Bug in IncidenceStructure.automorphism_group when the ground set consists of pairs

[nathanncohen]

Before

sage: designs.IncidenceStructure([[(1,2),(3,4)]]).automorphism_group()
...
ValueError: Invalid permutation vector: [(3, 4), (1, 2)]

After

sage: designs.IncidenceStructure([[(1,2),(3,4)]]).automorphism_group()
Permutation Group with generators [((1,2),(3,4))]

CC: @videlec @KPanComputes @dimpase

Component: combinatorial designs

Author: Nathann Cohen

Branch: aad8ee5

Reviewer: Dima Pasechnik

Issue created by migration from https://trac.sagemath.org/ticket/16741

[nathanncohen]

Branch: u/ncohen/16741

[nathanncohen]

comment:1

This really was triky to fix. The permutation code tries to handle several incompatible input i.e. tries to guess whether input is given in list notation or cycle notation, and guesses that from the type list/tuple of input.

Plus some permutation code WANTS permutations on 1...n, others on 0,...,n-1 and others handle everything...

[sagetrac-git]

Commit: aad8ee5

[sagetrac-git]

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

aad8ee5

trac #16741: Bug in IncidenceStructure.relabel when the ground set consists of pairs

[dimpase]

comment:3

when I read the permgroups code, I think: "FORTRAN" :-)

[dimpase]

comment:5

LGTM

[nathanncohen]

comment:6

Thaaaaaaaannks !

Nathann

[vbraun]

comment:7

name

[dimpase]

Reviewer: Dima Pasechnik

[vbraun]

Changed branch from u/ncohen/16741 to aad8ee5

[vbraun]

comment:10

PDF docs:

LaTeX Warning: Hyper reference `sage/combinat/designs/incidence_structures:sage
.combinat.designs.incidence_structures.IncidenceStructure.is_t_design' on page 
244 undefined on input line 20504.

[244] [245] [246]
! Double subscript.
l.20746 ...tion, see the ;wikipedia:$Packing_in_a_
                                                  hypergraph$.

[vbraun]

Changed commit from aad8ee5 to none

[vbraun]

comment:11

wrong ticket