Le 18/02/2015 18:42, Jeroen Demeyer a écrit :
On 2015-02-18 18:04, Julien Puydt wrote:
In fact, whenever sage-the-distribution pretends to package a software
"foo" version "3.14", but patches it left and right, breaking both the
API and the ABI, that is a problem not only for debian, but for all
serious distributions...
But what's the solution?
I think it's not realistic to require Sage to only depend on unpatched
stable versions of packages. PARI is especially bad because PARI has an
enormous gap between GIT master. Excluding bug-fix releases, there are
*years* between stable releases of PARI.
When I asked Bill about the cb_pari_err_handle patch, he asked why sage
wasn't using the standard pari error trapping mechanism.
I'm still discussing the matter with him, but perhaps I'll understand
enough of the issue to propose a nice patch. That wouldn't be the first
time I patch away things in a sage package in favor of actually using
upstream api to do the same.
I don't quite understand your point about pari : if you want a newer
version, the right solution is to contribute upstream and get it
shipped. Don't fork.
Snark on #sagemath
--
You received this message because you are subscribed to the Google Groups
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.
