| 1 | To test an alpha version of the upcoming FLINT release on Sage 5.1.beta2, install: |
| 2 | * the updated zn_poly spkg from #12433, |
| 3 | * the (functional but definitely not definitive) FLINT spkg at http://perso.telecom-paristech.fr/~flori/sage/flint-2.3.jp.spkg; |
| 4 | apply: |
| 5 | * [attachment:flint-2.3-sage-5.1.beta2.patch], |
| 6 | * [attachment:doctests.patch] (optional), |
| 7 | * [attachment:flint_stack_cleanup.patch], |
| 8 | * [attachment:n_factor.patch], |
| 9 | * [attachment:degree.patch], |
| 10 | * [attachment:sqrt_normalization_fix.patch], |
| 11 | * [attachment:headers.patch], |
| 12 | * [attachment:multrunc.patch], |
| 13 | * [attachment:fmpz_cleanup.patch], |
| 14 | * [attachment:jack.patch]. |
| 15 | |
| 16 | There remain some issues: |
| 17 | - a long test fails in sage/misc/cython.py because it uses a Sage specific file (fmpq_poly) which was integrated into FLINT 2, |
| 18 | - a long test timeouts in sage/rings/polynomial/polynomial_rational_flint.pyx, |
| 19 | - errors in sage/graphs/matchpoly.pyx caused by a bug in FLINT (only on 32 bits systems?). |