Date: Fri, 03 Jul 1998 23:58:05 -0400 From: "Christopher R. Bowman" <crb@ChrisBowman.com> To: Mike Smith <mike@smith.net.au> Cc: Thomas David Rivers <rivers@dignus.com>, drosih@rpi.edu, wjw@surf.IAE.nl, hackers@FreeBSD.ORG Subject: Re: Variant Link implementation, continued Message-ID: <199807040513.AAA15582@quark.ChrisBowman.com> In-Reply-To: <199807040226.TAA07461@antipodes.cdrom.com> References: <Your message of "Fri, 03 Jul 1998 13:02:38 EDT." <199807031702.NAA19145@lakes.dignus.com>
next in thread | previous in thread | raw e-mail | index | archive | help
At 10:26 PM 7/3/98 , Mike Smith wrote:
>> >
>> > Then I'll be thinking about haveing 2 rules of resolution:
>> > @{....}
>> > and ${....}
>> >
>>
>> I don't mean to badger... but what if you, in an existing installation,
>> already have symlinks that contain that text? Won't adding this
>> facility break those existing links?
>>
>> [And, don't laugh, but I do have links and files that begin with '$',
>> and, even worse, have '$' embedded in the middle of them...]
>
>In the existing sample implementation, you would have to have links
>whose names comply explicitly with the syntax ...${<tag>}... where <tag>
>is a valid tag in the variant link namespace.
>
>I think that this is sufficiently unlikely given that there have been
>only two respondents that actually use '$' in names at all...
Does anybody else get the feeling we are reinventing plan9 here?
I only know what little I remember of the few plan9 papers I read, but this
does sorta seem like a solution to the problems they were having with
different architectures, and seems like a sort of hack attempt at their
directory/name space manipulations.
--------
Christopher R. Bowman
crb@ChrisBowman.com
http://www.ChrisBowman.com/~crb
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-hackers" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199807040513.AAA15582>