Re: [Boost-build] bjam 4.0.. in C++

Subject: Re: [Boost-build] bjam 4.0.. in C++
From: Rene Rivera (grafikrobot_at_[hidden])
Date: 2010-05-27 14:36:23

On 5/27/2010 12:41 PM, Matt Chambers wrote:
> Spencer E. Olson wrote:
>> I'd like to hear feedback with regards to the future of
>> Boost.Build/bjam.

To respond to Spencer briefly.. Volodya and myself are committed to 
supporting BBv2/bjam for the long term future. Myself because it's what 
I use every day. And my efforts with rewriting bjam in C++ are to make 
that long term support easier. And more importantly, more likely to 
attract outside developers since as it stands the bjam C code base is 
almost impenetrable to first timers. It took me years to get the hang of 
the low level code enough to start extending it.

> I'm very grateful for Rene and Volodya's continued efforts on BBv2.
> However, I'm somewhat more supportive of Volodya's effort to port BB's
> higher level logic from Jam to Python than I am of replacing bjam's low
> level C parsing code with C++, even if it would mean the parser was
> easier to maintain. Also, isn't it easier to call simple C routines from
> Python than it is to call C++ when high performance is needed?

First, it's not really the parser that is the goal in the rewriting of 
bjam. It's the core make functionality. The parser is currently an 
impediment in rewriting because it's essentially required to make sure a 
complete bjam is working correctly (i.e. works as it does now). We have 
to consider that we are working with a live system and can't really do 
complete reworks and guarantee stability. So we approach the build 
system incrementally, we change and replicate and replace parts at a 
time so we can test that everything still works. The high-level BBv2 
logic being rewritten in Python is close to done. Which leaves the low 
level bjam to rewrite. Perhaps a bigger picture is more informative at 
this point. In some near future the goal is have a modular build system 
that is usable from many context. We should be able to easily integrate 
it into IDEs, or even something like directly into Windows Explorer. 
Hence we need to move towards modular architecture.

In summary, the parser is a rather small part of this. And it's not the 
important part. The important parts are the core bjam and BBv2. For the 
core I, still, promise to minimize the requirements of C++ as much as 
possible. Although I will certainly use templates, it will not be the 
universal hammer that Boost Libraries are known for ;-) Basically, the 
requirements on the C++ compiler will be tantamount to just having an 
STD of some sort.

-- 
-- Grafik - Don't Assume Anything
-- Redshift Software, Inc. - http://redshift-software.com
-- rrivera/acm.org (msn) - grafik/redshift-software.com
-- 102708583/icq - grafikrobot/aim,yahoo,skype,efnet,gmail