Ribbon Graphs

This file implements objects called ribbon graphs. These are graphs together with a cyclic ordering of the darts adjacent to each vertex. This data allows us to unambiguously “thicken” the ribbon graph to an orientable surface with boundary. Also, every orientable surface with non-empty boundary is the thickening of a ribbon graph.

AUTHORS:

• Pablo Portilla (2016)

class sage.geometry.ribbon_graph.RibbonGraph(sigma, rho)

Bases: sage.structure.sage_object.SageObject, sage.structure.unique_representation.UniqueRepresentation

A ribbon graph codified as two elements of a certain permutation group.

A comprehensive introduction on the topic can be found in the beginning of [GGD2011] Chapter 4. More concretely, we will use a variation of what is called in the reference “The permutation representation pair of a dessin”. Note that in that book, ribbon graphs are called “dessins d’enfant”. For the sake on completeness we reproduce an adapted version of that introduction here.

Brief introduction

Let \(\Sigma\) be an orientable surface with non-empty boundary and let \(\Gamma\) be the topological realization of a graph that is embedded in \(\Sigma\) in such a way that the graph is a strong deformation retract of the surface.

Let \(v(\Gamma)\) be the set of vertices of \(\Gamma\), suppose that these are white vertices. Now we mark black vertices in an interior point of each edge. In this way we get a bipartite graph where all the black vertices have valency 2 and there is no restriction on the valency of the white vertices. We call the edges of this new graph darts (sometimes they are also called half edges of the original graph). Observe that each edge of the original graph is formed by two darts.

Given a white vertex \(v \in v(\Gamma)\), let \(d(v)\) be the set of darts adjacent to \(v\). Let \(D(\Gamma)\) be the set of all the darts of \(\Gamma\) and suppose that we enumerate the set \(D(\Gamma)\) and that it has \(n\) elements.

With the orientation of the surface and the embedding of the graph in the surface we can produce two permutations:

• A permutation that we denote by \(\sigma\). This permutation is a product of as many cycles as white vertices (that is vertices in \(\Gamma\)). For each vertex consider a small topological circle around it in \(\Sigma\). This circle intersects each adjacent dart once. The circle has an orientation induced by the orientation on \(\Sigma\) and so defines a cycle that sends the number associated to one dart to the number associated to the next dart in the positive orientation of the circle.
• A permutation that we denote by \(\rho\). This permutation is a product of as many \(2\)-cycles as edges has \(\Gamma\). It just tells which two darts belong to the same edge.

Abstract definition

Consider a graph \(\Gamma\) (not a priori embedded in any surface). Now we can again consider one vertex in the interior of each edge splitting each edge in two darts. We label the darts with numbers.

We say that a ribbon structure on \(\Gamma\) is a set of two permutations \((\sigma, \rho)\). Where \(\sigma\) is formed by as many disjoint cycles as vertices had \(\Gamma\). And each cycle is a cyclic ordering of the darts adjacent to a vertex. The permutation \(\rho\) just tell us which two darts belong to the same edge.

For any two such permutations there is a way of “thickening” the graph to a surface with boundary in such a way that the surface retracts (by a strong deformation retract) to the graph and hence the graph is embedded in the surface in a such a way that we could recover \(\sigma\) and \(\rho\).

INPUT:

• sigma – a permutation a product of disjoint cycles of any length; singletons (vertices of valency 1) need not be specified
• rho – a permutation which is a product of disjoint 2-cycles

Alternatively, one can pass in 2 integers and this will construct a ribbon graph with genus sigma and rho boundary components. See make_ribbon().

One can also construct the bipartite graph modeling the corresponding Brieskorn-Pham singularity by passing 2 integers and the keyword bipartite=True. See bipartite_ribbon_graph().

EXAMPLES:

Consider the ribbon graph consisting of just \(1\) edge and \(2\) vertices of valency \(1\):

sage: s0 = PermutationGroupElement('(1)(2)')
sage: r0 = PermutationGroupElement('(1,2)')
sage: R0 = RibbonGraph(s0,r0); R0
Ribbon graph of genus 0 and 1 boundary components

Consider a graph that has \(2\) vertices of valency \(3\) (and hence \(3\) edges). That is represented by the following two permutations:

sage: s1 = PermutationGroupElement('(1,3,5)(2,4,6)')
sage: r1 = PermutationGroupElement('(1,2)(3,4)(5,6)')
sage: R1 = RibbonGraph(s1, r1); R1
Ribbon graph of genus 1 and 1 boundary components

By drawing the picture in a piece of paper, one can see that its thickening has only \(1\) boundary component. Since the thickening is homotopically equivalent to the graph and the graph has Euler characteristic \(-1\), we find that the thickening has genus \(1\):

sage: R1.number_boundaries()
1
sage: R1.genus()
1

The following example corresponds to the complete bipartite graph of type \((2,3)\), where we have added one more edge \((8,15)\) that ends at a vertex of valency \(1\). Observe that it is not necessary to specify the vertex \((15)\) of valency \(1\) when we define sigma:

sage: s2 = PermutationGroupElement('(1,3,5,8)(2,4,6)')
sage: r2 = PermutationGroupElement('(1,2)(3,4)(5,6)(8,15)')
sage: R2 = RibbonGraph(s2, r2); R1
Ribbon graph of genus 1 and 1 boundary components
sage: R2.sigma()
(1,3,5,8)(2,4,6)

This example is constructed by taking the bipartite graph of type \((3,3)\):

sage: s3 = PermutationGroupElement('(1,2,3)(4,5,6)(7,8,9)(10,11,12)(13,14,15)(16,17,18)')
sage: r3 = PermutationGroupElement('(1,16)(2,13)(3,10)(4,17)(5,14)(6,11)(7,18)(8,15)(9,12)')
sage: R3 = RibbonGraph(s3, r3); R3
Ribbon graph of genus 1 and 3 boundary components

The labeling of the darts can omit some numbers:

sage: s4 = PermutationGroupElement('(3,5,10,12)')
sage: r4 = PermutationGroupElement('(3,10)(5,12)')
sage: R4 = RibbonGraph(s4,r4); R4
Ribbon graph of genus 1 and 1 boundary components

The next example is the complete bipartite graph of type \((3,3)\), where we have added an edge that ends at a vertex of valency 1:

sage: s5 = PermutationGroupElement('(1,2,3)(4,5,6)(7,8,9)(10,11,12)(13,14,15)(16,17,18,19)')
sage: r5 = PermutationGroupElement('(1,16)(2,13)(3,10)(4,17)(5,14)(6,11)(7,18)(8,15)(9,12)(19,20)')
sage: R5 = RibbonGraph(s5,r5); R5
Ribbon graph of genus 1 and 3 boundary components
sage: C = R5.contract_edge(9); C
Ribbon graph of genus 1 and 3 boundary components
sage: C.sigma()
(1,2,3)(4,5,6)(7,8,9)(10,11,12)(13,14,15)(16,17,18)
sage: C.rho()
(1,16)(2,13)(3,10)(4,17)(5,14)(6,11)(7,18)(8,15)(9,12)
sage: S = R5.reduced(); S
Ribbon graph of genus 1 and 3 boundary components
sage: S.sigma()
(5,6,8,9,14,15,11,12)
sage: S.rho()
(5,14)(6,11)(8,15)(9,12)
sage: R5.boundary()
[[1, 16, 17, 4, 5, 14, 15, 8, 9, 12, 10, 3],
 [2, 13, 14, 5, 6, 11, 12, 9, 7, 18, 19, 20, 20, 19, 16, 1],
 [3, 10, 11, 6, 4, 17, 18, 7, 8, 15, 13, 2]]
sage: S.boundary()
[[5, 14, 15, 8, 9, 12], [6, 11, 12, 9, 14, 5], [8, 15, 11, 6]]
sage: R5.homology_basis()
[[[5, 14], [13, 2], [1, 16], [17, 4]],
 [[6, 11], [10, 3], [1, 16], [17, 4]],
 [[8, 15], [13, 2], [1, 16], [18, 7]],
 [[9, 12], [10, 3], [1, 16], [18, 7]]]
sage: S.homology_basis()
[[[5, 14]], [[6, 11]], [[8, 15]], [[9, 12]]]

We construct a ribbon graph corresponding to a genus 0 surface with 5 boundary components:

sage: R = RibbonGraph(0, 5); R
Ribbon graph of genus 0 and 5 boundary components
sage: R.sigma()
(1,9,7,5,3)(2,4,6,8,10)
sage: R.rho()
(1,2)(3,4)(5,6)(7,8)(9,10)

We construct the Brieskorn-Pham singularity of type \((2,3)\):

sage: B23 = RibbonGraph(2, 3, bipartite=True); B23
Ribbon graph of genus 1 and 1 boundary components
sage: B23.sigma()
(1,2,3)(4,5,6)(7,8)(9,10)(11,12)
sage: B23.rho()
(1,8)(2,10)(3,12)(4,7)(5,9)(6,11)

boundary()

Return the labeled boundaries of self.

If you cut the thickening of the graph along the graph. you get a collection of cylinders (recall that the graph was a strong deformation retract of the thickening). In each cylinder one of the boundary components has a labelling of its edges induced by the labelling of the darts.

OUTPUT:

A list of lists. The number of inner lists is the number of boundary components of the surface. Each list in the list consists of an ordered tuple of numbers, each number comes from the number assigned to the corresponding dart before cutting.

EXAMPLES:

We start with a ribbon graph whose thickening has one boundary component. We compute its labeled boundary, then reduce it and compute the labeled boundary of the reduced ribbon graph:

sage: s1 = PermutationGroupElement('(1,3,5)(2,4,6)')
sage: r1 = PermutationGroupElement('(1,2)(3,4)(5,6)')
sage: R1 = RibbonGraph(s1,r1); R1
Ribbon graph of genus 1 and 1 boundary components
sage: R1.boundary()
[[1, 2, 4, 3, 5, 6, 2, 1, 3, 4, 6, 5]]
sage: H1 = R1.reduced(); H1
Ribbon graph of genus 1 and 1 boundary components
sage: H1.sigma()
(3,5,4,6)
sage: H1.rho()
(3,4)(5,6)
sage: H1.boundary()
[[3, 4, 6, 5, 4, 3, 5, 6]]

We now consider a ribbon graph whose thickening has 3 boundary components. Also observe that in one of the labeled boundary components, a numbers appears twice in a row. That is because the ribbon graph has a vertex of valency 1:

sage: s2=PermutationGroupElement('(1,2,3)(4,5,6)(7,8,9)(10,11,12)(13,14,15)(16,17,18,19)')
sage: r2=PermutationGroupElement('(1,16)(2,13)(3,10)(4,17)(5,14)(6,11)(7,18)(8,15)(9,12)(19,20)')
sage: R2 = RibbonGraph(s2,r2)
sage: R2.number_boundaries()
3
sage: R2.boundary()
[[1, 16, 17, 4, 5, 14, 15, 8, 9, 12, 10, 3],
 [2, 13, 14, 5, 6, 11, 12, 9, 7, 18, 19, 20, 20, 19, 16, 1],
 [3, 10, 11, 6, 4, 17, 18, 7, 8, 15, 13, 2]]

contract_edge(k)

Return the ribbon graph resulting from the contraction of the k-th edge in self.

For a ribbon graph \((\sigma, \rho)\), we contract the edge corresponding to the \(k\)-th transposition of \(\rho\).

INPUT:

• k – non-negative integer; the position in \(\rho\) of the transposition that is going to be contracted

OUTPUT:

• a ribbon graph resulting from the contraction of that edge

EXAMPLES:

We start again with the one-holed torus ribbon graph:

sage: s1 = PermutationGroupElement('(1,3,5)(2,4,6)')
sage: r1 = PermutationGroupElement('(1,2)(3,4)(5,6)')
sage: R1 = RibbonGraph(s1,r1); R1
Ribbon graph of genus 1 and 1 boundary components
sage: S1 = R1.contract_edge(1); S1
Ribbon graph of genus 1 and 1 boundary components
sage: S1.sigma()
(1,6,2,5)
sage: S1.rho()
(1,2)(5,6)

However, this ribbon graphs is formed only by loops and hence it cannot be longer reduced, we get an error if we try to contract a loop:

sage: S1.contract_edge(1)
Traceback (most recent call last):
...
ValueError: the edge is a loop and cannot be contracted

In this example, we consider a graph that has one edge (19,20) such that one of its ends is a vertex of valency 1. This is the vertex (20) that is not specified when defining \(\sigma\). We contract precisely this edge and get a ribbon graph with no vertices of valency 1:

sage: s2 = PermutationGroupElement('(1,2,3)(4,5,6)(7,8,9)(10,11,12)(13,14,15)(16,17,18,19)')
sage: r2 = PermutationGroupElement('(1,16)(2,13)(3,10)(4,17)(5,14)(6,11)(7,18)(8,15)(9,12)(19,20)')
sage: R2 = RibbonGraph(s2,r2); R2
Ribbon graph of genus 1 and 3 boundary components
sage: R2.sigma()
(1,2,3)(4,5,6)(7,8,9)(10,11,12)(13,14,15)(16,17,18,19)
sage: R2c = R2.contract_edge(9); R2; R2c.sigma(); R2c.rho()
Ribbon graph of genus 1 and 3 boundary components
(1,2,3)(4,5,6)(7,8,9)(10,11,12)(13,14,15)(16,17,18)
(1,16)(2,13)(3,10)(4,17)(5,14)(6,11)(7,18)(8,15)(9,12)

extrude_edge(vertex, dart1, dart2)

Return a ribbon graph resulting from extruding an edge from a vertex, pulling from it, all darts from dart1 to dart2 including both.

INPUT:

• vertex – the position of the vertex in the permutation \(\sigma\), which must have valency at least 2
• dart1 – the position of the first in the cycle corresponding to vertex
• dart2 – the position of the second dart in the cycle corresponding to vertex

OUTPUT:

A ribbon graph resulting from extruding a new edge that pulls from vertex a new vertex that is, now, adjacent to all the darts from dart1``to ``dart2 (not including dart2) in the cyclic ordering given by the cycle corresponding to vertex. Note that dart1 may be equal to dart2 allowing thus to extrude a contractible edge from a vertex.

EXAMPLES:

We try several possibilities in the same graph:

sage: s1 = PermutationGroupElement('(1,3,5)(2,4,6)')
sage: r1 = PermutationGroupElement('(1,2)(3,4)(5,6)')
sage: R1 = RibbonGraph(s1,r1); R1
Ribbon graph of genus 1 and 1 boundary components
sage: E1 = R1.extrude_edge(1,1,2); E1
Ribbon graph of genus 1 and 1 boundary components
sage: E1.sigma()
(1,3,5)(2,8,6)(4,7)
sage: E1.rho()
(1,2)(3,4)(5,6)(7,8)
sage: E2 = R1.extrude_edge(1,1,3); E2
Ribbon graph of genus 1 and 1 boundary components
sage: E2.sigma()
(1,3,5)(2,8)(4,6,7)
sage: E2.rho()
(1,2)(3,4)(5,6)(7,8)

We can also extrude a contractible edge from a vertex. This new edge will end at a vertex of valency 1:

sage: E1p = R1.extrude_edge(0,0,0); E1p
Ribbon graph of genus 1 and 1 boundary components
sage: E1p.sigma()
(1,3,5,8)(2,4,6)
sage: E1p.rho()
(1,2)(3,4)(5,6)(7,8)

In the following example we first extrude one edge from a vertex of valency 3 generating a new vertex of valency 2. Then we extrude a new edge from this vertex of valency 2:

sage: s1 = PermutationGroupElement('(1,3,5)(2,4,6)')
sage: r1 = PermutationGroupElement('(1,2)(3,4)(5,6)')
sage: R1 = RibbonGraph(s1,r1); R1
Ribbon graph of genus 1 and 1 boundary components
sage: E1 = R1.extrude_edge(0,0,1); E1
Ribbon graph of genus 1 and 1 boundary components
sage: E1.sigma()
(1,7)(2,4,6)(3,5,8)
sage: E1.rho()
(1,2)(3,4)(5,6)(7,8)
sage: F1 = E1.extrude_edge(0,0,1); F1
Ribbon graph of genus 1 and 1 boundary components
sage: F1.sigma()
(1,9)(2,4,6)(3,5,8)(7,10)
sage: F1.rho()
(1,2)(3,4)(5,6)(7,8)(9,10)

genus()

Return the genus of the thickening of self.

OUTPUT:

• g – non-negative integer representing the genus of the thickening of the ribbon graph

EXAMPLES:

sage: s1 = PermutationGroupElement('(1,3,5)(2,4,6)')
sage: r1 = PermutationGroupElement('(1,2)(3,4)(5,6)')
sage: R1 = RibbonGraph(s1,r1)
sage: R1.genus()
1

sage: s3=PermutationGroupElement('(1,2,3)(4,5,6)(7,8,9)(10,11,12)(13,14,15,16)(17,18,19,20)(21,22,23,24)')
sage: r3=PermutationGroupElement('(1,21)(2,17)(3,13)(4,22)(7,23)(5,18)(6,14)(8,19)(9,15)(10,24)(11,20)(12,16)')
sage: R3 = RibbonGraph(s3,r3); R3.genus()
3

homology_basis()

Return an oriented basis of the first homology group of the graph.

OUTPUT:

• A 2-dimensional array of ordered edges in the graph (given by pairs). The length of the first dimension is \(\mu\). Each row corresponds to an element of the basis and is a circle contained in the graph.

EXAMPLES:

sage: R = RibbonGraph(0,6); R
Ribbon graph of genus 0 and 6 boundary components
sage: R.mu()
5
sage: R.homology_basis()
[[[3, 4], [2, 1]],
 [[5, 6], [2, 1]],
 [[7, 8], [2, 1]],
 [[9, 10], [2, 1]],
 [[11, 12], [2, 1]]]

sage: R = RibbonGraph(1,1); R
Ribbon graph of genus 1 and 1 boundary components
sage: R.mu()
2
sage: R.homology_basis()
[[[2, 5], [4, 1]], [[3, 6], [4, 1]]]
sage: H = R.reduced(); H
Ribbon graph of genus 1 and 1 boundary components
sage: H.sigma()
(2,3,5,6)
sage: H.rho()
(2,5)(3,6)
sage: H.homology_basis()
[[[2, 5]], [[3, 6]]]

sage: s3 = PermutationGroupElement('(1,2,3,4,5,6,7,8,9,10,11,27,25,23)(12,24,26,28,13,14,15,16,17,18,19,20,21,22)')
sage: r3 = PermutationGroupElement('(1,12)(2,13)(3,14)(4,15)(5,16)(6,17)(7,18)(8,19)(9,20)(10,21)(11,22)(23,24)(25,26)(27,28)')
sage: R3 = RibbonGraph(s3,r3); R3
Ribbon graph of genus 5 and 4 boundary components
sage: R3.mu()
13
sage: R3.homology_basis()
[[[2, 13], [12, 1]],
 [[3, 14], [12, 1]],
 [[4, 15], [12, 1]],
 [[5, 16], [12, 1]],
 [[6, 17], [12, 1]],
 [[7, 18], [12, 1]],
 [[8, 19], [12, 1]],
 [[9, 20], [12, 1]],
 [[10, 21], [12, 1]],
 [[11, 22], [12, 1]],
 [[23, 24], [12, 1]],
 [[25, 26], [12, 1]],
 [[27, 28], [12, 1]]]
sage: H3 = R3.reduced(); H3
Ribbon graph of genus 5 and 4 boundary components
sage: H3.sigma()
(2,3,4,5,6,7,8,9,10,11,27,25,23,24,26,28,13,14,15,16,17,18,19,20,21,22)
sage: H3.rho()
(2,13)(3,14)(4,15)(5,16)(6,17)(7,18)(8,19)(9,20)(10,21)(11,22)(23,24)(25,26)(27,28)
sage: H3.homology_basis()
[[[2, 13]],
 [[3, 14]],
 [[4, 15]],
 [[5, 16]],
 [[6, 17]],
 [[7, 18]],
 [[8, 19]],
 [[9, 20]],
 [[10, 21]],
 [[11, 22]],
 [[23, 24]],
 [[25, 26]],
 [[27, 28]]]

make_generic()

Return a ribbon graph equivalent to self but where every vertex has valency 3.

OUTPUT:

• a ribbon graph that is equivalent to self but is generic in the sense that all vertices have valency 3

EXAMPLES:

sage: R = RibbonGraph(1,3); R
Ribbon graph of genus 1 and 3 boundary components
sage: R.sigma()
(1,2,3,9,7)(4,8,10,5,6)
sage: R.rho()
(1,4)(2,5)(3,6)(7,8)(9,10)
sage: G = R.make_generic(); G
Ribbon graph of genus 1 and 3 boundary components
sage: G.sigma()
(2,3,11)(5,6,13)(7,8,15)(9,16,17)(10,14,19)(12,18,21)(20,22)
sage: G.rho()
(2,5)(3,6)(7,8)(9,10)(11,12)(13,14)(15,16)(17,18)(19,20)(21,22)
sage: R.genus() == G.genus() and R.number_boundaries() == G.number_boundaries()
True

sage: R = RibbonGraph(5,4); R
Ribbon graph of genus 5 and 4 boundary components
sage: R.sigma()
(1,2,3,4,5,6,7,8,9,10,11,27,25,23)(12,24,26,28,13,14,15,16,17,18,19,20,21,22)
sage: R.rho()
(1,12)(2,13)(3,14)(4,15)(5,16)(6,17)(7,18)(8,19)(9,20)(10,21)(11,22)(23,24)(25,26)(27,28)
sage: G = R.reduced(); G
Ribbon graph of genus 5 and 4 boundary components
sage: G.sigma()
(2,3,4,5,6,7,8,9,10,11,27,25,23,24,26,28,13,14,15,16,17,18,19,20,21,22)
sage: G.rho()
(2,13)(3,14)(4,15)(5,16)(6,17)(7,18)(8,19)(9,20)(10,21)(11,22)(23,24)(25,26)(27,28)
sage: G.genus() == R.genus() and G.number_boundaries() == R.number_boundaries()
True

sage: R = RibbonGraph(0,6); R
Ribbon graph of genus 0 and 6 boundary components
sage: R.sigma()
(1,11,9,7,5,3)(2,4,6,8,10,12)
sage: R.rho()
(1,2)(3,4)(5,6)(7,8)(9,10)(11,12)
sage: G = R.reduced(); G
Ribbon graph of genus 0 and 6 boundary components
sage: G.sigma()
(3,4,6,8,10,12,11,9,7,5)
sage: G.rho()
(3,4)(5,6)(7,8)(9,10)(11,12)
sage: G.genus() == R.genus() and G.number_boundaries() == R.number_boundaries()
True

mu()

Return the rank of the first homology group of the thickening of the ribbon graph.

EXAMPLES:

sage: s1 = PermutationGroupElement('(1,3,5)(2,4,6)')
sage: r1 = PermutationGroupElement('(1,2)(3,4)(5,6)')
sage: R1 = RibbonGraph(s1,r1);R1
Ribbon graph of genus 1 and 1 boundary components
sage: R1.mu()
2

normalize()

Return an equivalent graph such that the enumeration of its darts exhausts all numbers from 1 to the number of darts.

OUTPUT:

• a ribbon graph equivalent to self such that the enumeration of its darts exhausts all numbers from 1 to the number of darts.

EXAMPLES:

sage: s0 = PermutationGroupElement('(1,22,3,4,5,6,7,15)(8,16,9,10,11,12,13,14)')
sage: r0 = PermutationGroupElement('(1,8)(22,9)(3,10)(4,11)(5,12)(6,13)(7,14)(15,16)')
sage: R0 = RibbonGraph(s0,r0); R0
Ribbon graph of genus 3 and 2 boundary components
sage: RN0 = R0.normalize(); RN0; RN0.sigma(); RN0.rho()
Ribbon graph of genus 3 and 2 boundary components
(1,16,2,3,4,5,6,14)(7,15,8,9,10,11,12,13)
(1,7)(2,9)(3,10)(4,11)(5,12)(6,13)(8,16)(14,15)

sage: s1 = PermutationGroupElement('(5,10,12)(30,34,78)')
sage: r1 = PermutationGroupElement('(5,30)(10,34)(12,78)')
sage: R1 = RibbonGraph(s1,r1); R1
Ribbon graph of genus 1 and 1 boundary components
sage: RN1 = R1.normalize(); RN1; RN1.sigma(); RN1.rho()
Ribbon graph of genus 1 and 1 boundary components
(1,2,3)(4,5,6)
(1,4)(2,5)(3,6)

number_boundaries()

Return number of boundary components of the thickening of the ribbon graph.

EXAMPLES:

The first example is the ribbon graph corresponding to the torus with one hole:

sage: s1 = PermutationGroupElement('(1,3,5)(2,4,6)')
sage: r1 = PermutationGroupElement('(1,2)(3,4)(5,6)')
sage: R1 = RibbonGraph(s1,r1)
sage: R1.number_boundaries()
1

This example is constructed by taking the bipartite graph of type \((3,3)\):

sage: s2 = PermutationGroupElement('(1,2,3)(4,5,6)(7,8,9)(10,11,12)(13,14,15)(16,17,18)')
sage: r2 = PermutationGroupElement('(1,16)(2,13)(3,10)(4,17)(5,14)(6,11)(7,18)(8,15)(9,12)')
sage: R2 = RibbonGraph(s2,r2)
sage: R2.number_boundaries()
3

reduced()

Return a ribbon graph with 1 vertex and \(\mu\) edges (where \(\mu\) is the first betti number of the graph).

OUTPUT:

• a ribbon graph whose \(\sigma\) permutation has only 1 non-singleton cycle and whose \(\rho\) permutation is a product of \(\mu\) disjoint 2-cycles

EXAMPLES:

sage: s1 = PermutationGroupElement('(1,3,5)(2,4,6)')
sage: r1 = PermutationGroupElement('(1,2)(3,4)(5,6)')
sage: R1 = RibbonGraph(s1,r1); R1
Ribbon graph of genus 1 and 1 boundary components
sage: G1 = R1.reduced(); G1
Ribbon graph of genus 1 and 1 boundary components
sage: G1.sigma()
(3,5,4,6)
sage: G1.rho()
(3,4)(5,6)

sage: s2 = PermutationGroupElement('(1,2,3)(4,5,6)(7,8,9)(10,11,12)(13,14,15)(16,17,18,19)')
sage: r2 = PermutationGroupElement('(1,16)(2,13)(3,10)(4,17)(5,14)(6,11)(7,18)(8,15)(9,12)(19,20)')
sage: R2 = RibbonGraph(s2,r2); R2
Ribbon graph of genus 1 and 3 boundary components
sage: G2 = R2.reduced(); G2
Ribbon graph of genus 1 and 3 boundary components
sage: G2.sigma()
(5,6,8,9,14,15,11,12)
sage: G2.rho()
(5,14)(6,11)(8,15)(9,12)

sage: s3 = PermutationGroupElement('(1,2,3)(4,5,6)(7,8,9)(10,11,12)(13,14,15,16)(17,18,19,20)(21,22,23,24)')
sage: r3 = PermutationGroupElement('(1,21)(2,17)(3,13)(4,22)(7,23)(5,18)(6,14)(8,19)(9,15)(10,24)(11,20)(12,16)')
sage: R3 = RibbonGraph(s3,r3); R3
Ribbon graph of genus 3 and 1 boundary components
sage: G3 = R3.reduced(); G3
Ribbon graph of genus 3 and 1 boundary components
sage: G3.sigma()
(5,6,8,9,11,12,18,19,20,14,15,16)
sage: G3.rho()
(5,18)(6,14)(8,19)(9,15)(11,20)(12,16)

rho()

Return the permutation \(\rho\) of self.

EXAMPLES:

sage: s1 = PermutationGroupElement('(1,3,5,8)(2,4,6)')
sage: r1 = PermutationGroupElement('(1,2)(3,4)(5,6)(8,15)')
sage: R = RibbonGraph(s1, r1)
sage: R.rho()
(1,2)(3,4)(5,6)(8,15)

sigma()

Return the permutation \(\sigma\) of self.

EXAMPLES:

sage: s1 = PermutationGroupElement('(1,3,5,8)(2,4,6)')
sage: r1 = PermutationGroupElement('(1,2)(3,4)(5,6)(8,15)')
sage: R = RibbonGraph(s1, r1)
sage: R.sigma()
(1,3,5,8)(2,4,6)

sage.geometry.ribbon_graph.bipartite_ribbon_graph(p, q)

Return the bipartite graph modeling the corresponding Brieskorn-Pham singularity.

Take two parallel lines in the plane, and consider \(p\) points in one of them and \(q\) points in the other. Join with a line each point from the first set with every point with the second set. The resulting is a planar projection of the complete bipartite graph of type \((p,q)\). If you consider the cyclic ordering at each vertex induced by the positive orientation of the plane, the result is a ribbon graph whose associated orientable surface with boundary is homeomorphic to the Milnor fiber of the Brieskorn-Pham singularity \(x^p + y^q\). It satisfies that it has \(\gcd(p,q)\) number of boundary components and genus \((pq - p - q - \gcd(p,q) - 2) / 2\).

INPUT:

• p – a positive integer
• q – a positive integer

EXAMPLES:

sage: B23 = RibbonGraph(2,3,bipartite=True); B23; B23.sigma(); B23.rho()
Ribbon graph of genus 1 and 1 boundary components
(1,2,3)(4,5,6)(7,8)(9,10)(11,12)
(1,8)(2,10)(3,12)(4,7)(5,9)(6,11)

sage: B32 = RibbonGraph(3,2,bipartite=True); B32; B32.sigma(); B32.rho()
Ribbon graph of genus 1 and 1 boundary components
(1,2)(3,4)(5,6)(7,8,9)(10,11,12)
(1,9)(2,12)(3,8)(4,11)(5,7)(6,10)

sage: B33 = RibbonGraph(3,3,bipartite=True); B33; B33.sigma(); B33.rho()
Ribbon graph of genus 1 and 3 boundary components
(1,2,3)(4,5,6)(7,8,9)(10,11,12)(13,14,15)(16,17,18)
(1,12)(2,15)(3,18)(4,11)(5,14)(6,17)(7,10)(8,13)(9,16)

sage: B24 = RibbonGraph(2,4,bipartite=True); B24; B24.sigma(); B24.rho()
Ribbon graph of genus 1 and 2 boundary components
(1,2,3,4)(5,6,7,8)(9,10)(11,12)(13,14)(15,16)
(1,10)(2,12)(3,14)(4,16)(5,9)(6,11)(7,13)(8,15)

sage: B47 = RibbonGraph(4,7, bipartite=True); B47; B47.sigma(); B47.rho()
Ribbon graph of genus 9 and 1 boundary components
(1,2,3,4,5,6,7)(8,9,10,11,12,13,14)(15,16,17,18,19,20,21)(22,23,24,25,26,27,28)(29,30,31,32)(33,34,35,36)(37,38,39,40)(41,42,43,44)(45,46,47,48)(49,50,51,52)(53,54,55,56)
(1,32)(2,36)(3,40)(4,44)(5,48)(6,52)(7,56)(8,31)(9,35)(10,39)(11,43)(12,47)(13,51)(14,55)(15,30)(16,34)(17,38)(18,42)(19,46)(20,50)(21,54)(22,29)(23,33)(24,37)(25,41)(26,45)(27,49)(28,53)

sage.geometry.ribbon_graph.make_ribbon(g, r)

Return a ribbon graph whose thickening has genus g and r boundary components.

INPUT:

• g – non-negative integer representing the genus of the thickening
• r – positive integer representing the number of boundary components of the thickening

OUTPUT:

• a ribbon graph that has 2 vertices (two non-trivial cycles in its sigma permutation) of valency \(2g + r\) and it has \(2g + r\) edges (and hence \(4g + 2r\) darts)

EXAMPLES:

sage: from sage.geometry.ribbon_graph import make_ribbon
sage: R = make_ribbon(0,1); R
Ribbon graph of genus 0 and 1 boundary components
sage: R.sigma()
()
sage: R.rho()
(1,2)

sage: R = make_ribbon(0,5); R
Ribbon graph of genus 0 and 5 boundary components
sage: R.sigma()
(1,9,7,5,3)(2,4,6,8,10)
sage: R.rho()
(1,2)(3,4)(5,6)(7,8)(9,10)

sage: R = make_ribbon(1,1); R
Ribbon graph of genus 1 and 1 boundary components
sage: R.sigma()
(1,2,3)(4,5,6)
sage: R.rho()
(1,4)(2,5)(3,6)

sage: R = make_ribbon(7,3); R
Ribbon graph of genus 7 and 3 boundary components
sage: R.sigma()
(1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,33,31)(16,32,34,17,18,19,20,21,22,23,24,25,26,27,28,29,30)
sage: R.rho()
(1,16)(2,17)(3,18)(4,19)(5,20)(6,21)(7,22)(8,23)(9,24)(10,25)(11,26)(12,27)(13,28)(14,29)(15,30)(31,32)(33,34)