Re: SLIB support

[Rob Browning]

Andreas Rottmann <address@hidden> writes:

> I've now prepared a patch (attached) that's hopefully good enough
> for upstream use.

> +(define-module (ice-9 slib)
> +  :export (slib:load slib:load-source defmacro:load
> +        implementation-vicinity library-vicinity home-vicinity
> +        scheme-implementation-type scheme-implementation-version
> +        output-port-width output-port-height array-indexes
> +        make-random-state
> +        -1+ <? <=? =? >? >=?
> +        require slib:error slib:exit slib:warn slib:eval
> +        defmacro:eval logical:logand logical:logior logical:logxor
> +        logical:lognot logical:ash logical:logcount logical:integer-length
> +        logical:bit-extract logical:integer-expt logical:ipow-by-squaring
> +        slib:eval-load slib:tab slib:form-feed difftime offset-time
> +        software-type)
> +  :replace (provide provided?)
> +  :no-backtrace)
> +
> +
> +(load "@SLIB_LOCATION@/init/guile.init")

In the long run, or perhaps even in the short run if Aubrey Jaffer's
interested in a patch and can make a fairly quick release, I wondered
if we might want to go even further, i.e. change it so that the *only*
thing in ice-9/slib.scm is

  (load-from-path "slib/guile.init")

I suggest this in part because I believe SLIB already has code (which
might or might not need adjustment) to handle defining the module,
making the exports, etc.  It's just conditionalized on the Guile
version.

If I recall correctly, SLIB's module related code is fairly automated
and is actually part of an implementation independent module
framework.  So the definition in ice-9/slib.scm is probably just a
duplication of effort.

Given that, it seemed like it might be better to just work with Aubrey
Jaffer to maintain all of the code in one place.

-- 
Rob Browning
rlb @defaultvalue.org and @debian.org; previously @cs.utexas.edu
GPG as of 2002-11-03 14DD 432F AE39 534D B592 F9A0 25C8 D377 8C7E 73A4