Re: Re: Boost::Ruby

* David Abrahams <[email protected]> [2005-10-12 12:06]:
> Alan Gutierrez <[email protected]> writes:
> 
> > Hello. Responding to Rene's message on the boost mailing list to
> >     post here regarding a Ruby langbinding.
> >
> >     I've checked out a lot of languaged bindings for reference, as I
> >     noted, these are the ones that looked relevant...
> >
> >     http://engrm.com/wiki/Refresh#Ruby.2FC.2B.2B_Bindings
> >
> >     The Java bindings are a better reference for me than Python
> >     bindings, since I have no Python experience.

> However, the design of that particular binding facility may not
> map very well onto langbinding, FWIW.  We do have a rough strategy
> for Java, but IIUC it will have to be a bit different from the
> others, because the langbinding library will have to generate JNI
> code that is eventually compiled.

    The JNI requires the native side to use a naming convention. I'd
    not thought this through that far, but that does sound like more
    of an imposition on the C implementation that other bindings I'm
    familiar with.
    
    The JNI implementation that caught my interest was Noodle Glue,
    with it's long list of supported language features. It does use
    Doxygen, and multiple files to solve the problem (did some
    reading since last posting), which is too much of a solution.
    
    Other languages bind with less.

--
Alan Gutierrez - [email protected] - http://engrm.com/blogometer/


-------------------------------------------------------
This SF.Net email is sponsored by:
Power Architecture Resource Center: Free content, downloads, discussions,
and more. http://solutions.newsforge.com/ibmarch.tmpl