#14514: A constructor for the Brouwer-Haemers graph
---------------------------------+------------------------------------------
Reporter: ncohen | Owner: jason, ncohen, rlm
Type: enhancement | Status: needs_review
Priority: major | Milestone: sage-5.10
Component: graph theory | Resolution:
Keywords: | Work issues:
Report Upstream: N/A | Reviewers:
Authors: Nathann Cohen | Merged in:
Dependencies: | Stopgaps:
---------------------------------+------------------------------------------
Comment (by azi):
Replying to [comment:3 ncohen]:
> Yoooooooooo !!
>
> > The patch is fine! I would only add the following test.
>
> Done !
>
> > So that we have as much new tests as possible! Otherwise the whole
testing of the graph theory module finishes too quickly !!!
>
> Ahem. Yeah, good point `:-P`
>
> > BTW. I was wondering if its time to redesign this graph database
thing. If we keep adding "specific" graphs the codebase will explode with
code that basically just constructs new objects.
>
> Yep. I don't like it either.
>
> > One thing that I would suggest for all fixed graphs, compute their
sparse6/graph6 string (whichever is shorter) and simply have
Graph(thestring) in the given method?
>
> Once again : I don't like the way we do things now either. But replacing
the methods with sparse6 string means that our graphs are "proprietary"
graphs how they are built is also a nice information to have around. Plus
we lose layouts, the vertices' names (which may contain some information
too). And it would only shorten the smallgraphs file, not the constructors
that actually build families of graphs.
> And we lose the doctests and documentation too, perhaps ?
>
> > Or perhaps have a data file with graphs/sparse6 strings/layouts and
load that at runtime or something?
> >
> > What do you think ??
I agree. Somehow we want to balance usability and performance without
making a bloated module. I don't see a solution for that though :-)
>
> I think that we will need to have an index of all sparse6 string of the
graphs we have in Sage at some point. When we will want Sage to answer
questions like "Have you ever seen this graph ?". But if we do have such
an index, I think that we will still have the methods around at the same
time. They don't have the same purpose.
Yes. I was also thinking about that. To have some method of the form
"nameGraph()" returning the name of a given graph (if known) or perhaps a
construction of it in the sense "cartesian product of petersen and
5-cycle"
But this again sounds like a messy thing to implement :-)
>
> But still, I don't like it either `:-P`
>
> What do you think ?
>
> Patch updated, by the way !
Patch is fine I'm gonna change the status to reflect that.
>
> Nathann
--
Ticket URL: <http://trac.sagemath.org/sage_trac/ticket/14514#comment:4>
Sage <http://www.sagemath.org>
Sage: Creating a Viable Open Source Alternative to Magma, Maple, Mathematica,
and MATLAB
--
You received this message because you are subscribed to the Google Groups
"sage-trac" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To post to this group, send email to [email protected].
Visit this group at http://groups.google.com/group/sage-trac?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.
