Incidence matrix method doesnt work for G = DiGraph({0: ['a']})

[vipul79321]

Current behavior:

sage: G = DiGraph({0: ['a']})
sage: G.incidence_matrix()
Traceback (most recent call last)
...
TypeError: '<' not supported between instances of 'int' and 'str'

incidence_matrix method should be able to handle the case when vertices are of mismatched types.

Depends on #22349

CC: @dcoudert

Component: graph theory

Issue created by migration from https://trac.sagemath.org/ticket/29277

[dcoudert]

comment:1

When vertices are of different types, a solution is to give as input an ordering of the vertices

sage: G = DiGraph({0: ['a']})
sage: G.incidence_matrix(vertices=[0, 'a'])
[-1]
[ 1]
sage: G.incidence_matrix(vertices=list(G))
[ 1]
[-1]

and we will have to proceed with #22349 to stop sorting vertices and edges by default (not easy).

[vipul79321]

comment:2

Replying to @dcoudert:

When vertices are of different types, a solution is to give as input an ordering of the vertices

sage: G = DiGraph({0: ['a']})
sage: G.incidence_matrix(vertices=[0, 'a'])
[-1]
[ 1]
sage: G.incidence_matrix(vertices=list(G))
[ 1]
[-1]

and we will have to proceed with #22349 to stop sorting vertices and edges by default (not easy).

I was wondering why can't we make the following changes:
Instead of calling self.vertices(), we can call self.vertices(sort=False) which will solve our problem. And then we can modify doc-tests and documentation accordingly.

Also, We are kind of expecting the same behavior as above from fixing of #22349

[dcoudert]

comment:3

We did a hard work on the graph module to make it compatible with Python3. We let this case for compatibility with other codes for the moment. The compromise was to introduce parameter vertices.
Also, I prefer to first deprecate default sorting of vertices and edges before changing the behavior of this method. Meanwhile, we can document the issue in the doctests of the method to should that it is better to specify the order of vertices and edges.

[mkoeppe]

comment:4

Moving tickets to milestone sage-9.2 based on a review of last modification date, branch status, and severity.

[mkoeppe]

comment:6

Sage development has entered the release candidate phase for 9.3. Setting a new milestone for this ticket based on a cursory review.

[mkoeppe]

Dependencies: #22349