On Thu, 2023-03-23 at 22:05 +0000, Dima Pasechnik wrote: > On Thu, Mar 23, 2023 at 9:14 PM Michael Orlitzky <mich...@orlitzky.com> wrote: > > > > Does someone have access to https://github.com/sagemath/FlintQS ? > > > > The build is broken with clang-16 and there's an open pull request to > > fix it. A new release would be helpful. > > I promoted you to a maintainer of this repo. >
Thanks. After spending some time with the code, I'd rather replace it. There are unreported security issues with its use of /tmp, and it's just generally pretty fragile. It also looks like a more reliable (but often slower) implementation has been added to flint in the meantime, as qsieve_factor(). On the examples I've tried, FlintQS is usually capable of factoring a product of two primes, and relatively quickly. Switching the implementation to qsieve_factor() slows it down a bit (2s to 3s, for example). However, in cases like qsieve(329884789374968270239373010571299817801707558079749117449) which is a product of THREE primes, FlintQS gives you the wrong answer in four seconds whereas qsieve_factor() takes six to get it right. Number theorists, please speak up if you prefer to get less reliable answers more quickly in some scenario that I haven't imagined. -- You received this message because you are subscribed to the Google Groups "sage-devel" group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-devel+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/sage-devel/7a04728a6fbe3b8fc739ada6e2589e2bb7acde1a.camel%40orlitzky.com.
