Changes between Initial Version and Version 1 of Ticket #27695


Ignore:
Timestamp:
04/17/19 19:36:16 (4 months ago)
Author:
etn40ff
Comment:

Legend:

Unmodified
Added
Removed
Modified
  • Ticket #27695 – Description

    initial v1  
    2525
    2626Solution 2: enforce the fact that graph6 and dig6 representations make sense
    27 only for graphs with vertex set `range(n)` and raise an error otherwise. (The current code silently encodes an isomorphic graph.)
     27only for graphs with vertex set `range(n)` and raise an error otherwise. (The current code silently encodes an isomorphic graph.) Then use this directly to access vertices.
    2828
    2929To put things in perspective, these representations are barely used outside of `cluster_algebra_quiver` so neither solution would have a big impact on other code. On the contrary in `cluster_algebra_quiver` they are used extensively as a way of comparing isomorphism classes of quivers. This is not the right way of doing so (one should use `sage.graphs.generic_graph.GenericGraph.canonical_label` instead) but, personally, I do not thing that reimplementing  `cluster_algebra_quiver` is worth the effort.