"maximum recursion depth exceeded" when doing change_ring(RR) on an elliptic curve

[williamstein]

Behold this bug:

K.<a> = NumberField(x^2-5)
EllipticCurve(K,[1..5]).change_ring(RR)

BOOM!

...
  File "coerce_maps.pyx", line 82, in sage.structure.coerce_maps.DefaultConvertMap_unique._call_ (sage/structure/coerce_maps.c:3345)
  File "coerce_maps.pyx", line 77, in sage.structure.coerce_maps.DefaultConvertMap_unique._call_ (sage/structure/coerce_maps.c:3248)
  File "/Users/wstein/sage/build/sage-5.0/local/lib/python2.7/site-packages/sage/rings/complex_field.py", line 286, in _element_constructor_
    if not isinstance(x, (real_mpfr.RealNumber, tuple)):
RuntimeError: maximum recursion depth exceeded while calling a Python object

Component: elliptic curves

Keywords: sd40.5

Author: R. Andrew Ohana

Reviewer: William Stein

Merged: sage-5.1.beta5

Issue created by migration from https://trac.sagemath.org/ticket/13005

[JohnCremona]

comment:1

It's nothing to do with elliptic curves. Try

sage: K.<a> = NumberField(x^2-5)
sage: RR(K(1))     

So the expected "natural" coercion from elements of K to elements of RR is blowing up, instead of using the default embedding.

[ohanar]

Changed keywords from none to sd40.5

[ohanar]

Author: R. Andrew Ohana

[roed314]

comment:3

Can you add doctests for number fields that already embeds into a real field (possibly of precision different from that of R)?

[ohanar]

comment:4

Replying to @roed314:

Can you add doctests for number fields that already embeds into a real field (possibly of precision different from that of R)?

Ok, done