problem parsing arguments to NumberField.order()

[ncalexan]

sage: y = ZZ['y'].0; K = NumberField(y^4 + 4*y^2 + 2, 'a'); K
Number Field in a with defining polynomial y^4 + 4*y^2 + 2
sage: B = K.integral_basis()
sage: B
[1, a, a^2, a^3]
sage: K.order(B)
Order in Number Field in a with defining polynomial y^4 + 4*y^2 + 2
sage: K.order(gens=B)
+Infinity

CC: @ncalexan @robertwb @mwhansen

Component: number fields

Keywords: number field order arguments

Author: Craig Citro

Reviewer: Mike Hansen

Merged: sage-4.3.2.alpha0

Issue created by migration from https://trac.sagemath.org/ticket/2480

[craigcitro]

comment:2

This wasn't so bad -- the problem was that gens= put the list of gens in the kwds dict, instead of in the *-argument. I've attached a fix, but I'd love for someone to tell me if deleting gens out of the kwds dict is sufficiently pythonic. (If you don't, the call to absolute_order_from_ring_generators rightfully complains that gens is specified twice.) Another option would be to reassign kwds['dict'] at the end, but I don't think that's any nicer. (In fact, that might be epsilon slower, since it's another argument to unpack from the dictionary on the other side.)

Also, the comment block in the docstring really looks like something was accidentally cut off at some point. Amusingly, this isn't the case: I actually dug through the hg logs, and it was really committed just like that.

[craigcitro]

Author: Craig Citro

[craigcitro]

comment:3

Mike and Robert, I'm adding you on the cc so that you can tell me if I'm being sufficiently pythonic. :)

[mwhansen]

comment:4

Hey Craig,

gens = kwds.pop('gens')

is probably better.

[mwhansen]

comment:5

Err,

gens = kwds.pop('gens', args)