[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Chicken-hackers] too many core modules?
From: |
John Cowan |
Subject: |
Re: [Chicken-hackers] too many core modules? |
Date: |
Tue, 1 Sep 2009 11:51:57 -0400 |
User-agent: |
Mutt/1.5.13 (2006-08-11) |
Kon Lovett scripsit:
> Personally I would like to go further in splitting files (Unit library
> itself is kinda unwieldy), creating units like 'pathname' and
> 'system'.
I wasn't even going to touch that one yet, since everything in "library"
is included by default. But in principle you are right.
> I do want to endorse the suggestion by Alex for an '(extend
> MODULE ...)' form; I build these manually. Would a consumer need to
> 'require-library' to get the binary of the extended module or just
> require the extending module or both? Should extending a subset of a
> module and/or renaming/prefixing the identifiers be supported?
One problem is that every module must declare what it exports, which
creates bad coupling between the compound module and its components.
I'd rather see a simpler (compound-module newname . oldnames) that would
just allow combining modules rather than extending them.
--
Unless it was by accident that I had John Cowan
offended someone, I never apologized. address@hidden
--Quentin Crisp http://www.ccil.org/~cowan
- Re: [Chicken-hackers] too many core modules?, (continued)
Re: [Chicken-hackers] too many core modules?, Kon Lovett, 2009/09/01
- Re: [Chicken-hackers] too many core modules?,
John Cowan <=