Re: sinint/cosint specfun review

[Michele Ginesi]

On January 24, 2018, at 09:19, Colin Macdonald <address@hidden> wrote:

>Hi Michele and others,
>I'm trying to pick away at the GSoCCheers,pecfun work.  So far I've
>looked at sinint and cosint which look great.  I've made some minor
>changes, patches are at [1].
>But then unfortunately I found a serious problem in cosint for large
>negative real values.  The last patch at [1] is an xtest, but
>unfortunately I don't have a fix yet.
>@Michele are you still able to work on this?
>I'm not sure what the workflow should look like.  Should I fork and send
>merge requests to Michele?  Or continue with a series of patches on Savanah.
>[1] https://savannah.gnu.org/patch/index.php?9551
>cheers,
>Colin

Hi Colin.
I could still work on it, unfortunately not with all the time I had this summer. But I think I can try to figure out something. If the problem is only on big negative real part and small imaginary one it should be easy to make the modification once a good approximation is found. I will start searching for some good approximation in the next days.

Cheers,

Michele