EllipticCurve(E.a_invariants()) doesn't work

[williamstein]

At some point E.a_invariants() was changed to return a 5-tuple instead of a list of 5 numbers. I have no idea why. But this means that:

sage: E = EllipticCurve([1..5])
sage: EllipticCurve(E.a_invariants())
TypeError: invalid input to EllipticCurve constructor

Ugh. Fix this, i.e., make the constructor take a tuple or a list.

Component: elliptic curves

Author: William Stein

Reviewer: Robert Bradshaw

Merged: sage-5.0.beta5

Issue created by migration from https://trac.sagemath.org/ticket/12517

[williamstein]

Attachment: trac_12715.patch.gz

[williamstein]

Author: William Stein

[JohnCremona]

comment:3

See #4262 for the origin of this change, after a report by was (!) and #4264 by the patch which implemented it written mosstly by me. As it says at that ticket, I did (of course) try to get the EllipticCurve constructor to accept tuples, but for some reason at the time I failed to get it to work. So either you people are cleverer now than you or I were then, or other changes have made that easier.

Of course I approve of the change!

[williamstein]

comment:4

All my patch does is change an "isinstance(., list)" to "isinstance(., (list, tuple))" in one line. That's it. You worry me.

[JohnCremona]

comment:5

Replying to @williamstein:

All my patch does is change an "isinstance(., list)" to "isinstance(., (list, tuple))" in one line. That's it. You worry me.

Don't worry, it was over 3 years ago and needed masses of doctest output changes which is the main thing I remember doing. Looking at my old comments in #4264 it looks as if I was trying to change {{{init}} functions in classes instead of what you did.

[jdemeyer]

Reviewer: Robert Bradshaw

[jdemeyer]

Merged: sage-5.0.beta5