On 18/10/2010, at 3:35 AM, Erick Tryzelaar wrote:

> On Sun, Oct 17, 2010 at 6:12 AM, john skaller
> <skal...@users.sourceforge.net> wrote:
>> For "Unix" I'm thinking: how to get around the mess of variants?
>> 
>> The answer is: follow the standards, in particular:
>> 
>> [...]
> 
> That's what I ended up doing with fbuild.config:
> 
> http://github.com/erickt/fbuild/tree/master/lib/fbuild/config/
> 
> It'd be nice to have some parallels between the two. One thing to keep
> in mind is how to deal with incompatibilities between standards. Some
> functions didn't get well defined return types until posix04, some
> mmap implementations have extra macros, and etc. Any good ways of how
> to deal with that?

Well, if we have say Linux and OSX using different types,
we just have to add tags for that too :)

It is tricky, depending on how we want to handle type stupidity
present in C and Unix .. and Windows .. and .. yeah, all of it.
Sometimes there are two functions, one for each type and
sometimes just one and a typedef, and sometimes we're using
a abstract primitive in Felix to munge over the difference.

In read(fd ...  is fd an int, or is it a typedef for and int or perhaps a
long, or is it instead abstract? If abstract you can't, for example,
print it (without a conversion function).

Clearly at the higher levels it should be abstract, but in the base
primitives? Dunno.



--
john skaller
skal...@users.sourceforge.net





------------------------------------------------------------------------------
Download new Adobe(R) Flash(R) Builder(TM) 4
The new Adobe(R) Flex(R) 4 and Flash(R) Builder(TM) 4 (formerly 
Flex(R) Builder(TM)) enable the development of rich applications that run
across multiple browsers and platforms. Download your free trials today!
http://p.sf.net/sfu/adobe-dev2dev
_______________________________________________
Felix-language mailing list
Felix-language@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/felix-language

Reply via email to