[with bundle, needs additional review] Defined Hom parent of group homomorphisms.

[kohel]

Defined parent of a group homomorphism such that the following example works (and similar for permutation groups):

sage: G.<x,y> = AbelianGroup(2,[2,3]); G 
Multiplicative Abelian Group isomorphic to C2 x C3
sage: H.<a,b,c> = AbelianGroup(3,[2,3,4]); H
Multiplicative Abelian Group isomorphic to C2 x C3 x C4
sage: phi = AbelianGroupMorphism(G,H,[x,y],[a,b])
sage: phi

AbelianGroup morphism:
  From: Multiplicative Abelian Group isomorphic to C2 x C3
  To:   Multiplicative Abelian Group isomorphic to C2 x C3 x C4
sage: phi.parent()
Set of Morphisms from Multiplicative Abelian Group isomorphic to C2 x C3 to Multiplicative Abelian Group isomorphic to C2 x C3 x C4 in Category of groups
sage: Hom(G,H) == phi.parent()
True

[ncalexan]

Apply aa5061b07eef. Documentation and tests are fantastic!

45b75671c677 and 25e03ebc5b0d appear to be the same. Whether it should be applied... I say yes, but I'm worried that it will fail on infinite domains.

02e052cfe50a should be applied, although I don't like leaving commented code around.

[ncalexan]

BTW, why isn't the motivating example a doctest in the attached patch?

[ncalexan]

#1840 is aa5061b07eef.

#1893 is 45b75671c677 and 25e03ebc5b0d.

So the only new patch is 02e052cfe50a.

[mabshoff]

There are various points raised in the previous comments which should be addressed.

Cheers,

Michael

[wdj]

tried to create patch based on sage-3.0.alphaa0 from code D Kohel sent

[wdj]

Maybe others could apply the bundle but I could not. I asked David Kohel for his code, which he emailed me. I tried to create a patch from that. It seems to pass sage -t (I cannot get sage -testall to work without locking up on this machine). Since I get

sage: G.<x,y> = AbelianGroup(2,[2,3]); G
Multiplicative Abelian Group isomorphic to C2 x C3
sage: H.<a,b,c> = AbelianGroup(3,[2,3,4]); H
Multiplicative Abelian Group isomorphic to C2 x C3 x C4
sage: phi = AbelianGroupMorphism(G,H,[x,y],[a,b])
sage: phi
<sage.groups.abelian_gps.abelian_group_morphism.AbelianGroupMorphism instance at 0x364cef0>
sage: phi.parent()
---------------------------------------------------------------------------
<type 'exceptions.AttributeError'>        Traceback (most recent call last)

/mnt/drive_hda1/sagefiles/sage-3.0.alpha0/<ipython console> in <module>()

<type 'exceptions.AttributeError'>: AbelianGroupMorphism instance has no attribute 'parent'

I apparently made a mistake somewhere....