[sage-devel] Re: Finite Fields broken in sage-4.4

2010-04-26 Thread Simon King
Hi William! On Apr 26, 2:20 pm, William Stein wrote: > Yes, you should definitely import from sage.all or sage.rings.all > whenever possible, instead of directly from some module.  The API of > the "all.py" files are much more stable (especially the higher they > are in the directory structure).

Re: [sage-devel] Re: Finite Fields broken in sage-4.4

2010-04-26 Thread John Cremona
I think the problem here is that David Roe's major rewrite of the finite field code was (sensibly) split into several separate tickets, of which the first 2 or 3 have been merged in 4.4, while the rest are still awaiting review / work after review. And David tried hard to make each ticket in the s

Re: [sage-devel] Re: Finite Fields broken in sage-4.4

2010-04-26 Thread William Stein
On Mon, Apr 26, 2010 at 5:10 AM, Simon King wrote: > Hi David! > > On 26 Apr., 13:49, David Roe wrote: >> Marshall's right that you need to change where the import comes from, but >> you actually want >> from sage.rings.finite_rings.constructor import FiniteField >> >> The class in sage.rings.fin

[sage-devel] Re: Finite Fields broken in sage-4.4

2010-04-26 Thread Jason Grout
On 04/26/2010 06:49 AM, David Roe wrote: Marshall's right that you need to change where the import comes from, but you actually want from sage.rings.finite_rings.constructor import FiniteField Is this a compatibility issue that should have a deprecation period? I wonder how many people it wi

[sage-devel] Re: Finite Fields broken in sage-4.4

2010-04-26 Thread Simon King
Hi David! On 26 Apr., 13:49, David Roe wrote: > Marshall's right that you need to change where the import comes from, but > you actually want > from sage.rings.finite_rings.constructor import FiniteField > > The class in sage.rings.finite_rings.finite_field_base is the base class, > whereas Finit

Re: [sage-devel] Re: Finite Fields broken in sage-4.4

2010-04-26 Thread David Roe
Marshall's right that you need to change where the import comes from, but you actually want from sage.rings.finite_rings.constructor import FiniteField The class in sage.rings.finite_rings.finite_field_base is the base class, whereas FiniteField in constructor is the UniqueFactory that makes finit

[sage-devel] Re: Finite Fields broken in sage-4.4

2010-04-26 Thread Simon King
Hi Marshall! On 26 Apr., 13:32, mhampton wrote: > A number of tickets such as #8218 refactored quite a bit of finite > field code.  The FiniteField class you were using can be imported by: > > from sage.rings.finite_rings.finite_field_base import FiniteField Thank you! And "from sage.all" works

[sage-devel] Re: Finite Fields broken in sage-4.4

2010-04-26 Thread mhampton
A number of tickets such as #8218 refactored quite a bit of finite field code. The FiniteField class you were using can be imported by: from sage.rings.finite_rings.finite_field_base import FiniteField -Marshall On Apr 26, 6:08 am, Simon King wrote: > Hi! > > I just upgraded to sage-4.4. I get