[gentoo-science] sage 5.7

[gentoo-science] sage 5.7

[Thomas Kahle]

[-- Attachment #1: Type: text/plain, Size: 724 bytes --]

Hi guys,

I'm using the keywordfile from s-o-g, but I seem to need some further
unmasks.  Is it my fault, or should they be added?

#required by dev-python/sympy-0.7.2-r2::sage-on-gentoo[opengl], required by sci-mathematics/sage-5.7-r1::sage-on-gentoo, required by sci-mathematics/sage-notebook-0.10.4-r1::sage-on-gentoo
=dev-python/pyopengl-3.0.2-r1 ~amd64
#required by dev-python/wxpython-2.8.12.1-r1[cairo], required by dev-python/ipython-0.13.1-r1::sage-on-gentoo[wxwidgets], required by sci-mathematics/sage-5.7-r1::sage-on-gentoo, required by sci-mathematics/sage-notebook-0.10.4-r1::sage-on-gentoo
=dev-python/pycairo-1.10.0-r4 ~amd64

Thanks,
Thomas


-- 
Thomas Kahle
http://dev.gentoo.org/~tomka/

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 316 bytes --]

Re: [gentoo-science] sage 5.7

[Francois Bissey]

On 28/02/13 21:52, Thomas Kahle wrote:
> On 10:04 Thu 28 Feb 2013, fbissey@slingshot.co.nz wrote:
>> Quoting Thomas Kahle <tomka@gentoo.org>:
>>
>>> Hi guys,
>>>
>>> I'm using the keywordfile from s-o-g, but I seem to need some further
>>> unmasks.  Is it my fault, or should they be added?
>>>
>>> #required by dev-python/sympy-0.7.2-r2::sage-on-gentoo[opengl], 
>>> required by sci-mathematics/sage-5.7-r1::sage-on-gentoo, required by 
>>> sci-mathematics/sage-notebook-0.10.4-r1::sage-on-gentoo
>>> =dev-python/pyopengl-3.0.2-r1 ~amd64
>>> #required by dev-python/wxpython-2.8.12.1-r1[cairo], required by 
>>> dev-python/ipython-0.13.1-r1::sage-on-gentoo[wxwidgets], required by 
>>> sci-mathematics/sage-5.7-r1::sage-on-gentoo, required by 
>>> sci-mathematics/sage-notebook-0.10.4-r1::sage-on-gentoo
>>> =dev-python/pycairo-1.10.0-r4 ~amd64
>>>
>> Hi Thomas,
>>
>> I masked ipython-0.13.1-r1 for now. It is a python-r1 migrated ipython
>> and we would need to keywords all the dependencies. We just keep it ready.
> 
> The main tree also has an ipythin-0.13.1-r1 which applies to patches
> that the s-o-g version does not apply.  Maybe that's the problem?
> 

This must be new patches. I "forked" the sage-on-gentoo" from the main
tree 7 days ago. At that time the difference was patches I added based
on the spkg shipped in sage (one backport from an unreleased version,
the other won't be merged because what it fixes is being revamped).

>> pyopengl on the other hand is very curious. It is asked by sympy and
>> I specifically didn't depend on python-r1 pyopengl which is what portage
>> wants to install here. The output of emerge -puDtv sage could be helpful.
> 
> probably related...
> 

I doubt it but it would be nice.

Francois

Re: [gentoo-science] sage 5.7

[Francois Bissey]

On 28/02/13 22:27, Thomas Kahle wrote:
> On 22:11 Thu 28 Feb 2013, Francois Bissey wrote:
>> This must be new patches. I "forked" the sage-on-gentoo" from the main
>> tree 7 days ago. At that time the difference was patches I added based
>> on the spkg shipped in sage (one backport from an unreleased version,
>> the other won't be merged because what it fixes is being revamped).
> 
> Oh, my bad.  The patches are in your version:
> 
> # diff ~up/dev-python/ipython/ipython-0.13.1-r1.ebuild /var/lib/layman/sage-on-gentoo/dev-python/ipython/ipython-0.13.1-r1.ebuild
> 3c3
> < # $Header: /var/cvsroot/gentoo-x86/dev-python/ipython/ipython-0.13.1-r1.ebuild,v 1.1 2013/02/11 15:59:01 mgorny Exp $
> ---
>> # $Header: $
> 52a53,54
>>       epatch "${FILESDIR}"/${PN}-history-backport.patch
>>       epatch "${FILESDIR}"/prun_timeit_magics.patch
> 
> 
> OK.  What I'm trying to say is: Your mask is irrelevant since the
> main-tree version is not masked and emerge want's to install it anyway,
> I think?  If you intended to not unmask any python-1 ebuilds, then it
> failed.
>

Not all but you are right. The mask only applies to the content of the
overlay. That's by design. I don't like it but we either have to provide
keywording for the python-r1 dep or enforce ipython-0.13.1 exactly
which is about as bad.

Opinions Christopher?

Francois

Re: [gentoo-science] sage 5.7

[Francois Bissey]

On 28/02/13 22:44, Francois Bissey wrote:
> On 28/02/13 22:27, Thomas Kahle wrote:
>> On 22:11 Thu 28 Feb 2013, Francois Bissey wrote:
>>> This must be new patches. I "forked" the sage-on-gentoo" from the main
>>> tree 7 days ago. At that time the difference was patches I added based
>>> on the spkg shipped in sage (one backport from an unreleased version,
>>> the other won't be merged because what it fixes is being revamped).
>>
>> Oh, my bad.  The patches are in your version:
>>
>> # diff ~up/dev-python/ipython/ipython-0.13.1-r1.ebuild /var/lib/layman/sage-on-gentoo/dev-python/ipython/ipython-0.13.1-r1.ebuild
>> 3c3
>> < # $Header: /var/cvsroot/gentoo-x86/dev-python/ipython/ipython-0.13.1-r1.ebuild,v 1.1 2013/02/11 15:59:01 mgorny Exp $
>> ---
>>> # $Header: $
>> 52a53,54
>>>       epatch "${FILESDIR}"/${PN}-history-backport.patch
>>>       epatch "${FILESDIR}"/prun_timeit_magics.patch
>>
>>
>> OK.  What I'm trying to say is: Your mask is irrelevant since the
>> main-tree version is not masked and emerge want's to install it anyway,
>> I think?  If you intended to not unmask any python-1 ebuilds, then it
>> failed.
>>
> 
> Not all but you are right. The mask only applies to the content of the
> overlay. That's by design. I don't like it but we either have to provide
> keywording for the python-r1 dep or enforce ipython-0.13.1 exactly
> which is about as bad.
> 
> Opinions Christopher?
> 

I know I have just asked Christopher's opinion but I will make an
arbitrary decision now.

One of the options of ipython is matplotlib and the only python-r1
ebuild for this is at version 1.2.0 while I have locked sage to 1.1.0
(this may be reviewed later). Therefore I am locking ipython to 0.13.1
precisely.

Francois

Re: [gentoo-science] sage 5.7

[Christopher Schwan]

On Thursday 28 February 2013 22:52:43 Francois Bissey wrote:
> On 28/02/13 22:44, Francois Bissey wrote:
> > On 28/02/13 22:27, Thomas Kahle wrote:
> >> On 22:11 Thu 28 Feb 2013, Francois Bissey wrote:
> >>> This must be new patches. I "forked" the sage-on-gentoo" from the main
> >>> tree 7 days ago. At that time the difference was patches I added based
> >>> on the spkg shipped in sage (one backport from an unreleased version,
> >>> the other won't be merged because what it fixes is being revamped).
> >> 
> >> Oh, my bad.  The patches are in your version:
> >> 
> >> # diff ~up/dev-python/ipython/ipython-0.13.1-r1.ebuild
> >> /var/lib/layman/sage-on-gentoo/dev-python/ipython/ipython-0.13.1-r1.ebui
> >> ld 3c3
> >> < # $Header:
> >> /var/cvsroot/gentoo-x86/dev-python/ipython/ipython-0.13.1-r1.ebuild,v
> >> 1.1 2013/02/11 15:59:01 mgorny Exp $ ---
> >> 
> >>> # $Header: $
> >> 
> >> 52a53,54
> >> 
> >>>       epatch "${FILESDIR}"/${PN}-history-backport.patch
> >>>       epatch "${FILESDIR}"/prun_timeit_magics.patch
> >> 
> >> OK.  What I'm trying to say is: Your mask is irrelevant since the
> >> main-tree version is not masked and emerge want's to install it anyway,
> >> I think?  If you intended to not unmask any python-1 ebuilds, then it
> >> failed.
> > 
> > Not all but you are right. The mask only applies to the content of the
> > overlay. That's by design. I don't like it but we either have to provide
> > keywording for the python-r1 dep or enforce ipython-0.13.1 exactly
> > which is about as bad.
> > 
> > Opinions Christopher?
> 
> I know I have just asked Christopher's opinion but I will make an
> arbitrary decision now.
> 
> One of the options of ipython is matplotlib and the only python-r1
> ebuild for this is at version 1.2.0 while I have locked sage to 1.1.0
> (this may be reviewed later). Therefore I am locking ipython to 0.13.1
> precisely.

I am fine with that.

> 
> Francois

Christopher