Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 01 Apr 2002 18:55:20 -0800
From:      Jordan Hubbard <jkh@winston.freebsd.org>
To:        Mikhail Teterin <mi@aldan.algebra.com>
Cc:        obrien@FreeBSD.org, cvs-committers@FreeBSD.org, cvs-all@FreeBSD.org
Subject:   Re: cvs commit: src/usr.sbin/sysinstall install.c installUpgrade.c media.c package.c 
Message-ID:  <23637.1017716120@winston.freebsd.org>
In-Reply-To: Message from Mikhail Teterin <mi@aldan.algebra.com>  of "Mon, 01 Apr 2002 20:30:04 EST." <200204020130.g321U4FX074483@aldan.algebra.com> 

next in thread | previous in thread | raw e-mail | index | archive | help
> On  1 Apr, Jordan Hubbard wrote:
> >> Why can't the packages be extracted "in place" by the sysinstall?
> > 
> > Because that's just not the way it works. If you wanted to do that,
> > you'd need to invert the logic of pkg_add to extract in place and then
> > clean up the non-essentials, rather than extracting in a temporary
> > directory, moving only the essentials into place, and then deleting
> > the temp dir.
> 
> I just wonder how, then, the ``@option extract-in-place'' is implemented
> and why is it not used by default...

You wonder only because you haven't reviewed the code exection path
for any actual package addition operations. :-)

@option extract-in-place is not for anything produced by
pkg_create(1).  It's for tarballs masquerading as packages that don't
depend on the package support routines for anything other than
extraction.  As far as I know, no one has ever used this feature and
it's basically one of those use-at-your-own-risk experimental sort of
features.

> I'm not so certain of that, as a matter of fact. My last attempt to
> enhance pkg-code is still waiting (after two months) in the bin/34628:
> 
> 	http://www.freebsd.org/cgi/query-pr.cgi?pr=bin/34628

Probably because not too many people understand that code and your
proposed enhancement is a little obscure in any case.  The only thing
it really visibly adds is an extra warning message.  You're also responsible
for checking the result of sscanf() in your code since not all @comment
lines are md5 data - they can be anything.  And yeah, in retrospect I'm
sorry I didn't add an @md5 directive.

- Jordan

To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe cvs-all" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?23637.1017716120>