Mailing List Archive

On Sat, 2011-02-05 at 11:46 -0200, Paulo Cavalcanti wrote:
>
> http://people.atrpms.net/~pcavalcanti/srpms/blender-2.49b-22.fc14.src.rpm

> this version does not link to gettextlib any more and fix the problem
> with YafaRay crashing
> with the GTK+ qt4 style.

Thanks! BTW it does not build on F15, looks like some implicit include
for ptrdiff types needs to be explicitly included, but F15 isn't due for
a couple of months, so it's something we can fix later on. :)
--
http://thimm.gr/ - http://ATrpms.net/

On Mon, Feb 7, 2011 at 8:02 AM, Axel Thimm <Axel.Thimm@atrpms.net> wrote:

> On Sat, 2011-02-05 at 11:46 -0200, Paulo Cavalcanti wrote:
> >
> >
> http://people.atrpms.net/~pcavalcanti/srpms/blender-2.49b-22.fc14.src.rpm
>
> > this version does not link to gettextlib any more and fix the problem
> > with YafaRay crashing
> > with the GTK+ qt4 style.
>
> Thanks! BTW it does not build on F15, looks like some implicit include
> for ptrdiff types needs to be explicitly included, but F15 isn't due for
> a couple of months, so it's something we can fix later on. :)
>
>
F15 is shipping blender 2.56, which is not considered stable yet.

By the way, your macros do not include .pyo and .pyc in the packages.,
opposed to Fedora.

Of course, the packages become smaller, but if the dates are correct, those
bytecode files should increase
the performance, or this is irrelevant?

Thanks.

--
Paulo Roma Cavalcanti
LCG - UFRJ

On Mon, 2011-02-07 at 10:42 -0200, Paulo Cavalcanti wrote:
> On Mon, Feb 7, 2011 at 8:02 AM, Axel Thimm <Axel.Thimm@atrpms.net>
> wrote:
> On Sat, 2011-02-05 at 11:46 -0200, Paulo Cavalcanti wrote:
> http://people.atrpms.net/~pcavalcanti/srpms/blender-2.49b-22.fc14.src.rpm

> Thanks! BTW it does not build on F15, looks like some implicit
> [...]
> F15 is shipping blender 2.56, which is not considered stable yet.

That happens from time to time: Some packagers hope that the package
will be out of beta once the current Fedora spin gets released and
sometimes they stay in beta. The problem is that Fedora has no other
place to put experimental packages, e.g. something like bleeding.

> By the way, your macros do not include .pyo and .pyc in the packages.,
> opposed to Fedora.
>
> Of course, the packages become smaller, but if the dates are correct,
> those bytecode files should increase the performance, or this is
> irrelevant?

I'm not sure what I have done to make rpm not package what it would
normally package (?). I'm using native rpm as much as possible, does
this happen on f14, too?
--
http://thimm.gr/ - http://ATrpms.net/

On Tue, Feb 8, 2011 at 6:37 AM, Axel Thimm <Axel.Thimm@atrpms.net> wrote:

> On Mon, 2011-02-07 at 10:42 -0200, Paulo Cavalcanti wrote:
> > On Mon, Feb 7, 2011 at 8:02 AM, Axel Thimm <Axel.Thimm@atrpms.net>
> > wrote:
> > On Sat, 2011-02-05 at 11:46 -0200, Paulo Cavalcanti wrote:
> >
> http://people.atrpms.net/~pcavalcanti/srpms/blender-2.49b-22.fc14.src.rpm
>
> > Thanks! BTW it does not build on F15, looks like some implicit
> > [...]
> > F15 is shipping blender 2.56, which is not considered stable yet.
>
> That happens from time to time: Some packagers hope that the package
> will be out of beta once the current Fedora spin gets released and
> sometimes they stay in beta. The problem is that Fedora has no other
> place to put experimental packages, e.g. something like bleeding.
>

The ptrdiff bug is also there. I think it just needs to be "typefef
typename",
but I did not tried it yet.


>
> > By the way, your macros do not include .pyo and .pyc in the packages.,
> > opposed to Fedora.
> >
> > Of course, the packages become smaller, but if the dates are correct,
> > those bytecode files should increase the performance, or this is
> > irrelevant?
>
> I'm not sure what I have done to make rpm not package what it would
> normally package (?). I'm using native rpm as much as possible, does
> this happen on f14, too?
>
>
Yes. I noticed your packages were much smaller than the ones created by
mock, in F14.
The reason was the missing .pyo and .pyc bits.... Fedora 14 seems to be
pushing
the .pyc and .pyo bits into the rpms again. I guess rhel does not included
them, though...
You will have to check, but I do not even know if this is an important issue
on not.

This is from the Fedora blender package I fixed, because of the
libgettextlib problem:

http://download.fedora.redhat.com/pub/fedora/linux/updates/testing/14/x86_64/blender-2.49b-11.fc14.x86_64.rpm

.......


/usr/share/blender/VERSION
/usr/share/blender/scripts
/usr/share/blender/scripts/3ds_export.py
/usr/share/blender/scripts/3ds_export.pyc
/usr/share/blender/scripts/3ds_export.pyo
/usr/share/blender/scripts/3ds_import.py
/usr/share/blender/scripts/3ds_import.pyc
/usr/share/blender/scripts/3ds_import.pyo
/usr/share/blender/scripts/Axiscopy.py
/usr/share/blender/scripts/Axiscopy.pyc
/usr/share/blender/scripts/Axiscopy.pyo
/usr/share/blender/scripts/DirectX8Exporter.py
/usr/share/blender/scripts/DirectX8Exporter.pyc
/usr/share/blender/scripts/DirectX8Exporter.pyo



--
Paulo Roma Cavalcanti
LCG - UFRJ

On Tue, Feb 8, 2011 at 7:19 AM, Paulo Cavalcanti <promac@gmail.com> wrote:

>
>
> On Tue, Feb 8, 2011 at 6:37 AM, Axel Thimm <Axel.Thimm@atrpms.net> wrote:
>
>> On Mon, 2011-02-07 at 10:42 -0200, Paulo Cavalcanti wrote:
>> > On Mon, Feb 7, 2011 at 8:02 AM, Axel Thimm <Axel.Thimm@atrpms.net>
>> > wrote:
>> > On Sat, 2011-02-05 at 11:46 -0200, Paulo Cavalcanti wrote:
>> >
>> http://people.atrpms.net/~pcavalcanti/srpms/blender-2.49b-22.fc14.src.rpm
>>
>> > Thanks! BTW it does not build on F15, looks like some implicit
>> > [...]
>> > F15 is shipping blender 2.56, which is not considered stable yet.
>>
>> That happens from time to time: Some packagers hope that the package
>> will be out of beta once the current Fedora spin gets released and
>> sometimes they stay in beta. The problem is that Fedora has no other
>> place to put experimental packages, e.g. something like bleeding.
>>
>
> The ptrdiff bug is also there. I think it just needs to be "typefef
> typename",
> but I did not tried it yet.
>


In fact, it is a bunch of missing

#include <stddef.h>


--
Paulo Roma Cavalcanti
LCG - UFRJ

On Tue, Feb 8, 2011 at 8:43 PM, Paulo Cavalcanti <promac@gmail.com> wrote:

>
>
> On Tue, Feb 8, 2011 at 7:19 AM, Paulo Cavalcanti <promac@gmail.com> wrote:
>
>>
>>
>> On Tue, Feb 8, 2011 at 6:37 AM, Axel Thimm <Axel.Thimm@atrpms.net> wrote:
>>
>>> On Mon, 2011-02-07 at 10:42 -0200, Paulo Cavalcanti wrote:
>>> > On Mon, Feb 7, 2011 at 8:02 AM, Axel Thimm <Axel.Thimm@atrpms.net>
>>> > wrote:
>>> > On Sat, 2011-02-05 at 11:46 -0200, Paulo Cavalcanti wrote:
>>> >
>>> http://people.atrpms.net/~pcavalcanti/srpms/blender-2.49b-22.fc14.src.rpm
>>>
>>> > Thanks! BTW it does not build on F15, looks like some implicit
>>> > [...]
>>> > F15 is shipping blender 2.56, which is not considered stable yet.
>>>
>>> That happens from time to time: Some packagers hope that the package
>>> will be out of beta once the current Fedora spin gets released and
>>> sometimes they stay in beta. The problem is that Fedora has no other
>>> place to put experimental packages, e.g. something like bleeding.
>>>
>>
>> The ptrdiff bug is also there. I think it just needs to be "typefef
>> typename",
>> but I did not tried it yet.
>>
>
>
> In fact, it is a bunch of missing
>
> #include <stddef.h>
>
>
>
I have a version that compiles either for F14 or F15:

http://people.atrpms.net/~pcavalcanti/srpms/blender-2.56-8.fc14.src.rpm

However, it does require python3.

For F14, you should use

--without py32

because F14 only has python3-3.1. In this case, I disabled the bytecompile
because
it seems to be broken on python3-3.1, anyway.

You can push it to testing, because it is not considered a stable version
yet.

Thanks.

--
Paulo Roma Cavalcanti
LCG - UFRJ

On Wed, 2011-02-09 at 13:16 -0200, Paulo Cavalcanti wrote:

> I have a version that compiles either for F14 or F15:
>
> http://people.atrpms.net/~pcavalcanti/srpms/blender-2.56-8.fc14.src.rpm
>
> However, it does require python3.
>
> For F14, you should use
>
> --without py32
>
> because F14 only has python3-3.1. In this case, I disabled the
> bytecompile because
> it seems to be broken on python3-3.1, anyway.
>
> You can push it to testing, because it is not considered a stable
> version yet.

Thanks, done.

--
http://thimm.gr/ - http://ATrpms.net/