OK, so I now I see that while I did not think that the interface should have changed, it has changed in a very small way which is enough to break thinks. So I did mislead you.
I separated out the old class Point which contained both a point (x:y:z) in P^2(Q) and a pointer to an elliptic curve containing the point, to a base class P2Point for bare points (hence new files called p2points.h/cc with test programs in procs/), and the class Point for a point on a specific curve is derived from that. The reason for this is that to handle non-integral models without rewriting all the elliptic curve code I just work out a scaling to make the model integral and work on that scaled model internally, applying the scaling on input and output of points. As a result, the points returned by the two_descent class are P2Points and not Points. That probably does not make any sense, but it should be very easy to replace Point by P2Point in the wrapping code and have everything working again. John On 01/01/2008, mabshoff <[EMAIL PROTECTED]> wrote: > > > > On Jan 1, 1:05 pm, "John Cremona" <[EMAIL PROTECTED]> wrote: > > Michael, > > > > I think we have some crossed wires again, presumably my fault as usual. > > Nope, it was my fault in this case for sure. I just overflew your > message and misunderstood "The Sage interface to the library functions > for mwrank (etc) will need to be adapted to handle this." to mean > "adapted to use the extra functionality" implying "every that used to > work will keep working". I guess you need to be more direct as in > "Caution: this *breaks* the interface". Then there wouldn't be a doubt > in my mind that something needs to be fixed first ;) > > > The spkg I posted yesterday, and also the slightly patched one you > > then made from it, build perfectly if you untar the spkg and cd into > > src then do make and make check. > > I send you the details off-list, but I can reproduce the failure I > observed with your unchanged eclib-20071231.spkg. > > > But despite having the spkg suffix > > on this I made no claims at all that this version would work with Sage > > 2.9.1! > > :) > > > In fact I said specifically that the Sage interface would need > > to change to take into account the changes I made in the > > mwarank->cremona->eclib migration. > > > > Hence no-one should be trying to use this withing Sage until the > > wrapping has been changed to match the (fairly minor) changes in > > interface between the last cremona* and the first eclib*. These > > changes were only made to provide the extra functionality asked for by > > Sage users.... > > Well, by compiling the old cremona.spkg everything is back to normal > and works again. > > > Happy New Year to all, > > +1 > > > John > > Cheers, > > Michael > > > -- John Cremona --~--~---------~--~----~------------~-------~--~----~ To post to this group, send email to sage-devel@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/sage-devel URLs: http://sage.scipy.org/sage/ and http://modular.math.washington.edu/sage/ -~----------~----~----~----~------~----~------~--~---
