Path: utzoo!attcan!uunet!samsung!umich!yale!husc6!cmcl2!adm!smoke!gwyn
From: gwyn@smoke.BRL.MIL (Doug Gwyn)
Newsgroups: comp.sys.apple2
Subject: Re: Parent pointers
Message-ID: <13745@smoke.BRL.MIL>
Date: 5 Sep 90 10:11:05 GMT
References: <5928366@ub.cc.umich.edu> <1990Sep5.031342.27820@cunixf.cc.columbia.edu>
Organization: U.S. Army Ballistic Research Laboratory, APG, MD.
Lines: 19

In article <1990Sep5.031342.27820@cunixf.cc.columbia.edu> vw3@cunixf.cc.columbia.edu (Vernon Williams) writes:
>In article <5928366@ub.cc.umich.edu> Tabakal@UB.CC.UMICH.EDU writes:
>>Do standard ProDOS subdirectories have a pointer to their parent 
>>directory?  Or do they have to scan from root to find it?  It seems
>>that noone implements the equivalent of a .. parameter when they
>>implement a change directory command (with the exception of APW).

I've created for my own purposes an implementation of a canonicalizer
for GS/OS pathnames, which I plan to post to the sources group some day.
chdir("..") is implementable simply by trimming off the trailing piece
of the current prefix #0, which is the GS/OS equivalent of UNIX CWD.
By interposing such a canonicalizer in the path taken by an argument
to any filename-driven function, you automatically get UNIX-like path
interpretation for the function.

By the way, I hereby stake out a claim to the external name subspace
matching the regular expression "_Q8Q_*" for my own use in internal
support for UNIX-like C library extensions.  (Unfortunately, there is
no central registry for allocating portions of global name space.)