[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in net-print/cups/files: cups-1.5.0-systemd-socket-2.patch

[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in net-print/cups/files: cups-1.5.0-systemd-socket-2.patch

[Samuli Suominen]

On 11/04/13 14:05, Fabio Erculiani (lxnay) wrote:
> lxnay       13/04/11 11:05:05
>
>    Modified:             cups-1.5.0-systemd-socket-2.patch
>    Log:
>    fix systemd units path variable setup, see bug #465474
>
>    (Portage version: 2.2.0_alpha166/cvs/Linux x86_64, signed Manifest commit with key ADC916E5)
>
> Revision  Changes    Path
> 1.2                  net-print/cups/files/cups-1.5.0-systemd-socket-2.patch
>
> file : http://sources.gentoo.org/viewvc.cgi/gentoo-x86/net-print/cups/files/cups-1.5.0-systemd-socket-2.patch?rev=1.2&view=markup
> plain: http://sources.gentoo.org/viewvc.cgi/gentoo-x86/net-print/cups/files/cups-1.5.0-systemd-socket-2.patch?rev=1.2&content-type=text/plain
> diff : http://sources.gentoo.org/viewvc.cgi/gentoo-x86/net-print/cups/files/cups-1.5.0-systemd-socket-2.patch?r1=1.1&r2=1.2
>
> Index: cups-1.5.0-systemd-socket-2.patch
> ===================================================================
> RCS file: /var/cvsroot/gentoo-x86/net-print/cups/files/cups-1.5.0-systemd-socket-2.patch,v
> retrieving revision 1.1
> retrieving revision 1.2
> diff -u -r1.1 -r1.2
> --- cups-1.5.0-systemd-socket-2.patch	10 Apr 2013 16:39:19 -0000	1.1
> +++ cups-1.5.0-systemd-socket-2.patch	11 Apr 2013 11:05:05 -0000	1.2
> @@ -20,7 +20,7 @@
>   +++ cups-1.5.0/config-scripts/cups-systemd.m4	2011-10-18 15:33:16.861989058 +0100
>   @@ -0,0 +1,36 @@
>   +dnl
> -+dnl "$Id: cups-1.5.0-systemd-socket-2.patch,v 1.1 2013/04/10 16:39:19 lxnay Exp $"
> ++dnl "$Id: cups-1.5.0-systemd-socket-2.patch,v 1.2 2013/04/11 11:05:05 lxnay Exp $"
>   +dnl
>   +dnl   systemd stuff for CUPS.
>   +
> @@ -32,11 +32,11 @@
>   +        [], [with_systemdsystemunitdir=$($PKGCONFIG --variable=systemdsystemunitdir systemd)])
>   +if test "x$with_systemdsystemunitdir" != xno; then
>   +        AC_MSG_CHECKING(for libsystemd-daemon)
> ++	AC_SUBST([systemdsystemunitdir], [$with_systemdsystemunitdir])
>   +        if $PKGCONFIG --exists libsystemd-daemon; then
>   +		AC_MSG_RESULT(yes)
>   +        	SDCFLAGS=`$PKGCONFIG --cflags libsystemd-daemon`
>   +        	SDLIBS=`$PKGCONFIG --libs libsystemd-daemon`
> -+		AC_SUBST([systemdsystemunitdir], [$with_systemdsystemunitdir])
>   +		AC_DEFINE(HAVE_SYSTEMD)
>   +	else
>   +		AC_MSG_RESULT(no)
> @@ -53,7 +53,7 @@
>   +AC_SUBST(SDLIBS)
>   +
>   +dnl
> -+dnl "$Id: cups-1.5.0-systemd-socket-2.patch,v 1.1 2013/04/10 16:39:19 lxnay Exp $"
> ++dnl "$Id: cups-1.5.0-systemd-socket-2.patch,v 1.2 2013/04/11 11:05:05 lxnay Exp $"

^ This will likely cause the patch not to apply, at least with older 
patch versions
You should be able to delete this section of the patch to avoid the CVS 
tag polluting it

[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in net-print/cups/files: cups-1.5.0-systemd-socket-2.patch

[Ulrich Mueller]

>>>>> On Thu, 11 Apr 2013, Samuli Suominen wrote:

>> -+dnl "$Id: cups-1.5.0-systemd-socket-2.patch,v 1.1 2013/04/10 16:39:19 lxnay Exp $"
>> ++dnl "$Id: cups-1.5.0-systemd-socket-2.patch,v 1.2 2013/04/11 11:05:05 lxnay Exp $"

> ^ This will likely cause the patch not to apply, at least with older 
> patch versions
> You should be able to delete this section of the patch to avoid the CVS 
> tag polluting it

Or use "cvs add -ko" to suppress keyword expansion.

Ulrich

Re: [gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in net-print/cups/files: cups-1.5.0-systemd-socket-2.patch

[Rich Freeman]

On Thu, Apr 11, 2013 at 7:07 AM, Samuli Suominen <ssuominen@gentoo.org> wrote:
> ^ This will likely cause the patch not to apply, at least with older patch
> versions
> You should be able to delete this section of the patch to avoid the CVS tag
> polluting it

The CVS tags will also create issues during the git migration - we
have a bunch of ancient commits that get mangled during conversion
largely due to keywords in patch files.  I believe -ko will also fix
those, but if you don't need the keyword in the patch I'd just avoid
it entirely.  Brian has been working wonders with fixing as many as
possible, but let's not keep adding more (this one will still cause
problems since it will be in the commit history even if you fix it -
though I suspect in the end we'll just have to live with stuff like
this frozen in our history since there really isn't any one-right-way
to handle them).

IMHO CVS keywords are right up there with 2600Hz when it comes to
in-band signaling mechanisms that have unforeseen consequences.

Rich