Subject: Re: pkg/36662: pkgtools/pkg_install's audit_packages appears to be platform-specific
To: None <joerg@NetBSD.org, gnats-admin@netbsd.org, pkgsrc-bugs@netbsd.org,>
From: Stuart Shelton <srcshelton@gmail.com>
List: pkgsrc-bugs
Date: 07/23/2007 11:05:06
The following reply was made to PR pkg/36662; it has been noted by GNATS.
From: "Stuart Shelton" <srcshelton@gmail.com>
To: gnats-bugs@netbsd.org
Cc: joerg@netbsd.org, gnats-admin@netbsd.org, pkgsrc-bugs@netbsd.org
Subject: Re: pkg/36662: pkgtools/pkg_install's audit_packages appears to be platform-specific
Date: Mon, 23 Jul 2007 12:00:14 +0100
Still two outstanding issues:
* <err.h> is still being unconditionally included. This is wrong. It
should be replaced with a conditional include from nbcompat/, as with
sha2.h a few lines below.
* There's something screwey in nbcompat/err.h, which breaks IRIX'
cdefs.h - and cdefs.h must be included in order to get the __P macro.
So <sys/cdefs.h> must be included *before* <nbcompat/err.h>
Patch:
--- audit-packages/audit-packages.c.dist 2007-07-23
11:16:00.420162080 +0100
+++ audit-packages/audit-packages.c 2007-07-23 11:49:27.043021280 +0100
@@ -44,7 +44,12 @@
#ifdef HAVE_INTTYPES_H
#include <stdint.h>
#endif
+#include <sys/cdefs.h>
+#ifdef PKGSRC
+#include <nbcompat/err.h>
+#else
#include <err.h>
+#endif
#include <string.h>
#include <sys/types.h>
#include <sys/stat.h>
Cheers,
Stuart
On 21/07/07, Adrian Portelli <adrianp@stindustries.net> wrote:
> The following reply was made to PR pkg/36662; it has been noted by GNATS.
>
> From: Adrian Portelli <adrianp@stindustries.net>
> To: gnats-bugs@NetBSD.org
> Cc:
> Subject: Re: pkg/36662: pkgtools/pkg_install's audit_packages appears to be
> platform-specific
> Date: Sat, 21 Jul 2007 16:04:07 +0100
>
> Hi,
>
> Could you please do a 'cvs update' to make sure you have all the latest
> changes and then apply the following patch:
>
> http://www.stindustries.net/NetBSD/stuff/pkg_install.diff
>
> (cd ${PREFIX} && cat pkg_install.diff | patch -p0)
>
> This should fix the remainder of the issues in this PR.
>
> thanks,
>
> adrian.
>
>