http://trac.sagemath.org/ticket/19030 fixes two small bugs in the creation of isogeny classes of elliptic curves.
1. When the codomain of a computed isogeny is isomorphic but not identical to a curve already computed it must be composed with a suitable isomorphism, but that was not happening. (Only revealed over number fields since over Q when you compute an isogeny from a kernel it always arranges for the codomain to be the unique reduced minimal model). 2. Half the maps stored were being overwritten with 0. Now the matrix of maps contains all prime degree isogenies in both directions. This is important when transferring points around the class since it means that whichever curve you start on you can get to all others, i.e. the sub-directed graph of edges where the maps are stored is completely transitive. I hope to get this merged before 6.9 closes its doors. I have used the patched code on several hundreds of thousands of curves in the LMFDB.... Thanks! John -- You received this message because you are subscribed to the Google Groups "sage-nt" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send an email to [email protected]. Visit this group at http://groups.google.com/group/sage-nt. For more options, visit https://groups.google.com/d/optout.
