pretty printing of vectors over callable symbolic rings

[jason]

This patch makes vectors of callable symbolic rings print nicer, in the form arguments mapsto vector

sage: f(x,y)=[3*x,e^x,2*x*y]
sage: f
(x, y) |--> (3*x, e^x, 2*x*y)

[jason]

Doctests need to be fixed up.

[jason]

The doctests depend on a patch in 4.4.2.alpha0.

I think the right way to change printing for a vector over callable symbolic expressions is to subclass as I did in the patch and override the _repr_ and _latex_ methods. Can was or robertwb comment on this?

[jason]

#5506 adds some further ideas about what to do about this class.

[robertwb]

Yep, that's the way to do it. The code looks good, but I only glanced at it quickly haven't tested it.

[rbeezer]

Hi Jason,

Nice patch.

1. Do you think this should be called "free_module_element_callable_symbolic_dense"? Technically these don't have to be vectors, and they do have entries over a ring? Yes, that's an unwieldy name. ;-) Just asking.

2. Do you want to add it into the documentation, doesn't seem to be included when I build the docs?

3. Docstring for _latex_ looks like it has an indentation problem that needs fixing.

Running tests right now.

Rob

[jason]

Replying to rbeezer:

Hi Jason,

Nice patch.

1. Do you think this should be called "free_module_element_callable_symbolic_dense"? Technically these don't have to be vectors, and they do have entries over a ring? Yes, that's an unwieldy name. ;-) Just asking.

Well, Sage thinks that it is a field:

sage: f(x,y)=x^2+y                             
sage: R=f.parent()
sage: R 
Callable function ring with arguments (x, y)
sage: R.is_field()
True

2. Do you want to add it into the documentation, doesn't seem to be included when I build the docs?

3. Docstring for _latex_ looks like it has an indentation problem that needs fixing.

Looks like (2) and (3) both probably need to be fixed.

[rbeezer]

Replying to jason:

Callable function ring with arguments (x, y) sage: R.is_field() True

And I trusted the _repr_ output?

Failing a few tests, in the obvious way, in

/sage/dev/devel/sage-main/sage/symbolic/expression.pyx
/sage/dev/devel/sage-main/sage/calculus/calculus.py

(and not done testing).

[rbeezer]

On 4.4.2.rc0 I get the following failures. The one for R is totally unrelated I believe, the others I saw are just the obvious differences in format for functions involved in this patch.

The following tests failed:

        sage -t  devel/sage-main/sage/symbolic/expression.pyx # 4 doctests failed
        sage -t  devel/sage-main/sage/calculus/calculus.py # 2 doctests failed   
        sage -t  devel/sage-main/sage/interfaces/r.py # 1 doctests failed        
        sage -t  devel/sage-main/sage/modules/free_module_element.pyx # 3 doctests failed

[jason]

apply on top of previous patches

[jason]

Okay, the doctest issues you mentioned should all be fixed now.

[rbeezer]

sage/calculus, sage/modules, sage/symbolic directories pass all tests, and documentation for vector_callable_symbolic_dense looks good.

Positive review.