[gentoo-science] sage-scripts' dependencies

[gentoo-science] sage-scripts' dependencies

[Steven Trogdon]

[-- Attachment #1: Type: text/plain, Size: 676 bytes --]

Hi,

I encountered the following when making a fresh install of sage (~136  
packages) within a 32bit chroot. Emerge of sage terminated because of the  
following line

	dosym $(which gp) "${SAGE_LOCAL}"/bin/sage_pari || die

in the sage-scripts ebuild. In emerging sage, sage-scripts gets pulled in  
before pari and thus the failure. I suppose it would be awkward to have  
sage-scripts depend on pari when only a symlink is created? This can be  
replicated if one has a working sage by removing pari and sage-scripts and  
either updating world  or re-emerging sage-core. Here, sage-scripts gets  
pulled in before pari and the update/emerge fails.

Steve

[-- Attachment #2: Type: application/pgp-signature, Size: 198 bytes --]

Re: [gentoo-science] sage-scripts' dependencies

[Christopher Schwan]

Hi,

On Monday 05 July 2010 19:00:15 Steven Trogdon wrote:
> Hi,
> 
> I encountered the following when making a fresh install of sage (~136
> packages) within a 32bit chroot. Emerge of sage terminated because of the
> following line
> 
> 	dosym $(which gp) "${SAGE_LOCAL}"/bin/sage_pari || die
> 
> in the sage-scripts ebuild. In emerging sage, sage-scripts gets pulled in
> before pari and thus the failure. I suppose it would be awkward to have
> sage-scripts depend on pari when only a symlink is created? This can be
> replicated if one has a working sage by removing pari and sage-scripts and
> either updating world  or re-emerging sage-core. Here, sage-scripts gets
> pulled in before pari and the update/emerge fails.

Thats defintely a bug - I am currently testing if the link is actually needed.

> 
> Steve

Christopher

Re: [gentoo-science] sage-scripts' dependencies

[François Bissey]

> Hi,
> 
> I encountered the following when making a fresh install of sage (~136
> packages) within a 32bit chroot. Emerge of sage terminated because of the
> following line
> 
> 	dosym $(which gp) "${SAGE_LOCAL}"/bin/sage_pari || die
> 
> in the sage-scripts ebuild. In emerging sage, sage-scripts gets pulled in
> before pari and thus the failure. I suppose it would be awkward to have
> sage-scripts depend on pari when only a symlink is created? This can be
> replicated if one has a working sage by removing pari and sage-scripts and
> either updating world  or re-emerging sage-core. Here, sage-scripts gets
> pulled in before pari and the update/emerge fails.
> 
Hi Steve,

I have just added a svn snapshot of pari in the overlay but not all 
dependencies have been adjusted for the right slot. 
How did that go? Did you get pari 2.3 or 2.4 installed? 
I would imagine that if you had only pari 2.4 things failed to compile as 
I have slotted it so you have to request it explicitly which happens nowhere
at the moment.

Francois

Re: [gentoo-science] sage-scripts' dependencies

[Steven Trogdon]

[-- Attachment #1: Type: text/plain, Size: 664 bytes --]

On 07/05/2010 06:35:33 PM, François Bissey wrote:
> Hi Steve,
> 
> I have just added a svn snapshot of pari in the overlay but not all
> dependencies have been adjusted for the right slot.
> How did that go? Did you get pari 2.3 or 2.4 installed?
> I would imagine that if you had only pari 2.4 things failed to compile as
> I have slotted it so you have to request it explicitly which happens nowhere
> at the moment.
> 
> Francois
> 
Hi François,

Yes I had noticed that and had masked the 2.4 pari. I used pari-2.3.5. It  
would appear Christopher has removed the symlinks from sage-scripts. I haven't  
tested the latest changes yet.

Steve

[-- Attachment #2: Type: application/pgp-signature, Size: 198 bytes --]