some special cases of dimension_new_cusp_forms are broken

[was]

Ticket #11601 made some changes to congruence subgroups and broke some stuff that dims.py depends on, which resulted in breaking dimension_new_cusp_forms when the character is trivial and the level is 1 or 2.

sage: dimension_new_cusp_forms(DirichletGroup(1)(1), 12)
---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)

/Users/wstein/db/mfdb/<ipython console> in <module>()

/Users/wstein/sage/install/sage-5.0.beta2/local/lib/python2.7/site-packages/sage/modular/dims.pyc in dimension_new_cusp_forms(X, k, p)
    304         return X.dimension_new_cusp_forms(k,p=p)
    305     elif isinstance(X, dirichlet.DirichletCharacter):
--> 306         return Gamma1(X.modulus()).dimension_new_cusp_forms(k,eps=X,p=p)
    307     elif isinstance(X, (int,long,Integer)):
    308         return Gamma0(X).dimension_new_cusp_forms(k,p=p)

TypeError: dimension_new_cusp_forms() got an unexpected keyword argument 'eps'
sage: dimension_new_cusp_forms(DirichletGroup(2)(1), 12)
---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)

/Users/wstein/db/mfdb/<ipython console> in <module>()

/Users/wstein/sage/install/sage-5.0.beta2/local/lib/python2.7/site-packages/sage/modular/dims.pyc in dimension_new_cusp_forms(X, k, p)
    304         return X.dimension_new_cusp_forms(k,p=p)
    305     elif isinstance(X, dirichlet.DirichletCharacter):
--> 306         return Gamma1(X.modulus()).dimension_new_cusp_forms(k,eps=X,p=p)
    307     elif isinstance(X, (int,long,Integer)):
    308         return Gamma0(X).dimension_new_cusp_forms(k,p=p)

TypeError: dimension_new_cusp_forms() got an unexpected keyword argument 'eps'

It seems like bigger levels work fine...

sage: dimension_new_cusp_forms(DirichletGroup(3)(1), 12)
1
sage: dimension_new_cusp_forms(DirichletGroup(4)(1), 12)
1

[jen]

Looks great. All tests pass.

[davidloeffler]

Just for the record: that "somebody" was me, at ticket #11601. From that ticket description:

"Some "rationalisation" is also included: e.g. one can no longer create Gamma1(1) or GammaH(11, [2]), which previously existed as less-functional duplicates of SL2Z and Gamma0(11)."