Can't construct a rational fraction from a symbolic one.

[hivert]

Given a symbolic expression which is a rational fraction sage refuse to convert it to a element of the Field of rational fraction:

hivert@boxen:~$ sage
----------------------------------------------------------------------
| Sage Version 4.2.1, Release Date: 2009-11-14                       |
| Type notebook() for the GUI, and license() for information.        |
----------------------------------------------------------------------
sage: fr = (1+x)/(1+x+x^2)
sage: Fld = FractionField(PolynomialRing(QQ,x))
sage: Fld(fr)
ERROR: An unexpected error occurred while tokenizing input
The following traceback may be corrupted or invalid
The error message is: ('EOF in multi-line statement', (1181, 0))

[...]

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)

[...]

/usr/local/sage/local/lib/python2.6/site-packages/sage/rings/polynomial/polynomial_ring.pyc in _element_constructor_(self, x, check, is_gen, construct, **kwds)
    304                 x = x.numerator() * x.denominator().inverse_of_unit()
    305             else:
--> 306                 raise TypeError, "denominator must be a unit"
    307
    308         elif isinstance(x, pari_gen):

TypeError: denominator must be a unit

It seems that it needs to convert is to a polynomial. Of course if one convert numerator and denominator separately everything is Ok:

sage: Fld((1+x))/(1+x+x^2)
(x + 1)/(x^2 + x + 1)

I'm not sure about which component should be selected... Is it algebra, calculus or coercion...

Florent

CC: @burcin

Component: coercion

Keywords: Fraction Field

Author: Robert Bradshaw

Reviewer: Florent Hivert, Burcin Erocal

Merged: sage-4.6.alpha2

Issue created by migration from https://trac.sagemath.org/ticket/7741

[robertwb]

comment:2

Attachment: 7741-symbolic-frac.patch.gz

[hivert]

Author: Robert Bradshaw

[hivert]

Reviewer: Florent Hivert

[rlmill]

Work Issues: needs rebase

[loefflerd]

Attachment: trac_7741-symbolic-frac-rebase.patch.gz

rebased to 4.4.4

[ghost]

comment:6

William points out that denominator is 1 will fail unless denominator is the Python int 1, and nothing else...

Is this really preferable to denominator == 1?

PS, Passes tests in sage/rings/fraction_field_element.pyx

[robertwb]

comment:7

Attachment: 7741-symbolic-frac-fixed.patch.gz

Actually, the "is 1" was intentional, as I wasn't thinking of taking that path if the user passed in a ring element (as opposed to the default value) and didn't want to make the default value None and handle it everywhere. In retrospect, I think it's fine for it to be equal to 1.

[burcin]

comment:8

apply only attachment: 7741-symbolic-frac-fixed.patch.

[burcin]

Changed work issues from needs rebase to none

[burcin]

Changed reviewer from Florent Hivert to Florent Hivert, Burcin Erocal

[qed777]

Merged: sage-4.6.alpha2