From:	Boon van der RJ <rjvdboon AT cs DOT vu DOT nl>
Newsgroups:	comp.os.msdos.djgpp
Subject:	Re: Recursive make: portable technique?
Date:	23 Oct 1998 11:42:54 GMT
Organization:	Fac. Wiskunde & Informatica, VU, Amsterdam
Lines:	38
Message-ID:	<70pq3u$cju$1@star.cs.vu.nl>
References:	<362FF086 DOT BDA6266C AT montana DOT com>
NNTP-Posting-Host:	sloep68.cs.vu.nl
User-Agent:	tin/pre-1.4-981002 ("Phobia") (UNIX) (SunOS/5.5.1 (sun4m))
Originator:	rjvdboon AT sloep68 DOT cs DOT vu DOT nl
To:	djgpp AT delorie DOT com
DJ-Gateway:	from newsgroup comp.os.msdos.djgpp
Reply-To:	djgpp AT delorie DOT com

bowman <bowman AT montana DOT com> wrote:
> I found that doing an explicit  cd ../original_dir fixed the problem,
> and there were a few exchanges about this in the archives that indicated
> this was the case. 

> I ran a small test using this hack with bash available, and pointed to
> by SHELL, and it still seemed to work.

> So, my question, is the explicit cd safe in all configurations, or will
> I get burnt?

Don't know about all systems, but on our unix (Sun) system, using bash
with -P (follow physical links, not symbolic) could(not?) create
problems.[1]
But if SHELL=/bin/sh I think there should be no problem.

bye now,
 Robert.


[1] What I mean is, if you have a tree set up like:
/usr/package1
/usr/package2 -> /net/package2

you make in /usr/package1, this make says:
cd ../package2; make
cd ../package1

the cd ../package2 sets you to /net/package2, which means that cd
../package1 doesn't work (package1 is in /usr, not /net) but if make
returns to package1 by itself then the cd ../package1 can't hurt
either.
So bottom line is: I don't think there will be a problem with such an
approach (not even taking into acount the unlikeliness of the tree set
up and use of it described above)
-- 
rjvdboon AT cs DOT vu DOT nl        | "En dat is niet waar!" sprak ex-Staatsecretaris-
www.cs.vu.nl/~rjvdboon   |    van-Onderwijs Netelenbos fel.

- Raw text -