Date: Tue, 09 Jun 2009 11:34:35 -0700 From: Doug Barton <dougb@FreeBSD.org> To: Wesley Shields <wxs@FreeBSD.org> Cc: cvs-ports@FreeBSD.org, cvs-all@FreeBSD.org, ports-committers@FreeBSD.org Subject: Re: cvs commit: ports UPDATING Message-ID: <4A2EAB3B.6040205@FreeBSD.org> In-Reply-To: <20090609180020.GA40405@atarininja.org> References: <200906091346.n59Dk3vj093458@repoman.freebsd.org> <4A2EA0A1.6010607@FreeBSD.org> <20090609180020.GA40405@atarininja.org>
next in thread | previous in thread | raw e-mail | index | archive | help
Wesley Shields wrote:
> The problem was not with portmaster,
That's always good news. :)
> but was with the reliance on
> pkg_which. The attached patch should fit it. I currently have sent it
> to itetcu and miwi for review. The biggest downside is that pkg_info can
> not take multiple arguments on the command line like pkg_which does, so
> we have to use xargs -L 1, which makes things slow compared to using
> pkg_which.
Have you benchmarked that? I would imagine (although I don't know for
sure) that pkg_which is just calling whatever method it uses to get
the answer once each time for each command line argument, which would
mean that they are basically equivalent.
It seems that the current solution is needlessly complicated (although
I haven't benchmarked either). What would be wrong with something like:
for file in `find <pythondir> -type f`; do
pkg_info -qW $file
done | sort -u
Doug
--
This .signature sanitized for your protection
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4A2EAB3B.6040205>