Boost logo

Boost-Build :

From: Ken Smith (Ken.Smith_at_[hidden])
Date: 2004-05-05 17:15:45


Thank you for the detailed response. It's a shame that the product has
become fractured. Is there any hope in the future of reconciliation? I
think it would be better for the Jam community to be able to use and
talk about a single product.

Ken

David Abrahams wrote:

> Ken Smith <Ken.Smith_at_[hidden]> writes:
>
>
>>I am new to Jam and am weighing the benefits of each of the flavors of
>>Jam. Can anyone tell me why there are so many flavors of Jam and why
>>haven't some of the improvments introduced by the other flavors been
>>integrated into Perforce Jam?
>
>
> Perforce Jam has a very weak interpreted programming language built
> into it.
>
> Boost.Build is a high-level build system with a fair amount of code.
> We needed some basic language features that could support nontrivial
> software design. We added features like named arguments for rules,
> type checking, modules, rule indirection and while loops (Perforce got
> the last two eventually). We also routinely ran over Perforce Jam's
> internal fixed-size buffer limits for various string operations,
> crashing it, so we implemented a "mutable string" module that used
> dynamic allocation when neccessary. The list of enhancements has now
> grown to be quite a bit larger than this.
>
> http://www.boost.org/tools/build/jam_src/index.html#core_extensions
>
> I think Perforce's goal to keep Jam simple prevents them from wanting
> to pick up most of our changes.
>
>
>>As a brand new user of Jam, should I use Perforce Jam unless I run
>>into issues which Boost.Jam solves?
>
>
> It depends on your goals. If some of your needs are described here:
>
> http://www.boost.org/tools/build/v2/boost_build_v2.html
>
> or
>
> http://www.boost.org/tools/build/v1/build_system.htm#design_criteria
>
> Then use Boost.Build. Otherwise, you might go with one of the others
> if you're not afraid of the inherent limitations.
>
>
>>Is Perforce Jam considered the official version?
>
>
> I guess. Our forks have diverged.
>

 


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