Re: Specs using %define

[Eduardo Mayorga Téllez]

El jue, 24-12-2015 a las 15:01 -0600, Jason L Tibbitts III escribió:
> ngspice (chitlesh, mayorga)

Fix pushed and built in Rawhide.

Eduardo
--
devel mailing list
devel@lists.fedoraproject.org
http://lists.fedoraproject.org/admin/lists/devel@lists.fedoraproject.org

Re: Specs using %define

[Tom Callaway]

On 12/24/2015 04:01 PM, Jason L Tibbitts III wrote:
> To satisfy my curiosity, I grepped the convenient tarball of specfiles
> (http://pkgs.fedoraproject.org/repo/rpm-specs-latest.tar.xz) for lines
> matching "(? there were more than 1900 hits.
> 
> Here's a complete (long) list.  I don't think there's much point in
> doing anything about the vast majority of these, but if you're in
> cleaning up your packages then why not?  Also, feel free to let me know
> if any of these are false positives.  (They may be due to %define in the
> changelog or in a "comment", which is also something that should be
> fixed.)  And if you're using %define because you actually need one of
> the peculiar behaviors it exhibits, please let me know because I'd like
> to see an actual example of that.
> 
> (Own goal notice: I have a few packages on this list.  Off to fix them
> up now.)

Cleaned up the 50 packages on this list that I own.

~tom

==
Red Hat
--
devel mailing list
devel@lists.fedoraproject.org
http://lists.fedoraproject.org/admin/lists/devel@lists.fedoraproject.org

Re: Specs using %define

[Nils Philippsen]

On Thu, 2015-12-24 at 15:01 -0600, Jason L Tibbitts III wrote:
> ardour4 (nphilipp)
> gegl (deji, nphilipp)
> gimp-data-extras (nphilipp)
> gimp-help (nphilipp)
> glade2 (nphilipp, alexl, caillon, caolanm, johnp, mbarnes, rhughes, rstrode, 
> ssp, group::gnome-sig)
> python-augeas (xaeth, nphilipp)
> rss-glx (nphilipp, cheese)
> sane-backends (nphilipp)
> supervisor (nphilipp, arg)
> ufraw (nphilipp)

fixed in git

Nils
-- 
Nils Philippsen  "Those who would give up Essential Liberty to purchase 
Red Hat   a little Temporary Safety, deserve neither Liberty
n...@redhat.com   nor Safety."  --  Benjamin Franklin, 1759
PGP fingerprint:  C4A8 9474 5C4C ADE3 2B8F  656D 47D8 9B65 6951 3011

--
devel mailing list
devel@lists.fedoraproject.org
http://lists.fedoraproject.org/admin/lists/devel@lists.fedoraproject.org

Re: Specs using %define

[Richard Fearn]

> eclipse-findbugs (richardfearn)

Fixed - thanks!

Richard

-- 
Richard Fearn
richardfe...@gmail.com
--
devel mailing list
devel@lists.fedoraproject.org
http://lists.fedoraproject.org/admin/lists/devel@lists.fedoraproject.org

Re: Specs using %define

[Jakub Jelen]

On 12/24/2015 10:01 PM, Jason L Tibbitts III wrote:

To satisfy my curiosity, I grepped the convenient tarball of specfiles
(http://pkgs.fedoraproject.org/repo/rpm-specs-latest.tar.xz) for lines
matching "(?After fixing %define -> %global in my spec file, my vim is having 
several problems:


 * First of them is missing syntax highlight for %define macro - bug 
from 2014 [1]. Patch fixes it.
 * Second is c sequence for generating changelog record stopped 
working. Reported as [2] with local patch.


CCing vim maintainer. Can we have it in F23?

[1] https://bugzilla.redhat.com/show_bug.cgi?id=1058041
[2] https://bugzilla.redhat.com/show_bug.cgi?id=1297746

Greetings,

--
Jakub Jelen
Security Technologies
Red Hat
--
devel mailing list
devel@lists.fedoraproject.org
http://lists.fedoraproject.org/admin/lists/devel@lists.fedoraproject.org

Re: Specs using %define

[Ankur Sinha]

On Thu, 2015-12-24 at 15:01 -0600, Jason L Tibbitts III wrote:
> To satisfy my curiosity, I grepped the convenient tarball of
> specfiles
> (http://pkgs.fedoraproject.org/repo/rpm-specs-latest.tar.xz) for line
> s
> matching "(? there were more than 1900 hits.
> 
> Here's a complete (long) list.  I don't think there's much point in
> doing anything about the vast majority of these, but if you're in
> cleaning up your packages then why not?  Also, feel free to let me
> know
> if any of these are false positives.  (They may be due to %define in
> the
> changelog or in a "comment", which is also something that should be
> fixed.)  And if you're using %define because you actually need one of
> the peculiar behaviors it exhibits, please let me know because I'd
> like
> to see an actual example of that.
> 

> cf-bonveno-fonts (ankursinha)

Fixed. 



> gtg (ankursinha)

Fixed.


> oflb-notcouriersans-fonts (ankursinha)

Fixed.

> 
> oldstandard-sfd-fonts (ankursinha)

Fixed.


> subtitleeditor (mso, ankursinha)

Fixed.

-- 
Thanks,
Regards,
Ankur Sinha "FranciscoD"

http://fedoraproject.org/wiki/User:Ankursinha




signature.asc
Description: This is a digitally signed message part
--
devel mailing list
devel@lists.fedoraproject.org
http://lists.fedoraproject.org/admin/lists/devel@lists.fedoraproject.org

Re: Specs using %define

[Louis Lagendijk]

On Thu, 2015-12-24 at 15:01 -0600, Jason L Tibbitts III wrote:
> cups-bjnp
Is fixed and built in Rawhide, F22/F23 are fixed but not built.
Louid
--
devel mailing list
devel@lists.fedoraproject.org
http://lists.fedoraproject.org/admin/lists/devel@lists.fedoraproject.org

Re: Specs using %define

[Richard W.M. Jones]

On Tue, Jan 05, 2016 at 09:15:29PM +0200, Panu Matilainen wrote:
> On 01/05/2016 07:16 PM, Richard W.M. Jones wrote:
> >On Thu, Dec 24, 2015 at 03:01:02PM -0600, Jason L Tibbitts III wrote:
> >>coccinelle (rjones)
> >
> >coccinelle.spec has:
> >
> >%{!?python_sitelib: %define python_sitelib %(%{__python} -c "from 
> >distutils.sysconfig import get_python_lib; print get_python_lib()")}
> >
> >TBH I have no idea if this is correct or not, or perhaps should be
> >completely removed.  Thoughts?
> 
> That's one of those cases where using %define is actually *wrong*
> because of the macro scoping thing. But in practise it'll never get
> used since these days %python_sitelib is always defined so its not
> just wrong but also pointless as well.
> 
> >>mingw-freetype (rjones, lfarkas, epienbro)
> >
> >This uses:
> >
> >%{!?_with_subpixel_rendering: %{!?_without_subpixel_rendering: %define 
> >_without_subpixel_rendering --without-subpixel_rendering}}
> >
> >_without_subpixel_rendering is not used anywhere else in the file.
> >No idea if that is right or not.
> 
> Another case where %define is actually wrong. The whole construct
> looks like a workaround for %bcond related misunderstanding, but
> dunno.

Thanks Panu.  I think I will delete both lines, and see if we
have any problems.

Rich.

-- 
Richard Jones, Virtualization Group, Red Hat http://people.redhat.com/~rjones
Read my programming and virtualization blog: http://rwmj.wordpress.com
virt-top is 'top' for virtual machines.  Tiny program with many
powerful monitoring features, net stats, disk stats, logging, etc.
http://people.redhat.com/~rjones/virt-top
--
devel mailing list
devel@lists.fedoraproject.org
http://lists.fedoraproject.org/admin/lists/devel@lists.fedoraproject.org

Re: Specs using %define

[Richard W.M. Jones]

On Wed, Jan 06, 2016 at 11:54:06AM +, Richard W.M. Jones wrote:
> On Tue, Jan 05, 2016 at 09:15:29PM +0200, Panu Matilainen wrote:
> > On 01/05/2016 07:16 PM, Richard W.M. Jones wrote:
> > >On Thu, Dec 24, 2015 at 03:01:02PM -0600, Jason L Tibbitts III wrote:
> > >>mingw-freetype (rjones, lfarkas, epienbro)
> > >
> > >This uses:
> > >
> > >%{!?_with_subpixel_rendering: %{!?_without_subpixel_rendering: %define 
> > >_without_subpixel_rendering --without-subpixel_rendering}}
> > >
> > >_without_subpixel_rendering is not used anywhere else in the file.
> > >No idea if that is right or not.
> > 
> > Another case where %define is actually wrong. The whole construct
> > looks like a workaround for %bcond related misunderstanding, but
> > dunno.

Actually in this second case, mingw-freetype.spec is just following
the freetype.spec file.  Which also looks wrong because it defines
_without_subpixel_rendering which is never used anywhere.

http://pkgs.fedoraproject.org/cgit/rpms/freetype.git/tree/freetype.spec

Adding CC freetype-owner.

Rich.

-- 
Richard Jones, Virtualization Group, Red Hat http://people.redhat.com/~rjones
Read my programming and virtualization blog: http://rwmj.wordpress.com
virt-builder quickly builds VMs from scratch
http://libguestfs.org/virt-builder.1.html
--
devel mailing list
devel@lists.fedoraproject.org
http://lists.fedoraproject.org/admin/lists/devel@lists.fedoraproject.org

Re: Specs using %define

[Panu Matilainen]

On 01/06/2016 01:59 PM, Richard W.M. Jones wrote:

On Wed, Jan 06, 2016 at 11:54:06AM +, Richard W.M. Jones wrote:

On Tue, Jan 05, 2016 at 09:15:29PM +0200, Panu Matilainen wrote:

On 01/05/2016 07:16 PM, Richard W.M. Jones wrote:

On Thu, Dec 24, 2015 at 03:01:02PM -0600, Jason L Tibbitts III wrote:

mingw-freetype (rjones, lfarkas, epienbro)


This uses:

%{!?_with_subpixel_rendering: %{!?_without_subpixel_rendering: %define 
_without_subpixel_rendering --without-subpixel_rendering}}

_without_subpixel_rendering is not used anywhere else in the file.
No idea if that is right or not.


Another case where %define is actually wrong. The whole construct
looks like a workaround for %bcond related misunderstanding, but
dunno.


Actually in this second case, mingw-freetype.spec is just following
the freetype.spec file.  Which also looks wrong because it defines
_without_subpixel_rendering which is never used anywhere.

http://pkgs.fedoraproject.org/cgit/rpms/freetype.git/tree/freetype.spec


Right. Freetype is old enough to actually predate rpm's 
%bcond_with/without so custom variants have been implemented + evolved 
by various packagers over the years, but this should've really been 
using %bcond already:

http://pkgs.fedoraproject.org/cgit/rpms/freetype.git/commit/freetype.spec?id=417715cf8e484f8099e6c2561e1ba83e851d9751

This is what it should be changed to (untested but should be "obviously 
correct"):


diff --git a/freetype.spec b/freetype.spec
index c6d492d..e479acf 100644
--- a/freetype.spec
+++ b/freetype.spec
@@ -1,6 +1,6 @@
 # Patented subpixel rendering disabled by default.
 # Pass '--with subpixel_rendering' on rpmbuild command-line to enable.
-%{!?_with_subpixel_rendering: %{!?_without_subpixel_rendering: %define 
_without_subpixel_rendering --without-subpixel_rendering}}

+%bond_with subpixel_rendering

 %{!?with_xfree86:%define with_xfree86 1}

@@ -37,7 +37,7 @@ BuildRequires: zlib-devel
 BuildRequires: bzip2-devel

 Provides: %{name}-bytecode
-%if %{?_with_subpixel_rendering:1}%{!?_with_subpixel_rendering:0}
+%if %{with subpixel_rendering}
 Provides: %{name}-subpixel
 %endif

@@ -78,7 +78,7 @@ FreeType.
 %prep
 %setup -q -b 1 -a 2

-%if %{?_with_subpixel_rendering:1}%{!?_with_subpixel_rendering:0}
+%if %{with subpixel_rendering}
 %patch21  -p1 -b .enable-spr
 %endif


If this seems backwards, see 
http://rpm.org/wiki/PackagerDocs/ConditionalBuilds


- Panu -





Adding CC freetype-owner.

Rich.


--
devel mailing list
devel@lists.fedoraproject.org
http://lists.fedoraproject.org/admin/lists/devel@lists.fedoraproject.org

Re: Specs using %define

[Richard W.M. Jones]

On Wed, Jan 06, 2016 at 02:28:45PM +0200, Panu Matilainen wrote:
> On 01/06/2016 01:59 PM, Richard W.M. Jones wrote:
> >On Wed, Jan 06, 2016 at 11:54:06AM +, Richard W.M. Jones wrote:
> >>On Tue, Jan 05, 2016 at 09:15:29PM +0200, Panu Matilainen wrote:
> >>>On 01/05/2016 07:16 PM, Richard W.M. Jones wrote:
> On Thu, Dec 24, 2015 at 03:01:02PM -0600, Jason L Tibbitts III wrote:
> >mingw-freetype (rjones, lfarkas, epienbro)
> 
> This uses:
> 
> %{!?_with_subpixel_rendering: %{!?_without_subpixel_rendering: %define 
> _without_subpixel_rendering --without-subpixel_rendering}}
> 
> _without_subpixel_rendering is not used anywhere else in the file.
> No idea if that is right or not.
> >>>
> >>>Another case where %define is actually wrong. The whole construct
> >>>looks like a workaround for %bcond related misunderstanding, but
> >>>dunno.
> >
> >Actually in this second case, mingw-freetype.spec is just following
> >the freetype.spec file.  Which also looks wrong because it defines
> >_without_subpixel_rendering which is never used anywhere.
> >
> >http://pkgs.fedoraproject.org/cgit/rpms/freetype.git/tree/freetype.spec
> 
> Right. Freetype is old enough to actually predate rpm's
> %bcond_with/without so custom variants have been implemented +
> evolved by various packagers over the years, but this should've
> really been using %bcond already:
> http://pkgs.fedoraproject.org/cgit/rpms/freetype.git/commit/freetype.spec?id=417715cf8e484f8099e6c2561e1ba83e851d9751
> 
> This is what it should be changed to (untested but should be
> "obviously correct"):
> 
> diff --git a/freetype.spec b/freetype.spec
> index c6d492d..e479acf 100644
> --- a/freetype.spec
> +++ b/freetype.spec
> @@ -1,6 +1,6 @@
>  # Patented subpixel rendering disabled by default.
>  # Pass '--with subpixel_rendering' on rpmbuild command-line to enable.
> -%{!?_with_subpixel_rendering: %{!?_without_subpixel_rendering:
> %define _without_subpixel_rendering --without-subpixel_rendering}}
> +%bond_with subpixel_rendering
> 
>  %{!?with_xfree86:%define with_xfree86 1}
> 
> @@ -37,7 +37,7 @@ BuildRequires: zlib-devel
>  BuildRequires: bzip2-devel
> 
>  Provides: %{name}-bytecode
> -%if %{?_with_subpixel_rendering:1}%{!?_with_subpixel_rendering:0}
> +%if %{with subpixel_rendering}
>  Provides: %{name}-subpixel
>  %endif
> 
> @@ -78,7 +78,7 @@ FreeType.
>  %prep
>  %setup -q -b 1 -a 2
> 
> -%if %{?_with_subpixel_rendering:1}%{!?_with_subpixel_rendering:0}
> +%if %{with subpixel_rendering}
>  %patch21  -p1 -b .enable-spr
>  %endif
> 
> 
> If this seems backwards, see
> http://rpm.org/wiki/PackagerDocs/ConditionalBuilds

Thanks - I'll follow what freetype-owner decides, because we want to
try to keep mingw packaging as close to the native packages as
possible.

Rich.

-- 
Richard Jones, Virtualization Group, Red Hat http://people.redhat.com/~rjones
Read my programming and virtualization blog: http://rwmj.wordpress.com
virt-p2v converts physical machines to virtual machines.  Boot with a
live CD or over the network (PXE) and turn machines into KVM guests.
http://libguestfs.org/virt-v2v
--
devel mailing list
devel@lists.fedoraproject.org
http://lists.fedoraproject.org/admin/lists/devel@lists.fedoraproject.org

Re: Specs using %define

[Kevin Kofler]

Panu Matilainen wrote:
> Right. Freetype is old enough to actually predate rpm's
> %bcond_with/without so custom variants have been implemented + evolved
> by various packagers over the years, but this should've really been
> using %bcond already:

IMHO, that conditional should just be dropped entirely. I build a freetype-
freeworld in RPM Fusion which has subpixel rendering enabled and which 
overrides the libraries through ld.so.conf.d, which is IMHO a much better 
solution than having the user rebuild the freetype RPM with a non-default 
flag.

And I don't need this conditional for freetype-freeworld. I actually removed 
it entirely from freetype-freeworld, because now that the hinting patents 
expired, there are only 2 ways of building Freetype: with or without 
subpixel rendering. freetype is without, freetype-freeworld is with. So I 
just hardcoded it to always enabed in freetype-freeworld, which simplified 
my specfile.

Kevin Kofler
--
devel mailing list
devel@lists.fedoraproject.org
http://lists.fedoraproject.org/admin/lists/devel@lists.fedoraproject.org

Re: Specs using %define

[Daiki Ueno]

Jason L Tibbitts III  writes:

> baekmuk-bdf-fonts (ueno)
> baekmuk-ttf-fonts (ueno)
> ibus-gucharmap (ueno)
> un-core-fonts (ueno)
> un-extra-fonts (ueno)

Done in rawhide.

Regards,
-- 
Daiki Ueno
--
devel mailing list
devel@lists.fedoraproject.org
http://lists.fedoraproject.org/admin/lists/devel@lists.fedoraproject.org

Re: Specs using %define

[Richard W.M. Jones]

On Thu, Dec 24, 2015 at 03:01:02PM -0600, Jason L Tibbitts III wrote:
> coccinelle (rjones)

coccinelle.spec has:

%{!?python_sitelib: %define python_sitelib %(%{__python} -c "from 
distutils.sysconfig import get_python_lib; print get_python_lib()")}

TBH I have no idea if this is correct or not, or perhaps should be
completely removed.  Thoughts?

> mingw-atk (rjones, epienbro, kalev)

Fixed.

> mingw-freetype (rjones, lfarkas, epienbro)

This uses:

%{!?_with_subpixel_rendering: %{!?_without_subpixel_rendering: %define 
_without_subpixel_rendering --without-subpixel_rendering}}

_without_subpixel_rendering is not used anywhere else in the file.
No idea if that is right or not.

> mingw-gdk-pixbuf (epienbro, rjones, sailer)
> mingw-glib2 (fidencio, rjones, epienbro, lfarkas, sailer, kalev, elmarco)
> mingw-glibmm24 (sailer, elmarco, rjones)
> mingw-libgcrypt (rjones, epienbro)
> mingw-liboil (lfarkas, rjones)
> mingw-libsoup (epienbro, rjones)
> mingw-pangomm (sailer, rjones)
> mingw-pango (rjones, epienbro, kalev)

All done.

> mingw-libvirt (berrange, rjones, elmarco)

I didn't feel confident in changing this one.  I compared the
mingw-libvirt.spec file and ordinary libvirt.spec, and both use
%define extensively for with_* settings.  I assume there must be some
reason for that.

> nekovm (rjones, andyli)

Done.

> ocaml-ancient (rjones)
> ocamldsort (rjones)
> ocaml-omake (rjones)

Done.

> qemu (jforbes, dwmw2, berrange, quintela, ehabkost, amitshah, alon, rjones, 
> bonzini, crobinso)

I was told off the last time I broke qemu.spec, so I didn't
modify this :-)  It uses %define 7 times.

> unison213 (rjones, gemi)
> unison227 (rjones, gemi, brummbq)
> unison240 (brummbq, rjones)

All done.

Rich.

-- 
Richard Jones, Virtualization Group, Red Hat http://people.redhat.com/~rjones
Read my programming and virtualization blog: http://rwmj.wordpress.com
virt-builder quickly builds VMs from scratch
http://libguestfs.org/virt-builder.1.html
--
devel mailing list
devel@lists.fedoraproject.org
http://lists.fedoraproject.org/admin/lists/devel@lists.fedoraproject.org

Re: Specs using %define

[Panu Matilainen]

On 01/05/2016 07:16 PM, Richard W.M. Jones wrote:

On Thu, Dec 24, 2015 at 03:01:02PM -0600, Jason L Tibbitts III wrote:

coccinelle (rjones)


coccinelle.spec has:

%{!?python_sitelib: %define python_sitelib %(%{__python} -c "from 
distutils.sysconfig import get_python_lib; print get_python_lib()")}

TBH I have no idea if this is correct or not, or perhaps should be
completely removed.  Thoughts?


That's one of those cases where using %define is actually *wrong* 
because of the macro scoping thing. But in practise it'll never get used 
since these days %python_sitelib is always defined so its not just wrong 
but also pointless as well.



mingw-atk (rjones, epienbro, kalev)


Fixed.


mingw-freetype (rjones, lfarkas, epienbro)


This uses:

%{!?_with_subpixel_rendering: %{!?_without_subpixel_rendering: %define 
_without_subpixel_rendering --without-subpixel_rendering}}

_without_subpixel_rendering is not used anywhere else in the file.
No idea if that is right or not.


Another case where %define is actually wrong. The whole construct looks 
like a workaround for %bcond related misunderstanding, but dunno.


- Panu -
--
devel mailing list
devel@lists.fedoraproject.org
http://lists.fedoraproject.org/admin/lists/devel@lists.fedoraproject.org

Re: Specs using %define

[Jakub Jelen]

On 12/24/2015 10:01 PM, Jason L Tibbitts III wrote:

cyrus-sasl (plautrba, tmraz, jjelen)

Fixed, waiting for prepared upstream release.

openssh (plautrba, mitr, sgrubb, dwalsh, lkundrak, mgrepl, tmraz, jjelen)
tcp_wrappers (plautrba, herczy, jjelen)

Fixed, committed to rawhide. Will build with next occasion.

--
Jakub Jelen
Associate Software Engineer
Security Technologies
Red Hat
--
devel mailing list
devel@lists.fedoraproject.org
http://lists.fedoraproject.org/admin/lists/devel@lists.fedoraproject.org

Re: Specs using %define

[Florian Weimer]

On 12/24/2015 10:01 PM, Jason L Tibbitts III wrote:
> To satisfy my curiosity, I grepped the convenient tarball of specfiles
> (http://pkgs.fedoraproject.org/repo/rpm-specs-latest.tar.xz) for lines
> matching "(? there were more than 1900 hits.

Your message would have benefited from a link to the relevant rule from
the package guidelines. :)

http://fedoraproject.org/wiki/Packaging:Guidelines#.25global_preferred_over_.25define

However, the text is self-contradictory: “The two macro defining
statements behave the same when they are a the top level of rpm's
nesting level” vs “%define and %global differ in more ways than just scope”.

It seems that the key difference is that %define evaluates the macro
body at evaluation time, and %global evaluates it at definition time.
It would be nice to fix the description above to reflect that.

Florian
--
devel mailing list
devel@lists.fedoraproject.org
http://lists.fedoraproject.org/admin/lists/devel@lists.fedoraproject.org

Re: Specs using %define

[Vít Ondruch]

Dne 24.12.2015 v 22:01 Jason L Tibbitts III napsal(a):
> To satisfy my curiosity, I grepped the convenient tarball of specfiles
> (http://pkgs.fedoraproject.org/repo/rpm-specs-latest.tar.xz) for lines
> matching "(? there were more than 1900 hits.
>
> Here's a complete (long) list.  I don't think there's much point in
> doing anything about the vast majority of these, but if you're in
> cleaning up your packages then why not?  Also, feel free to let me know
> if any of these are false positives.  (They may be due to %define in the
> changelog or in a "comment", which is also something that should be
> fixed.)  And if you're using %define because you actually need one of
> the peculiar behaviors it exhibits, please let me know because I'd like
> to see an actual example of that.
>
> (Own goal notice: I have a few packages on this list.  Off to fix them
> up now.)
>
>
> ruby-activerecord (stahnma)
> ruby-activesupport (stahnma)

Are these some zombies or what is it?


Vít
--
devel mailing list
devel@lists.fedoraproject.org
http://lists.fedoraproject.org/admin/lists/devel@lists.fedoraproject.org

Re: Specs using %define

[Vít Ondruch]

Dne 24.12.2015 v 22:01 Jason L Tibbitts III napsal(a):
> http://pkgs.fedoraproject.org/repo/rpm-specs-latest.tar.xz


Btw why this ^^ was never announced to my knowledge nor it is
documented, except [1]. It seems useful indeed.


Vít



[1] https://fedorahosted.org/fedora-infrastructure/ticket/4941
--
devel mailing list
devel@lists.fedoraproject.org
http://lists.fedoraproject.org/admin/lists/devel@lists.fedoraproject.org

Re: Specs using %define

[Kevin Fenzi]

On Mon, 4 Jan 2016 16:45:01 +0100
Vít Ondruch  wrote:

> Dne 24.12.2015 v 22:01 Jason L Tibbitts III napsal(a):
> > http://pkgs.fedoraproject.org/repo/rpm-specs-latest.tar.xz  
> 
> 
> Btw why this ^^ was never announced to my knowledge nor it is
> documented, except [1]. It seems useful indeed.
> 
> 
> Vít
> 
> 
> 
> [1] https://fedorahosted.org/fedora-infrastructure/ticket/4941

Well, it was setup right before the holidays. 

I'm be happy to go announce it now...

kevin


pgp6N0KKGl3DF.pgp
Description: OpenPGP digital signature
--
devel mailing list
devel@lists.fedoraproject.org
http://lists.fedoraproject.org/admin/lists/devel@lists.fedoraproject.org

Re: Specs using %define

[Akira TAGOH]

On Fri, Dec 25, 2015 at 6:01 AM, Jason L Tibbitts III  wrote:
> anthy (tagoh)
> emacs-apel (tagoh)
> emacs-mew (tagoh)
> hanazono-fonts (tagoh)
> japanese-bitmap-fonts (tagoh)
> jisksp16-1990-fonts (tagoh)
> kinput2 (tagoh)
> linux-libertine-fonts (tagoh)
> lv (tagoh)
> mozc (tagoh, fujiwara)
> sazanami-fonts (tagoh)
> uim (tagoh)

Fixed in git but not yet build. thanks.


-- 
Akira TAGOH
--
devel mailing list
devel@lists.fedoraproject.org
http://lists.fedoraproject.org/admin/lists/devel@lists.fedoraproject.org

Re: Specs using %define

[Paul Howarth]

On Sat, 02 Jan 2016 09:29:01 -0600
Jason L Tibbitts III  wrote:

> > "PH" == Paul Howarth  writes:
> 
> PH> In perl-Module-Extract-VERSION I have this for provides filtering
> PH> that works in EPEL < 7:
> 
> Maybe I can grep that out, but you really should be using the proper
> macros in releases where they are supported.

I do:

# Filter bogus provide for perl(ExtUtils::MakeMaker::_version) (rpm 4.9 onwards)
#global __provides_exclude ^perl\\(ExtUtils::MakeMaker::_version\\)

I have both the old and new style filtering in place as one works in
Fedora, the other in EPEL and they don't interfere with each other, so
I can maintain the same spec everywhere.

> PH> It's been there for a long time and it's quite common in perl
> PH> module packages. I don't think this works with %global instead of
> PH> %define.
> 
> I'm not sure; has anyone actually tried it?

Just tried it now and replacing %define with %global worked in this
case for both EPEL-5 and EPEL-6 builds. I'm sure I had some issues with
this when I switched to %global initially so I'll get rid of the
%defines as and when I do package updates and check carefully that I'm
getting the expected results.

Paul.
--
devel mailing list
devel@lists.fedoraproject.org
http://lists.fedoraproject.org/admin/lists/devel@lists.fedoraproject.org

Re: Specs using %define

[drago01]

> bless (drago01)
> ghex (dodji, drago01, kalev)

-> fixed in rawhide.
--
devel mailing list
devel@lists.fedoraproject.org
http://lists.fedoraproject.org/admin/lists/devel@lists.fedoraproject.org

Re: Specs using %define

[Jason L Tibbitts III]

> "PH" == Paul Howarth  writes:

PH> In perl-Module-Extract-VERSION I have this for provides filtering
PH> that works in EPEL < 7:

Maybe I can grep that out, but you really should be using the proper
macros in releases where they are supported.

PH> It's been there for a long time and it's quite common in perl module
PH> packages. I don't think this works with %global instead of %define.

I'm not sure; has anyone actually tried it?

 - J<
--
devel mailing list
devel@lists.fedoraproject.org
http://lists.fedoraproject.org/admin/lists/devel@lists.fedoraproject.org

Re: Specs using %define

[Michael Cronenworth]

On 12/24/2015 03:01 PM, Jason L Tibbitts III wrote:

linux-igd (masahase, mooninite)


Fixed. Thanks.
--
devel mailing list
devel@lists.fedoraproject.org
http://lists.fedoraproject.org/admin/lists/devel@lists.fedoraproject.org

Re: Specs using %define

[Neal Gompa]

On Thu, Dec 24, 2015 at 3:01 PM, Jason L Tibbitts III  wrote:
> oggconvert (ngompa)

Took the opportunity to clean up the package spec a bit and pushed it
to build for rawhide. Thanks.

Ahh, the days gone by when we weren't sure that macros would be
available to us all the time. Now things are much simpler...


-- 
真実はいつも一つ！/ Always, there's only one truth!
--
devel mailing list
devel@lists.fedoraproject.org
http://lists.fedoraproject.org/admin/lists/devel@lists.fedoraproject.org

Re: Specs using %define

[Dave Johansen]

On Thu, Dec 24, 2015 at 2:01 PM, Jason L Tibbitts III 
wrote:

> odb (daveisfera)
>

Fixed in rawhide. No builds made.
Thanks,
Dave
--
devel mailing list
devel@lists.fedoraproject.org
http://lists.fedoraproject.org/admin/lists/devel@lists.fedoraproject.org

Re: Specs using %define

[Douglas Schilling Landgraf]

Hi Jason,

On 12/24/2015 04:01 PM, Jason L Tibbitts III wrote:

To satisfy my curiosity, I grepped the convenient tarball of specfiles
(http://pkgs.fedoraproject.org/repo/rpm-specs-latest.tar.xz) for lines
matching "(?


python-confparser (dougsland, aquini)
python-ovirt-register (dougsland)
vdsm (fsimonce, danken, dougsland, bronhaim)



For the above packages I have sent a patch.

Thanks!
--
devel mailing list
devel@lists.fedoraproject.org
http://lists.fedoraproject.org/admin/lists/devel@lists.fedoraproject.org

Re: Specs using %define

[Paul Howarth]

On Thu, 24 Dec 2015 15:01:02 -0600
Jason L Tibbitts III  wrote:

> To satisfy my curiosity, I grepped the convenient tarball of specfiles
> (http://pkgs.fedoraproject.org/repo/rpm-specs-latest.tar.xz) for lines
> matching "(? there were more than 1900 hits.
> 
> Here's a complete (long) list.  I don't think there's much point in
> doing anything about the vast majority of these, but if you're in
> cleaning up your packages then why not?  Also, feel free to let me
> know if any of these are false positives.  (They may be due to
> %define in the changelog or in a "comment", which is also something
> that should be fixed.)  And if you're using %define because you
> actually need one of the peculiar behaviors it exhibits, please let
> me know because I'd like to see an actual example of that.

In perl-Module-Extract-VERSION I have this for provides filtering that
works in EPEL < 7:

# Filter bogus provide for perl(ExtUtils::MakeMaker::_version) (prior to rpm 
4.9)
%global provfilt /bin/sh -c "%{__perl_provides} | grep -Fvx 
'perl(ExtUtils::MakeMaker::_version)'"
%define __perl_provides %{provfilt}

It's been there for a long time and it's quite common in perl module
packages. I don't think this works with %global instead of %define.


Paul.
--
devel mailing list
devel@lists.fedoraproject.org
http://lists.fedoraproject.org/admin/lists/devel@lists.fedoraproject.org

Re: Specs using %define

[Jason L Tibbitts III]

> "PS" == Petr Stodulka  writes:

PS> zip, unzip, git, gzip - are false positives: found "%%" or just
PS> "define*" in changelog

I'm confused; none of those were in the list I posted.

PS> sed - fixed

For fun, what was there:

%ifos linux
%define _bindir /bin
%endif

Wow.  I can't imagine the original purpose of that block, or how long
ago it must have been added.

However, why not actually remove it completely?  What Fedora or EL
release is not going to have _bindir defined?

 - J<
--
devel mailing list
devel@lists.fedoraproject.org
http://lists.fedoraproject.org/admin/lists/devel@lists.fedoraproject.org

Re: Specs using %define

[Petr Stodulka]

On 24.12.2015 22:01, Jason L Tibbitts III wrote:
> To satisfy my curiosity, I grepped the convenient tarball of specfiles
> (http://pkgs.fedoraproject.org/repo/rpm-specs-latest.tar.xz) for lines
> matching "(? there were more than 1900 hits.
> 
> Here's a complete (long) list.  I don't think there's much point in
> doing anything about the vast majority of these, but if you're in
> cleaning up your packages then why not?  Also, feel free to let me know
> if any of these are false positives.  (They may be due to %define in the
> changelog or in a "comment", which is also something that should be
> fixed.)  And if you're using %define because you actually need one of
> the peculiar behaviors it exhibits, please let me know because I'd like
> to see an actual example of that.
> 
> (Own goal notice: I have a few packages on this list.  Off to fix them
> up now.)
> 

zip, unzip, git, gzip - are false positives: found "%%" or just "define*" in 
changelog
sed - fixed

Happy new Year :-)
Petr



signature.asc
Description: OpenPGP digital signature
--
devel mailing list
devel@lists.fedoraproject.org
http://lists.fedoraproject.org/admin/lists/devel@lists.fedoraproject.org

Re: Specs using %define

[Michael Schwendt]

On Sat, 26 Dec 2015 14:37:13 -0600, Jason L Tibbitts III wrote:

> For fun, what was there:
> 
> %ifos linux
> %define _bindir /bin
> %endif
> 
> Wow.  I can't imagine the original purpose of that block, or how long
> ago it must have been added.
> 
> However, why not actually remove it completely?  What Fedora or EL
> release is not going to have _bindir defined?

%_bindir is not /bin
--
devel mailing list
devel@lists.fedoraproject.org
http://lists.fedoraproject.org/admin/lists/devel@lists.fedoraproject.org

Re: Specs using %define

[Reindl Harald]

Am 26.12.2015 um 22:45 schrieb Jason L Tibbitts III:

"MS" == Michael Schwendt  writes:


MS> %_bindir is not /bin

In Fedora there's not exactly much of a difference because of the
symlink.  But why conditionalize it on "%ifos linux" in any case?


there *is* a difference and hence of it i have a meta-package with 
Provides on any system because otherwise dependency repeatly can't be 
solved because "glibc" and "perl" where years after UsrMove not fixed 
and bugreports ignored


you get hit by it when perl/glibc are updated with the ame transaction 
while your own packages correctly using the macros for their Requires 
instead the old, hardcoded path


Provides:  %{_bindir}/perl
Provides:  %{_sbindir}/ldconfig

https://fedoraproject.org/wiki/Packaging:Guidelines#Effect_of_the_UsrMove_Fedora_Feature





signature.asc
Description: OpenPGP digital signature
--
devel mailing list
devel@lists.fedoraproject.org
http://lists.fedoraproject.org/admin/lists/devel@lists.fedoraproject.org

Re: Specs using %define

[Michael Schwendt]

On Sat, 26 Dec 2015 15:45:24 -0600, Jason L Tibbitts III wrote:

> MS> %_bindir is not /bin  
> 
> In Fedora there's not exactly much of a difference because of the
> symlink.

Not true.

While "rpm -q --whatprovides …" queries follow symlinks, the same
cannot be said about dependencies in repodata. Watch this:

  # repoquery --whatprovides /usr/bin/sed
  #

  # rpmls sed|grep bin
  -rwxr-xr-x  /bin/sed

  # repoquery --whatprovides /bin/sed
  sed-0:4.2.2-11.fc23.x86_64

  # repoquery --whatrequires /usr/bin/sed|wc -l
  0
  # repoquery --whatrequires /bin/sed|wc -l
  11

> But why conditionalize it on "%ifos linux" in any case?

That question is an entirely unrelated one.
--
devel mailing list
devel@lists.fedoraproject.org
http://lists.fedoraproject.org/admin/lists/devel@lists.fedoraproject.org

Re: Specs using %define

[Petr Stodulka]

On 26.12.2015 22:02, Michael Schwendt wrote:
> On Sat, 26 Dec 2015 14:37:13 -0600, Jason L Tibbitts III wrote:
> 
>> For fun, what was there:
>>
>> %ifos linux
>> %define _bindir /bin
>> %endif
>>
>> Wow.  I can't imagine the original purpose of that block, or how long
>> ago it must have been added.
>>
>> However, why not actually remove it completely?  What Fedora or EL
>> release is not going to have _bindir defined?
> 
> %_bindir is not /bin
> --
> devel mailing list
> devel@lists.fedoraproject.org
> http://lists.fedoraproject.org/admin/lists/devel@lists.fedoraproject.org
> 

Is there some reason why should be the script listed in rpm with "path" /bin
instead of /usr/bin? Ok, I am not sure what about few packages (list below)
which require /bin/sed instead of "sed" package.

It's possible, that sed will not be found correctly for them and may there
should be set provides of "/bin/sed" too in that case. I haven't try that yet.

What is your opinion? Use of actual paths seems correctly for me.

Packages which require directly /bin/sed:
libnetdude
libpcapnav
mod_fcgid
os-prober
policycoreutils
redhat-lsb
resource-agents
xchat



signature.asc
Description: OpenPGP digital signature
--
devel mailing list
devel@lists.fedoraproject.org
http://lists.fedoraproject.org/admin/lists/devel@lists.fedoraproject.org

Re: Specs using %define

[Jason L Tibbitts III]

> "MS" == Michael Schwendt  writes:

MS> %_bindir is not /bin

In Fedora there's not exactly much of a difference because of the
symlink.  But why conditionalize it on "%ifos linux" in any case?

 - J<
--
devel mailing list
devel@lists.fedoraproject.org
http://lists.fedoraproject.org/admin/lists/devel@lists.fedoraproject.org

Re: Specs using %define

[Björn Esser]

Am 24.12.2015 um 22:01 schrieb Jason L Tibbitts III:

To satisfy my curiosity, I grepped the convenient tarball of specfiles
(http://pkgs.fedoraproject.org/repo/rpm-specs-latest.tar.xz) for lines
matching "(?

All %define changed to %global in recent commit.


NLopt (besser82)


This spec-file uses %global, but in one special case:  I use %define for 
nested macro-expansion --->  %define lc_name %(echo ${%{name},,})
lc_name would be empty during parsing the spec-file, when declaring with 
%global…

--
devel mailing list
devel@lists.fedoraproject.org
http://lists.fedoraproject.org/admin/lists/devel@lists.fedoraproject.org

Re: Specs using %define

[Alec Leamas]

On 24/12/15 22:01, Jason L Tibbitts III wrote:
> To satisfy my curiosity, I grepped the convenient tarball of specfiles
> (http://pkgs.fedoraproject.org/repo/rpm-specs-latest.tar.xz) for lines
> matching "(? there were more than 1900 hits.

> iguanaIR (leamas)

Fixed in git, no builds made.

Cheers!

--alec
--
devel mailing list
devel@lists.fedoraproject.org
http://lists.fedoraproject.org/admin/lists/devel@lists.fedoraproject.org

Re: Specs using %define

[Ville Skyttä]

On Fri, Dec 25, 2015 at 2:10 AM, Jason L Tibbitts III  wrote:
>> "JJ" == Jerry James  writes:
>
> JJ> This uses a %define inside the %check script as a convenient macro,
> JJ> used only inside the %check script; i.e., it really is a local
> JJ> definition.
>
> But %define isn't a "local definition" in the sense that you're thinking
> of.  It's certainly visible outside of %check, for example.  Which
> actual property of %define are you relying upon here?  I don't think
> you're doing anything specific with nested macro expansions.

Also, if the definitions are really local to some specfile shell code
sections, there's probably no need to involve rpm in them at all --
regular shell variables or functions can be used instead.
--
devel mailing list
devel@lists.fedoraproject.org
http://lists.fedoraproject.org/admin/lists/devel@lists.fedoraproject.org

Re: Specs using %define

[Jason L Tibbitts III]

> "BE" == Björn Esser  writes:

BE> This spec-file uses %global, but in one special case: I use %define
BE> for nested macro-expansion --->

BE>   %define lc_name %(echo ${%{name},,})

BE> lc_name would be empty during parsing the spec-file, when declaring
BE> with %global…

Thanks for letting me know that there is a valid case in use there.  I
don't think there's an easy way to detect valid usage, but I can grep
out that macro when I run things again.

 - J<


-- devel mailing list devel@lists.fedoraproject.org
BE> http://lists.fedoraproject.org/admin/lists/devel@lists.fedoraproject.org


-- 
 Jason L Tibbitts III - ti...@math.uh.edu - 713/743-3486 - 660PGH
 System Manager:  University of Houston Department of Mathematics 
--
devel mailing list
devel@lists.fedoraproject.org
http://lists.fedoraproject.org/admin/lists/devel@lists.fedoraproject.org

Re: Specs using %define

[Christian Dersch]

On 12/24/2015 10:01 PM, Jason L Tibbitts III wrote:
> libapogee (sergiopr, lupinix)
> qwtplot3d (chitlesh, lupinix)
>
 Done :) Also added %license.
--
devel mailing list
devel@lists.fedoraproject.org
http://lists.fedoraproject.org/admin/lists/devel@lists.fedoraproject.org

Re: Specs using %define

[Michael Schwendt]

On Thu, 24 Dec 2015 15:01:02 -0600, Jason L Tibbitts III wrote:

> eiciel (mschwendt)

Fixed in rawhide git. No build requested, though.

> libmowgli (ertzing, mschwendt)

This has not been touched except for several mass-rebuilds.
No API users are left in the package collection.
It could be retired.
--
devel mailing list
devel@lists.fedoraproject.org
http://lists.fedoraproject.org/admin/lists/devel@lists.fedoraproject.org

Specs using %define

[Jason L Tibbitts III]

To satisfy my curiosity, I grepped the convenient tarball of specfiles
(http://pkgs.fedoraproject.org/repo/rpm-specs-latest.tar.xz) for lines
matching "(?http://lists.fedoraproject.org/admin/lists/devel@lists.fedoraproject.org

Re: Specs using %define

[Jason L Tibbitts III]

> "JJ" == Jerry James  writes:

JJ> This uses a %define inside the %check script as a convenient macro,
JJ> used only inside the %check script; i.e., it really is a local
JJ> definition.

But %define isn't a "local definition" in the sense that you're thinking
of.  It's certainly visible outside of %check, for example.  Which
actual property of %define are you relying upon here?  I don't think
you're doing anything specific with nested macro expansions.

It doesn't really make that much difference in the packages you mention,
but you seem to have some misconception of the differences between
%global and %define.

See:
  
https://fedoraproject.org/wiki/Packaging:Guidelines?rd=Packaging/Guidelines#.25global_preferred_over_.25define
and the thread beginning at
  https://www.redhat.com/archives/fedora-packaging/2009-May/msg00095.html
  
 - J<
--
devel mailing list
devel@lists.fedoraproject.org
http://lists.fedoraproject.org/admin/lists/devel@lists.fedoraproject.org

Re: Specs using %define

[Antonio]

Smuxi has been fixed right now. Thanks.

Sent with AquaMail for Android
http://www.aqua-mail.com


On December 24, 2015 10:01:36 PM Jason L Tibbitts III  
wrote:



To satisfy my curiosity, I grepped the convenient tarball of specfiles
(http://pkgs.fedoraproject.org/repo/rpm-specs-latest.tar.xz) for lines
matching "(?alacarte (magcius, alexl, caillon, caolanm, hadess, johnp, mbarnes, 
rhughes, ssp)

alexandria (mtasaka)
alex (petersen)
allegro (jwrdegoede)
alsa-firmware (timj, perex)
alsa-plugins (perex)
alsa-utils (perex)
alt-ergo (amdunn, jjames)
amanda (phracek, jgorig)
amarok (rdieter, tmz, jgrulich, group::kde-sig)
am-utils (iankent)
anaconda-realmd (stefw, vpodzime)
anaconda (anaconda-maint, dcantrel, clumens, pjones, dlehman, rvykydal, 
bcl, sbueno, dshea, vpodzime, wwoods)

ansible (kevin, wzzrd, toshio)
anthy (tagoh)
apitrace (smani)
apmud (dwmw2)
apper (rdieter, ltinkl, rhughes, group::kde-sig)
appliance-tools (ausil, goldmann, mattdm)
apricots (limb)
aprsd (lucilanga)
apr (jorton, bojan, jkaluza)
apr-util (jorton, bojan, jkaluza)
apt (athimm, moceap)
aqbanking (notting, rdieter)
arandr (mzatko, humaton)
archmage (lbazan)
ardour4 (nphilipp)
argus (limb, fab)
aria2 (athmane)
armacycles-ad (limb)
arm-none-eabi-gdb (mzatko)
arts (than, rdieter, kkofler)
artwiz-aleczapka-fonts (spot)
asio (uwog, fale, pocock)
aspell-af (kasal, jchaloup, nforro)
aspell-ar (danken)
aspell-bg (jchaloup, nforro)
aspell-bn (aalam, pnemade)
aspell-br (jchaloup, nforro)
aspell-ca (jchaloup, nforro)
aspell-cs (jchaloup, nforro)
aspell-cy (jchaloup, nforro)
aspell-da (jchaloup, nforro)
aspell-de (jchaloup, nforro)
aspell-el (jchaloup, nforro)
aspell-en (jchaloup, nforro)
aspell-es (jchaloup, nforro)
aspell-fi (opuk)
aspell-fo (jchaloup, nforro)
aspell-fr (jchaloup, nforro)
aspell-ga (jchaloup, nforro)
aspell-gd (jchaloup, nforro)
aspell-gl (jchaloup, nforro)
aspell-gu (aalam, pnemade)
aspell-he (danken)
aspell-hi (aalam, pnemade)
aspell-hr (jchaloup, nforro)
aspell-id (jchaloup, nforro)
aspell-is (jchaloup, nforro)
aspell-it (jchaloup, nforro)
aspell-la (opuk)
aspell-ml (rajeeshknambiar, pnemade)
aspell-mr (aalam, pnemade)
aspell-nl (jchaloup, nforro)
aspell-no (jchaloup, nforro)
aspell-or (aalam, pnemade)
aspell-pa (aalam)
aspell-pl (jchaloup, nforro)
aspell-pt_BR (jchaloup, nforro)
aspell-pt_PT (jchaloup, nforro)
aspell-sk (ondrejj)
aspell-sl (jchaloup, nforro)
aspell-sr (jchaloup, nforro)
aspell-sv (jchaloup, nforro)
aspell-ta (aalam, pnemade)
aspell-te (aalam)
asymptote (spot)
atf (jmmv)
atkmm (kalev, rishi, hguemar)
atk (mclasen, alexl, caillon, caolanm, hadess, johnp, mbarnes, rhughes, 
rstrode, ssp, group::gnome-sig)

atlas (deji, fkluknav, dsommers)
atmel-firmware (kwizart, linville)
atomorun (jwrdegoede)
atop (limb, hobbes1069, grover)
audacity (moezroy, dtimms)
audit (sgrubb, mitr)
autofs (iankent, jmoyer)
automoc (rdieter, tuxbrewr, jreznik)
autotest-framework (mkrizek, cleber, dzickus)
avahi (lennart, msekleta)
avogadro (kkofler)
avr-binutils (tnorth, trondd, mhlavink)
avr-gcc (tnorth, trondd, ndim, mhlavink)
avr-gdb (tnorth, trondd)
babeld (limb, greenscientist)
bacula2 (limb)
bacula-docs (slaanesh)
baekmuk-bdf-fonts (ueno)
baekmuk-ttf-fonts (ueno)
bakery (hguemar)
baloo (rdieter, jgrulich, jreznik, ltinkl, nucleo, rnovacek, than, dvratil, 
group::kde-sig)

balsa (pawsa)
banshee-community-extensions (chkr, npmccallum, elsupergomez)
banshee-mirage (retired?)
bashdb (roma)
bash (ooprala, kasal)
bcm283x-firmware (pbrobinson, pwhalen)
beanstalkd (jjh, gnat)
BEDTools (verdurin)
beecrypt (robert)
berusky-data (stransky)
biblesync (greghellings)
bigloo (jjames, salimma)
bind-dyndb-ldap (pspacek, atkac, thozza)
bindex (omajid)
binutils (nickc, aoliva, jakub, jankratochvil, law)
bitten (timn)
bkchem (limb)
bless (drago01)
bluecove (mbooth)
bochs (clalance, limb, aquini)
boost (pmachata, denisarnaud, jwakely)
botan (thm)
boxes (jhrozek)
brltty (limb, jskarvad)
btrfs-progs (josef, mmahut, sandeen)
bucardo (itamarjp, puiterwijk, lbazan)
bugzilla (itamarjp, eseyman, lbazan)
byacc (pfrankli)
byobu (hpejakle)
bytelist (mmorsi)
bzr-explorer (retired?)
cabal-install (petersen)
ca-certificates (kengert, tmraz, pwouters, jorton)
cadvisor (eparis, jchaloup, lsm5, vbatts)
cairo (company, ajax, alexl, caillon, caolanm, hadess, johnp, mbarnes, 
rhughes, rstrode, ssp, group::gnome-sig)

calamaris (robert)
calc (mattdm)
calendar (dcantrel)
calligra-l10n (rdieter, ltinkl)
calligra (rdieter, ltinkl, group::kde-sig)
cas (astokes)
catfish (mtasaka)
cbios (jwrdegoede)
ccache (scop)
cdargs (retired?)
ceph (kkeithle, jdieter, steve, stingray, ke4qqq, dmick, branto, ktdreyer, 
dachary)

certmaster (ssalevan, wakko666, alikins)
cf-bonveno-fonts (ankursinha)
chameleon (jortel)
chatzilla (hicham)
checkpolicy (dwalsh, mgrepl, plautrba)
cherokee (pali)
chisholm-to-be-continued-fonts (herlo)
chkrootkit (limb)
chm2pdf (hubbitus)
chmsee (bbbush, pwu)
cifs-utils (sprabhu, simo)
cjkuni-ukai-fonts (pwu, jamesni)
ClanLib1 (jwrdegoede)

Re: Specs using %define

[Antonio]

smuxi (sagitter, cicku)


On December 24, 2015 10:01:36 PM Jason L Tibbitts III  
wrote:



To satisfy my curiosity, I grepped the convenient tarball of specfiles
(http://pkgs.fedoraproject.org/repo/rpm-specs-latest.tar.xz) for lines
matching "(?alacarte (magcius, alexl, caillon, caolanm, hadess, johnp, mbarnes, 
rhughes, ssp)

alexandria (mtasaka)
alex (petersen)
allegro (jwrdegoede)
alsa-firmware (timj, perex)
alsa-plugins (perex)
alsa-utils (perex)
alt-ergo (amdunn, jjames)
amanda (phracek, jgorig)
amarok (rdieter, tmz, jgrulich, group::kde-sig)
am-utils (iankent)
anaconda-realmd (stefw, vpodzime)
anaconda (anaconda-maint, dcantrel, clumens, pjones, dlehman, rvykydal, 
bcl, sbueno, dshea, vpodzime, wwoods)

ansible (kevin, wzzrd, toshio)
anthy (tagoh)
apitrace (smani)
apmud (dwmw2)
apper (rdieter, ltinkl, rhughes, group::kde-sig)
appliance-tools (ausil, goldmann, mattdm)
apricots (limb)
aprsd (lucilanga)
apr (jorton, bojan, jkaluza)
apr-util (jorton, bojan, jkaluza)
apt (athimm, moceap)
aqbanking (notting, rdieter)
arandr (mzatko, humaton)
archmage (lbazan)
ardour4 (nphilipp)
argus (limb, fab)
aria2 (athmane)
armacycles-ad (limb)
arm-none-eabi-gdb (mzatko)
arts (than, rdieter, kkofler)
artwiz-aleczapka-fonts (spot)
asio (uwog, fale, pocock)
aspell-af (kasal, jchaloup, nforro)
aspell-ar (danken)
aspell-bg (jchaloup, nforro)
aspell-bn (aalam, pnemade)
aspell-br (jchaloup, nforro)
aspell-ca (jchaloup, nforro)
aspell-cs (jchaloup, nforro)
aspell-cy (jchaloup, nforro)
aspell-da (jchaloup, nforro)
aspell-de (jchaloup, nforro)
aspell-el (jchaloup, nforro)
aspell-en (jchaloup, nforro)
aspell-es (jchaloup, nforro)
aspell-fi (opuk)
aspell-fo (jchaloup, nforro)
aspell-fr (jchaloup, nforro)
aspell-ga (jchaloup, nforro)
aspell-gd (jchaloup, nforro)
aspell-gl (jchaloup, nforro)
aspell-gu (aalam, pnemade)
aspell-he (danken)
aspell-hi (aalam, pnemade)
aspell-hr (jchaloup, nforro)
aspell-id (jchaloup, nforro)
aspell-is (jchaloup, nforro)
aspell-it (jchaloup, nforro)
aspell-la (opuk)
aspell-ml (rajeeshknambiar, pnemade)
aspell-mr (aalam, pnemade)
aspell-nl (jchaloup, nforro)
aspell-no (jchaloup, nforro)
aspell-or (aalam, pnemade)
aspell-pa (aalam)
aspell-pl (jchaloup, nforro)
aspell-pt_BR (jchaloup, nforro)
aspell-pt_PT (jchaloup, nforro)
aspell-sk (ondrejj)
aspell-sl (jchaloup, nforro)
aspell-sr (jchaloup, nforro)
aspell-sv (jchaloup, nforro)
aspell-ta (aalam, pnemade)
aspell-te (aalam)
asymptote (spot)
atf (jmmv)
atkmm (kalev, rishi, hguemar)
atk (mclasen, alexl, caillon, caolanm, hadess, johnp, mbarnes, rhughes, 
rstrode, ssp, group::gnome-sig)

atlas (deji, fkluknav, dsommers)
atmel-firmware (kwizart, linville)
atomorun (jwrdegoede)
atop (limb, hobbes1069, grover)
audacity (moezroy, dtimms)
audit (sgrubb, mitr)
autofs (iankent, jmoyer)
automoc (rdieter, tuxbrewr, jreznik)
autotest-framework (mkrizek, cleber, dzickus)
avahi (lennart, msekleta)
avogadro (kkofler)
avr-binutils (tnorth, trondd, mhlavink)
avr-gcc (tnorth, trondd, ndim, mhlavink)
avr-gdb (tnorth, trondd)
babeld (limb, greenscientist)
bacula2 (limb)
bacula-docs (slaanesh)
baekmuk-bdf-fonts (ueno)
baekmuk-ttf-fonts (ueno)
bakery (hguemar)
baloo (rdieter, jgrulich, jreznik, ltinkl, nucleo, rnovacek, than, dvratil, 
group::kde-sig)

balsa (pawsa)
banshee-community-extensions (chkr, npmccallum, elsupergomez)
banshee-mirage (retired?)
bashdb (roma)
bash (ooprala, kasal)
bcm283x-firmware (pbrobinson, pwhalen)
beanstalkd (jjh, gnat)
BEDTools (verdurin)
beecrypt (robert)
berusky-data (stransky)
biblesync (greghellings)
bigloo (jjames, salimma)
bind-dyndb-ldap (pspacek, atkac, thozza)
bindex (omajid)
binutils (nickc, aoliva, jakub, jankratochvil, law)
bitten (timn)
bkchem (limb)
bless (drago01)
bluecove (mbooth)
bochs (clalance, limb, aquini)
boost (pmachata, denisarnaud, jwakely)
botan (thm)
boxes (jhrozek)
brltty (limb, jskarvad)
btrfs-progs (josef, mmahut, sandeen)
bucardo (itamarjp, puiterwijk, lbazan)
bugzilla (itamarjp, eseyman, lbazan)
byacc (pfrankli)
byobu (hpejakle)
bytelist (mmorsi)
bzr-explorer (retired?)
cabal-install (petersen)
ca-certificates (kengert, tmraz, pwouters, jorton)
cadvisor (eparis, jchaloup, lsm5, vbatts)
cairo (company, ajax, alexl, caillon, caolanm, hadess, johnp, mbarnes, 
rhughes, rstrode, ssp, group::gnome-sig)

calamaris (robert)
calc (mattdm)
calendar (dcantrel)
calligra-l10n (rdieter, ltinkl)
calligra (rdieter, ltinkl, group::kde-sig)
cas (astokes)
catfish (mtasaka)
cbios (jwrdegoede)
ccache (scop)
cdargs (retired?)
ceph (kkeithle, jdieter, steve, stingray, ke4qqq, dmick, branto, ktdreyer, 
dachary)

certmaster (ssalevan, wakko666, alikins)
cf-bonveno-fonts (ankursinha)
chameleon (jortel)
chatzilla (hicham)
checkpolicy (dwalsh, mgrepl, plautrba)
cherokee (pali)
chisholm-to-be-continued-fonts (herlo)
chkrootkit (limb)
chm2pdf (hubbitus)
chmsee (bbbush, pwu)
cifs-utils (sprabhu, simo)
cjkuni-ukai-fonts (pwu, jamesni)
ClanLib1 (jwrdegoede)
clearsilver (limb)
clement (jmrcpn)
clips (timn)
cloud-utils (juergh)

Re: Specs using %define

[Jerry James]

On Thu, Dec 24, 2015 at 2:01 PM, Jason L Tibbitts III  wrote:
> alt-ergo (amdunn, jjames)

This uses a %define inside the %check script as a convenient macro,
used only inside the %check script; i.e., it really is a local
definition.

> bigloo (jjames, salimma)

This uses a %define inside the %build script as a convenient macro,
used only inside the %build script; i.e., it really is a local
definition.

> mona (jjames)

These %defines should be %globals; will fix them.

> ntl (rdieter, jjames, jehane, pcpa)

This %define should be a %global; will fix it.

> sphinxtrain (jjames)

This uses a %define inside the %prep script as a convenient macro,
used only inside the %prep script; i.e., it really is a local
definition.

> why (amdunn, jjames)

This uses a %define inside the %prep script as a convenient macro,
used only inside the %prep script; i.e., it really is a local
definition.
-- 
Jerry James
http://www.jamezone.org/
--
devel mailing list
devel@lists.fedoraproject.org
http://lists.fedoraproject.org/admin/lists/devel@lists.fedoraproject.org

Re: Specs using %define

[gil]

hi
Il 24/12/2015 22:01, Jason L Tibbitts III ha scritto:

java_cup (gil, mizdebsk)

fixed in java_cup-0.11b-2.fc24
Task info: http://koji.fedoraproject.org/koji/taskinfo?taskID=12300790

proxool (gil)

fixed in proxool-0.9.1-17.fc24
Task info: http://koji.fedoraproject.org/koji/taskinfo?taskID=12300795


xpp3 (mbooth, dbhole, mizdebsk, gil)

fixed in xpp3-1.1.4-8.c.fc24
Task info: http://koji.fedoraproject.org/koji/taskinfo?taskID=12300792
thanks for reporting the problem
merry xmas
.g
--
devel mailing list
devel@lists.fedoraproject.org
http://lists.fedoraproject.org/admin/lists/devel@lists.fedoraproject.org

Re: Specs using %define

[Jason L Tibbitts III]

> "MS" == Michael Schwendt  writes:

>> libmowgli (ertzing, mschwendt)

MS> This has not been touched except for several mass-rebuilds.  No API
MS> users are left in the package collection.  It could be retired.

That's one of the reasons I like to poke around in the collection like
this; we fine all sorts of old things that should probably be cleaned
up.  I'll probably do a run like this a couple of times a year (if I can
remember).

 - J<
--
devel mailing list
devel@lists.fedoraproject.org
http://lists.fedoraproject.org/admin/lists/devel@lists.fedoraproject.org