Date: Mon, 17 May 2021 08:55:15 -0700 From: bob prohaska <fbsd@www.zefox.net> To: George Mitchell <george+freebsd@m5p.com> Cc: Kubilay Kocak <koobs@freebsd.org>, FreeBSD ports <freebsd-ports@freebsd.org> Subject: Re: Python 37/38 conflict, was Re: Trubles compiling lxqt on RPi4 Message-ID: <20210517155515.GA77731@www.zefox.net> In-Reply-To: <1eef122c-f10e-4010-bfd6-0bb1cfa5ac82@m5p.com> References: <515FCC01-19A2-463C-8416-85D0BF0B4845.ref@yahoo.com> <515FCC01-19A2-463C-8416-85D0BF0B4845@yahoo.com> <20210514013518.GA46967@www.zefox.net> <18651bb2-4093-af83-da8f-d57553fffc9d@FreeBSD.org> <20210514163514.GA52420@www.zefox.net> <802898bf-fb68-f648-b893-1bebb02d9c16@FreeBSD.org> <20210517021927.GA66198@www.zefox.net> <1eef122c-f10e-4010-bfd6-0bb1cfa5ac82@m5p.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, May 17, 2021 at 09:37:07AM -0400, George Mitchell wrote: > On 5/16/21 10:19 PM, bob prohaska wrote: > > [...] > > I'd like to see the ports system keep working as it has in the past, but that seemingly > > requires a kind of machine intelligence that hasn't evolved yet. Poudriere seems like > > a brute force approach. [...] > > You'll find quite a few remaining fans of portmaster. Occasionally it > puts you in dependency hell if you don't run "pkg check -ad" and "pkg > check -aB" often enough, but I've given up on poudriere more than once. 8-) Poudriere seems adapted to a closed-source system where the primary goal is expedient production of binary-only software. It doesn't explicitly close the source, but does discourage access. I fiddled with portmaster some time ago while trying to compile www/chromium on a Raspberry Pi3. It seemed more prone to getting stuck than a simple make -DBATCH when all the dust settled. A large fraction of stoppages were related to refusal to upgrade old ports that were already installed. Since portmaster was advertised as a way to "upgrade" existing ports that was surprising. The existing conflict between versions of python strikes me as more of a planning problem than a software bug. It may be naive, but I don't see why python37 and python38 can't use distinct names for files placed in system directories. It's rather more peculiar that deinstalling python37 does not solve the problem. Nor does deleting the offending file (link). Still, poudriere seems a huge hammer for a small tack. Thanks for reading, bob prohaska
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20210517155515.GA77731>