[NumericPrelude] Polynomial.divMod bug with respect to denormalized divisor (Was: Bug?)

Dominic Steinitz dominic at steinitz.org
Mon Feb 24 14:18:59 GMT 2014


Thanks for fixing it so quickly. I was just playing: I wanted to write a general Euclid’s algorithm that works on integers and also on polynomials. It should work for any integral domain (I think - my knowledge is very rusty). But now I have run out of time and have to work on a presentation.

Dominic Steinitz
dominic at steinitz.org
http://idontgetoutmuch.wordpress.com

On 24 Feb 2014, at 12:58, Henning Thielemann <schlepptop at henning-thielemann.de> wrote:

> Am 24.02.2014 11:09, schrieb Dominic Steinitz:
>> If I divide x + 1 by 0.x + 1 I think I should get x + 1
>> 
>>> *EuclidPoly> fromCoeffs [1,1] `div` fromCoeffs [1,0]
>>> Polynomial.fromCoeffs [1.0]
> 
> Although I wrote tests for Polynomials that one slipped through.
> 
>> I couldn’t see how to raise a new ticket on trac.
> 
> I have just fixed it and pushed to darcs repo. Please tell me when you need a new Hackage release.
> 




More information about the Numeric-Prelude mailing list