py3: fixing doctests in elliptic_curves/cm.py

[chapoton]

by more sorting of discriminants

[chapoton]

New commits:

​9d5f2f9

py3: fixes in elliptic_curves/cm.py

[cremona]

All looks fine apart from this:

-    dlist = sum([v for h, v in iteritems(discriminants_with_bounded_class_number(K.degree(), proof=proof))], [])
+    dlist = sorted(Df for v in discriminants_with_bounded_class_number(K.degree(), proof=proof).values() for Df in v)

whose syntax looks suspicious to me.

[chapoton]

Hmm. Why suspicious ? The "sorted" is the only real change we are making here. If my understanding of the mathematics involved is ok, this should not be a problem.

Or maybe you are troubled by the double for ? This is a perfectly valid python syntax. We can chain the "for", the second one depending on the argument of the first. Here the effect is to flatten a list of lists.

​https://www.python.org/dev/peps/pep-0289/

[cremona]

Replying to chapoton:

Hmm. Why suspicious ? The "sorted" is the only real change we are making here. If my understanding of the mathematics involved is ok, this should not be a problem.

Or maybe you are troubled by the double for ? This is a perfectly valid python syntax. We can chain the "for", the second one depending on the argument of the first. Here the effect is to flatten a list of lists.

​https://www.python.org/dev/peps/pep-0289/

Thanks for the lesson! I had never seen a double for so thought it was a mistake (and did not have time to check).

[git]

Branch pushed to git repo; I updated commit sha1. New commits:

​5403ce7

trac 26914 fix pyflakes warning

[chapoton]

green bot, ready for review