Modulo Errors

I’ve made a few cosmetic and minor functional tweaks to some of my SAGE offerings. The latest version of eds suppresses some diagnostic information, and the representation of an eds object now gives the defining terms rather than the entire dictionary (which runs to scores of pages for sequences over the integers). Further, heights will now catch the zero division exception that sometimes arises when you use Shipsey’s algorithm on a sequence corresponding to a point of finite order. It will now give a warning to this effect, and return a height of zero as required.
This, coupled with the singular curve catch introduced last time should mean that batch testing of large numbers of curves shouldn’t now fail- but I’m sure I’ll find some new exception in due course!

I’ve taken stock of my various coding efforts listed on my university space and trimmed out the least useful / developed / efficient ones from the past year. What’s left generally offers features not otherwise available for the given CAS- generic group operations and their use for [hyper]elliptic curve groups in Maple; EDS/height calculations for SAGE. The bug I found in hyperelliptic curve divisor class addition over finite fields has since been fixed, making my alternative class redundant; and the NTL interface seems to have changed, rendering my faster version broken; so those have been dropped from the site.

Finally, I’ll be back in the West Country for the SAGE Days 6 talks this weekend.

This entry was posted on Tuesday, November 6th, 2007 at 3:37 pm and is filed under PhD, SAGE.