Subject: Re: CVS commit: pkgsrc/net/mtr
To: NetBSD Packages Technical Discussion List <tech-pkg@NetBSD.ORG>
From: Johnny Lam <firstname.lastname@example.org>
Date: 10/25/2001 14:32:55
On Thu, Oct 25, 2001 at 04:56:56PM -0400, Greg A. Woods wrote:
> [ On Thursday, October 25, 2001 at 15:15:35 (-0400), Jim Wise wrote: ]
> > It is a standing policy of pkgsrc that when you patch configure.in, you
> > regen configure, and commit a patch to configure based on the result.
> That's not borne out by the evidence in pkgsrc today.
> That might be how you think it should work, but it's definitely not how
> I think it should work. Besides being a direct violation of the "never
> commit a generated file" rule, it's also an illogical and unnecessary
> complication to the pkgsrc patches.
There is no such rule to which you are alluding. What Jim expressed is
actually the current policy in pkgsrc, and I expect to see more packages
migrated to do this.
> > it also, as pointed out at the beginning of this
> > thread, saves people on slow machines from endless and meaningless
> > installations and invocations of autoconf and automake.
> 1) submit your pkgsrc changes to the package's build procedure back to
> the author
We do this anyway, but we have to wait for the next release of their
software to see our changes incorporated. It's reasonable to commit
both sets of changes, to the source as well as to the generated file,
in the meantime.
> 2) bite the bullet and eat the cost of running auto* with every "make"
> until the package's author accepts the changes (they really are not
> that expensive to run, especially not compared to the cost of
> actually running the generated configure script!)
I think it's because installing automake requires installing a lot of
cruft which isn't really required to build software. You're not generally
required to have autoconf/automake around when you build non-pkgsrc
software, and it pains some people (including myself) to need it for
pkgsrc software. Trying to regularly use a mac68k machine where compiling
perl is a pretty long task made me thing this way.
> Do not second guess auto* and do not assume that further changes might
> not be made by local pkgsrc patches.
I can't parse what you're saying. We patch a lot of files. Some of those
changes are very difficult to do in the configure.in but very easy to do
in the configure script, so we do it the way that is most appropriate.
-- Johnny Lam <email@example.com>