On Tuesday, June 11, 2013 10:47:48 AM UTC+1, Snark wrote:
> What would break if they were copied instead of linked?
>
Probably not in the case of ATLAS since that library is pretty
self-contained.
You might ask in general why do we compile libraries instead of copying
them from the build hos
On Tue, Jun 11, 2013 at 11:47:48AM +0200, Julien Puydt wrote:
> Le 11/06/2013 11:33, Volker Braun a écrit :
> >I'd say archiving the symlink is exactly what you'd want. If you use
> >system libraries for compiling Sage then they become dependencies of the
> >binary distribution, obviously.
>
> Wha
Le 11/06/2013 11:33, Volker Braun a écrit :
I'd say archiving the symlink is exactly what you'd want. If you use
system libraries for compiling Sage then they become dependencies of the
binary distribution, obviously.
What would break if they were copied instead of linked?
The current answers
I'd say archiving the symlink is exactly what you'd want. If you use system
libraries for compiling Sage then they become dependencies of the binary
distribution, obviously.
On Tuesday, June 11, 2013 10:08:20 AM UTC+1, Jeroen Demeyer wrote:
>
> On 06/11/2013 09:43 AM, Julien Puydt wrote:
> > H
On 06/11/2013 09:43 AM, Julien Puydt wrote:
Hi,
shouldn't the atlas spkg copy the static library instead of linking
them, in order for sage -bdist to produce a working archive?
Can't we just declare SAGE_ATLAS_LIB to be fundamentally incompatible
with making a bdist?
--
You received this mess
Hi,
shouldn't the atlas spkg copy the static library instead of linking
them, in order for sage -bdist to produce a working archive?
With the current situation, sage -bdist will archive a symlink, which
might not (hence will not -- it's not pessimism if it's true...) be
available on a target
