 
> FWIW, how is a rewrite into python addressing Doug's
> meta comment about the Jam language itself?

Simple -- most users have no problem writing:

        exe a : a.cpp ;

the problems start when they try to extend Boost.Build. And using
Python for that will greatly simplify things.

> The BBv1 to BBv2 transition was long and painful. Now, with BBv3 a
> total rewrite, I'd expect something similar in the future, and
> boost-1.3x.0 to experience similar delays.

For the record, at least 50% of 1.34 delay had nothing to do with
Boost.Build (and don't get me started on that). And given that
rewrite is supposed to be 100% compatible for Jamfiles, I'd
expect a drop-in replacement.

>> Judging from the current distribution of user complains, it seems to
>> me that Python port is going to be immediate improvement, and the end
>> result will be pretty good. Other approaches seem much more risky.
> 
> Could you elaborate on what you see are the top 10 user complaints? I'm
> curious.

Had you read my initial email? It has 2 top user complaints.

- Volodya