Re: problem with git-pull
- In reply to: Jan Beich : "Re: problem with git-pull"
- Go to: [ bottom of page ] [ top of archives ] [ this month ]
Date: Thu, 14 Sep 2023 09:27:45 UTC
On Thu, 14 Sep 2023 00:10:58 +0200 Jan Beich <jbeich@FreeBSD.org> wrote: > Martin Neubauer <fbsd@herrneubauer.de> writes: > > > On Wed, 13 Sep 2023, Matthias Apitz wrote: > > > >> El día miércoles, septiembre 13, 2023 a las 12:55:20 +0200, Kurt Jaeger escribió: > >> > >>> Hi! > >>> > >>>> In short: Is there no way with git to pull only one special port > >>>> for a recompilation? > >>> > >>> For things like that I always have a git ports tree besides the > >>> poudriere ports tree. If I need one little recompile, I > >>> update the non-poudriere tree, take the one change from that tree > >>> that I need, put it into the poudriere and re-run poudriere... > >>> > >>> There might be easier ways to achieve the same, but this works. > >> > >> I updated, first, the ports tree below /usr/ports while poudriere > >> was using /usr/local/poudriere/ports/ports20230806 (as defined when > >> creating the port within poudriere). I was faulty thinking that both > >> are physically the same place and /usr/local/poudriere/ports/ports20230806 > >> only a mount point of /usr/ports. > >> > >> The fist 'git pull' did nothimng within poudriere, the other recompiled > >> some ports which I did not wanted. I should have done a temp. > >> copy of /usr/ports/security/wpa_supplicant to > >> /usr/local/poudriere/ports/ports20230806/security/wpa_supplicant > > You might want to have a look at the -C switch (and maybe -S) to > > poudriere as well. It's not without pitfalls, but might be a viable > > option for the particular situation you described. > > -S switch is kinda broken per https://github.com/freebsd/poudriere/issues/806 > See also https://lists.freebsd.org/archives/freebsd-ports/2023-January/003292.html And -S tracks dependencies isufficiently (maybe direct dependencies only), causing build order reversal, thus causing failed build which can usually done on next run. -- Tomoaki AOKI <junchoon@dec.sakura.ne.jp>