From owner-freebsd-ports Sun Mar 11 19:30:56 2001 Delivered-To: freebsd-ports@freebsd.org Received: from guru.mired.org (okc-65-26-235-186.mmcable.com [65.26.235.186]) by hub.freebsd.org (Postfix) with SMTP id 2189537B71A for ; Sun, 11 Mar 2001 19:30:48 -0800 (PST) (envelope-from mwm@mired.org) Received: (qmail 84776 invoked by uid 100); 12 Mar 2001 03:30:46 -0000 From: Mike Meyer MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Message-ID: <15020.17126.838366.365347@guru.mired.org> Date: Sun, 11 Mar 2001 21:30:46 -0600 To: ports@freebsd.org Subject: cookies vs. sup updates X-Mailer: VM 6.89 under 21.1 (patch 14) "Cuyahoga Valley" XEmacs Lucid X-face: "5Mnwy%?j>IIV\)A=):rjWL~NB2aH[}Yq8Z=u~vJ`"(,&SiLvbbz2W`;h9L,Yg`+vb1>RG% *h+%X^n0EZd>TM8_IB;a8F?(Fb"lw'IgCoyM.[Lg#r\ Sender: owner-freebsd-ports@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org I've discovered an interesting systemic problem with the ports system The ports system has an implicit assumption that updating the ports tree will remove all the cookies. When this doesn't happen, the world stops behaving in a sane manner. The first noticable instance is that "pkg_version -c" doesn't actually rebuild or install things. In the sequence cd make && pkg_delete -f make intall the make & make install will do nothing, because the cookies are in place. The most interesting cases happen with ports that are partially installed and then updated. For instance, you've installed the port, then deinstalled it. At a later date, after updating the port in /usr/ports, a "make install" will jump straight into trying to install the port - but the software the port will be trying to install a different version of the software than it actually has. Since we now have PORTREVISION - and that is hopefully changed even if PORTVERSION isn't - the Makefile should change if the port changes. It seems like having the various cookies depend on the Makefile would solve the problems I'm having, and wouldn't break anything during normal use. Debugging ports might be more interesting, but possibly a knob could fix that. Since I'm not subscribed to -ports, please cc: me on any replies. Thanx, http://www.mired.org/home/mwm/ Independent WWW/Perforce/FreeBSD/Unix consultant, email for more information. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-ports" in the body of the message