py3: fix doctests in hyperbolicity and graph_coloring

[dcoudert]

• hyperbolicity: we avoid sorting vertex ids before returning the certificate, and we sort these ids in the doctests.
• graph_coloring: we sort the output of some doctests

Component: graph theory

Author: David Coudert

Branch/Commit: c3678ad

Reviewer: Jean-Florent Raymond

Issue created by migration from https://trac.sagemath.org/ticket/27160

[dcoudert]

New commits:

7dd655f	hyperbolicity: mark doctest as py2 and py3
4784cc2	graph_coloring: mark doctest as py2 and py3

[dcoudert]

Branch: u/dcoudert/27160_hyperbolicity_and_graph_coloring

[dcoudert]

Commit: 4784cc2

[sagetrac-git]

Changed commit from 4784cc2 to 2490bc9

[sagetrac-git]

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

0cccb17	trac #27160: Merged with 8.7.beta4
5e58fe7	trac #27160: avoid sorting vertex ids when returning certificate
2490bc9	trac #27160: avoid tag py2 and py3 in graph_coloring

[sagetrac-git]

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

c3678ad

trac #27160: avoid another py2/py3 tag in graph_coloring

[sagetrac-git]

Changed commit from 2490bc9 to c3678ad

[dcoudert]

comment:4

I changed the proposed fixes to avoid # py2 and #py 3 tags.

• hyperbolicity: it is more robust to avoid sorting vertex ids of the certificate before returning it, and to sort these ids in the doctests
• graph_coloring: we now sort the output of some doctests

[jfraymond]

comment:5

Note that the lines you edited about approximation of hyperbolicity for grids an cycles now go over 80 characters. (also true for other lines of both files, but outside of the aim of this ticket)

Is it a problem? If not I am fine with the patch.

[dcoudert]

comment:6

It's not a problem (at least for me).

[jfraymond]

Reviewer: Jean-Florent Raymond

[vbraun]

Changed branch from u/dcoudert/27160_hyperbolicity_and_graph_coloring to c3678ad