It might be useful for prototyping of filesystems, which is what I use
perl for a lot of times (prototyping in general, that is :-)
On Fri, 6 Nov 1998, Mikael Karpberg wrote:
> According to Ruslan Shevchenko:
> > Jordan K. Hubbard wrote:
> > > http://dd.sh/perlfs/
> > IMHO, Linux-ism
>
> IMO: "Maybe, but so what? That's WAY COOOOOL!"
> it's not a good idea to scrap all the filesystems and redo them in perl,
> or anything... but as an extension it can only bring good.
>
> It IS a rather neat idea... If we could make a similar thing that had the
> same API, we could share code with them too. And yes... it's less efficient,
> but then again... You don't always care, do you? I mean... if it gets to be
> a generic interface (version are written for all BSDs too, etc) then we
> might find a lot of obscure filesystems implmented like that... which
> might allow us to support reading/write such filsystems to people that
> need it, without having to do anything. And these people will be MUCH
> happier that they can read/write that file they needed to/from their
> obscure filesystem at half the speed then not at all.
>
> I wonder how much work it would be... How do they integrate perl with
> the kernel?? That could get ugly :)
>
> *starts diggint for details*
--
Christopher Nielsen
Scient: The eBusiness Systems Innovator
<http://www.scient.com>
cnielsen@scient.com
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-hackers" in the body of the message