|
Boost-Build : |
Subject: Re: [Boost-build] [future] Implementation language(s)..
From: Stefan Seefeld (stefan_at_[hidden])
Date: 2016-10-25 10:08:12
On 25.10.2016 09:58, Rene Rivera wrote:
> On Tuesday, October 18, 2016, Stefan Seefeld <stefan_at_[hidden]
> <mailto:stefan_at_[hidden]>> wrote:
>
> On 18.10.2016 10:28, Klemens Morgenstern wrote:
> > I'd provide both C++ and Python extensions, otherwise I do
> agree. I'll
> > see if I can come up wiht a way to build a plugin-system that may
> > understand more than one language, but with one interface; if that
> > works providing more than one language for extensions shouldn't be a
> > problem at all.
>
> I'd caution against that. A build system consisting of plugins written
> in many different languages is just begging for trouble for
> maintenance.
>
>
> After some thought for the past week.. I agree. Multiple languages at
> the plugin level would make it a nightmare for users to know what they
> need as far a requirements.
I was actually more thinking of the community as a whole. Even if it's
easy for the implementer to add new plugins in the language of his
choice, having a plethora of plugins written in many different languages
is is hard for others to embrace. (I'm assuming here that plugins aren't
just written by a user for his own consumption, but to be shared and
collaboratively maintained). It's much better for the community to have
a project with few "lingua francas" to focus on, and I think that the
combination of C++ and Python are an excellent choice in that domain.
> But I think it would be OK to write Python plugins that make use of
> native (i.e. C/C++) library/module. As in that case the compilation of
> that ends up in the module install instead of being up to BB to try
> and compile and use on the fly.
>
> Rene.
Stefan
-- ...ich hab' noch einen Koffer in Berlin...
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