Hugo van der Sanden

Keeping the two separate makes sense to me; it's always possible that particular patterns of use might find a more noticeable difference, so this would leave users the opportunity to...

Thanks, I hadn't appreciated there was so much difference between github and release (and sqrtmod doesn't even appear in the already lengthy list of changes). I'll see if I can...

Just taking a quick look through latest sqrtmod PP, I noticed this: ``` diff --git a/lib/Math/Prime/Util/PP.pm b/lib/Math/Prime/Util/PP.pm index 69b8cab..625e377 100644 --- a/lib/Math/Prime/Util/PP.pm +++ b/lib/Math/Prime/Util/PP.pm @@ -3879,7 +3879,7 @@ sub sqrtmod...

Well I originally looked because my own code has a comment `# TODO: use Tonelli-Shanks`, and I thought before I start exploring that I should check if MPU already has...

Am I right to guess that you are pushing towards an imminent release? I've pushed [commit 183d150be0](https://github.com/hvds/divrep/commit/183d150be0beeec7bffe77c13d6e34599e2bcd33) to my code for now, but the people using it are not sophisticated...

> I have to confess that in 12 years of working on the Perl core distribution, I had never heard of the [Andrew File System](https://en.wikipedia.org/wiki/Andrew_File_System) before encountering this ticket. So...

> * for t/re/opt.t Sorry I didn't notice this back in August: it is quite strange to me that `re/opt.t` is failing, but if I read the output correctly the...

If we show the flags on `svl` and `svr` at the start of `pp_add` for the first iteration we see: ``` svl->sv_flags = 0x8012202 = SVt_NV | SVf_NOK | SVp_NOK...

> @t-a-k @hvds Any further thoughts? Nothing new from me; I'll stick with my earlier comment "I agree that lossless_NV_to_IV should at the very least be consistent with SvIV_please_nomg, so...

> > I think it involves regex flags (`RXf_WHITE` and `RXf_SKIPWHITE`, > > Yeah, it does. I forgot they were regex flags rather than split's `op_private` flags. D'oh. Not sure...