get rid of some more useless .keys in combinat

[fchapoton]

sequel of #21296 and #21304

with an eye towards python3 compatibility

when iterating over a dict, one does not need to call .keys()

and when asking for the list of keys, it is better to call list(d) for compatibility with py3

Removing .keys() when possible is performed in some files in combinat folder

Component: combinatorics

Author: Frédéric Chapoton

Branch/Commit: 8021e24

Reviewer: Travis Scrimshaw

Issue created by migration from https://trac.sagemath.org/ticket/21379

[fchapoton]

New commits:

c92288b

more .keys() removal inside combinat

[fchapoton]

Branch: u/chapoton/21379

[fchapoton]

Commit: c92288b

[tscrim]

comment:2

Two stupid things. You made a PEP8 change which made me notice that while-we-are-at-it^(TM), could you make this change:

-        if len(positions) == 0:
+        if not positions:

Also, in CyclicSievingPolynomial, you can replace keys by orbit_sizes.

[tscrim]

Reviewer: Travis Scrimshaw

[sagetrac-git]

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

f9ab401

trac 21379 some details after review

[sagetrac-git]

Changed commit from c92288b to f9ab401

[fchapoton]

comment:5

done. And thank you again for taking time to do so many reviews.

[tscrim]

comment:6

It's nothing compared to what you've been doing.

[vbraun]

comment:7

See patchbot

[sagetrac-git]

Changed commit from f9ab401 to 8021e24

[sagetrac-git]

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

cae55cf	Merge branch 'u/chapoton/21379' in 7.4.b3
8021e24	trac 21379 correct cyclic sieving

[fchapoton]

comment:9

ok, I have corrected the problem.

[vbraun]

Changed branch from u/chapoton/21379 to 8021e24