bug in unit part of factorizations of multivariate polynomials

[williamstein]

Behold this behavior:

sage: R.<a,b,c,d> = QQ[]
sage: f =  (-1) * (a - d) * (-a + b) * (b - d) * (a - c) * (b - c) * (c - d)
sage: f.factor()
(-1) * (a - d) * (-a + b) * (b - d) * (a - c) * (b - c) * (c - d)
sage: F = f.factor()
sage: F[0][0]
-1
sage: F.unit_part ()
1

However it should be that F.unit_part() is -1 and F[0][0] is a-d.

Component: commutative algebra

Issue created by migration from https://trac.sagemath.org/ticket/1391

[sagetrac-mabshoff]

comment:1

Bug Day material?

Cheers,

Michael

[malb]

Attachment: trac_1391.patch.gz

[malb]

comment:2

The attached patch fixes the issue

[ncalexan]

comment:3

Patch looks reasonable. The parameter to factor() is not standard sage but seems appropriate. I say apply!

[sagetrac-mabshoff]

comment:4

The patch no longer applies to me in 2.10.1.alpha0, so it probably needs just a rebase:

sage-2.10.1.alpha0/devel/sage$ patch -p1 --dry-run < trac_1391.patch
patching file sage/rings/polynomial/multi_polynomial_libsingular.pyx
Hunk #1 FAILED at 3096.
Hunk #2 succeeded at 3159 (offset 22 lines).
1 out of 2 hunks FAILED -- saving rejects to file sage/rings/polynomial/multi_polynomial_libsingular.pyx.rej

Cheers,

Michael

[williamstein]

Attachment: trac-1391-apply_after_patch_rej_from_other_patch.patch.gz

[sagetrac-mabshoff]

comment:5

Merged in Sage 2.10.1.alpha0