[Debian-science-sagemath] Overview of doctest failures

[Jeroen Demeyer]

On 2017-10-18 16:10, Tobias Hansen wrote:
> Is there anything we should keep in mind when trying to use that with pari 2.9?

First of all, you *must* upgrade cypari2 in case you haven't done that: 
https://trac.sagemath.org/ticket/23929

I made sure that Sage is still compatible with pari 2.9. Here I define 
"compatible" to mean that Sage builds and that all computations are 
mathematically correct with both pari 2.9 and pari master.

This is *not* the same as all doctests passing, since one mathematically 
correct output may be replaced by a different mathematically correct 
one. In particular, you will need to revert the doctest changes from 
https://trac.sagemath.org/ticket/23544 if you use pari stable.