Date: Tue, 22 Jul 2014 19:43:45 +0200 From: Antoine Brodin <antoine@FreeBSD.org> To: freebsd-ports <freebsd-ports@freebsd.org> Cc: "portmgr@freebsd.org Management Team" <portmgr@freebsd.org>, Nick Hilliard <nick@foobar.org> Subject: Re: mass commit needed: python breakage in pkgng repo Message-ID: <CAALwa8=o0a%2Bv5OB7kFyZ_soeMkjt1e7KC3Nw=rD-yR7Uqj1Fvg@mail.gmail.com> In-Reply-To: <53CE9AD6.6040201@gmx.de> References: <53CE846E.8090800@foobar.org> <53CE9AD6.6040201@gmx.de>
next in thread | previous in thread | raw e-mail | index | archive | help
On Tue, Jul 22, 2014 at 7:09 PM, olli hauer <ohauer@gmx.de> wrote: >>>>On 2014-07-22 17:34, Nick Hilliard wrote:> sysutils/py-salt has been broken for the last couple of weeks. The >>>>> resulting package didn't include the python egg info directory and as a >>>>> result, py-salt crashed on startup. There a PR in bugzilla for this: >>>>> >>>>> https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=191986 >>>>> >>>>>>From what I can tell, this problem has been fixed by this commit: >>>>> >>>>> http://svnweb.freebsd.org/ports?view=revision&revision=362364 >>>>> >>>>> So py-salt will need to be rebuilt in the freebsd pkgng repo, but because >>>>> the fix was in bsd.python.mk, the port will need to have PORTREVISION >>>>> bumped in order to flush out the broken version from the pkgng repo. >>>>> >>>>> I've attached a list of other ports which may also be broken (i.e. ports >>>>> whose makefiles include the text USE_PYDISTUTILS but not >>>>> PYDISTUTILS_AUTOPLIST). Some of these ports may also need to rebuilt, but >>>>> without testing each one of them individually, it may not be be possible to >>>>> tell which. >>>>> >>>>> Could someone consider doing a mass commit to each of these ports to bump >>>>> PORTREVISION so that any potential breakage is flushed out of the pkg repos? >>>>> > > Sorry, my mobile has taken the discussion out of list ... > >>>>> isn't it easier to bump python instead ? >>>> yep sure would, but that will force a rebuild of ~2700 ports instead of 300. > >>> I suspect even more then the 2300 ports (depening ports also counting) but >>> this way really everything will be cached. >>> Dont know if the snap for weekly build was already taken if not it would be >>> an option >> >> tbh, i'm not familiar enough with the main repo build process + >> consequences to be able to assess how best to deal with this. >> > > Best to address this also to portmgr@ for decicission (added to CC) Hi, py-setuptools was updated on Wednesday 16th so all ports depending on it will be rebuilt during next bulk on the pkg builders. But I can still bump PORTREVISION on py-salt as for this specific port the missing egginfo is harmful? (for most other ports it's harmless) Cheers, Antoine
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CAALwa8=o0a%2Bv5OB7kFyZ_soeMkjt1e7KC3Nw=rD-yR7Uqj1Fvg>