Subject: Re: pkgsrc NetBSD 2.0/i386 bulk build results 2005-04-01
To: Thomas Klausner <wiz@NetBSD.org>
From: Jim Wise <jwise@draga.com>
List: pkgsrc-bulk
Date: 04/04/2005 11:33:16
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
On Mon, 4 Apr 2005, Thomas Klausner wrote:
>37a35,36
>> databases/jdbc-postgresql jwise@NetBSD.org
>
> Error loading class org.apache.tools.ant.Main: Bad major version number
>
>> databases/jdbc-postgresql7 jwise@NetBSD.org
>
> Error loading class org.apache.tools.ant.Main: Bad major version number
>39a39
Both of these look like an apache-ant problem (see the error messages
about org.apache.tools.ant.Main failing to load). In particular, they
look like the bulk build is trying to run an ant built with a modern
java compiler against the ancient native pkgsrc jdk. You may want to
ping the ant MAINTAINER about these -- if ant won't run, then packages
built using ant aren't going to build, after all.
>120a122
>> www/ap-jk jwise@NetBSD.org
>
> /bin/sh ../libtool ../libtool --mode=compile cc -I/usr/pkg/include/httpd -O2 -I/usr/pkg/include -I/usr/include -D_LARGEFILE_SOURC
> libtool: warning: cannot infer operation mode from `../libtool'
This was updated this last week by erh, who I am pulling into the CC on
this mail. That things like this can break -- and the MAINTAINER be
spammed with bulk-build notices -- is another reason to check with
MAINTAINER before updating a pkg.
erh: do you want to take over MAINTAINERship of this package? I'm
unlikely to get to it again soon, while you seem interested in it...
wiz: doesn't the high rate at which packages failing to build turns out
to be actually caused by other packages malfunctioning suggest that
spamming the MAINTAINER with each bulk build result may be a bad idea?
Just my $0.02 USD...
- --
Jim Wise
jwise@draga.com
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.1 (NetBSD)
iD8DBQFCUV5ApRpI6SYACmIRAhd+AJ9CgxfHF/luS3uSe6YXL8XhiR6UjACfaX7S
VbxQ6MYEjoSl+oesrcI0W5E=
=KVd9
-----END PGP SIGNATURE-----