Date: Thu, 28 Oct 2021 02:33:51 +0000 From: Alexey Dokuchaev <danfe@freebsd.org> To: Stefan Esser <se@freebsd.org> Cc: ports-committers@freebsd.org, dev-commits-ports-all@freebsd.org, dev-commits-ports-main@freebsd.org Subject: Re: git: 00e15405660f - main - textproc/html2text: Update CONFLICTS Message-ID: <YXoMD4Gk5Ceb757C@FreeBSD.org> In-Reply-To: <7e5125f5-2e8d-5be7-3b0c-bb4401a3c58d@freebsd.org> References: <202110270841.19R8fLH2046856@gitrepo.freebsd.org> <YXkVJfKXVfb%2BzYR9@FreeBSD.org> <7e5125f5-2e8d-5be7-3b0c-bb4401a3c58d@freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Wed, Oct 27, 2021 at 08:25:13PM +0200, Stefan Esser wrote: > Am 27.10.21 um 11:00 schrieb Alexey Dokuchaev: > >> ... > >> -CONFLICTS= py27-html2text-[0-9]* > >> +CONFLICTS= py*-html2text-[0-9]* > > > > I recal, last time I've tried to specify a conflict without the -[0-9]* > > suffix and it worked as expected. Are those really necessary in default > > "all versions" case? > > And quite a few CONFLICTS specifications should actually be > CONFLICTS_INSTALL. But it takes a lot of work to check whether there > really is a build conflict for each of the combinations... I'm not sure if CONFLICTS alone is justified, and why can't we always use correct CONFLICTS_$TYPE. Checking if CONFLICTS should really be CONFLICTS_INSTALL could be automated, or so it seems at least for cases when prerequisite are met, by comparing pkg-plists of ports in question. CONFLICTS_BUILD are tougher; whenever I added one, it was as a result of stumbling across local build problem and wondering why it does not happen on the cluster, i.e. in pristine environment. Overall, many good points Stefan, nice write up. So that even if you won't pick up on this task, someone might now that you've made it less of an uncharted territory. ./danfe
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?YXoMD4Gk5Ceb757C>