Ok If anyone wants to test the program I have it accessible in my home
directory (novocin) on the development machine
sage.math.washington.edu:
Here's the syntax:
the program is called newfact and stdin should be an NTL style
polynomial and it will stdout the factorization.
I have a file called inputpoly also in my home directory so you can
call
/home/novocin/newfact </home/novocin/inputpoly >resultpoly
There is also a file called ntlfact in my home directory which does
the same thing but with the old NTL library.
I have some test polynomials which are designed to be slow for the van
Hoeij approach in the folder called testpolys and in the same folder a
little sage script for loading them, it's called loadpolys.sage and if
you load it then you can create a polynomial in sage with the command
poly(N).
This polynomial will give most factoring programs a headache.
So to create a test input of size 301
I'll run in SAGE from my testpolys folder
load 'loadpolys.sage'
f=poly(301)
g=open('../inputpoly','w')
g.write(str(ntl.ZZX(f.coeffs())))
g.close()
So now if you want to run tests on my little code it's on the
development machine.
Thanks,
-Andy
On Oct 3, 8:56 pm, "William Stein" <[EMAIL PROTECTED]> wrote:
> On Fri, Oct 3, 2008 at 11:54 AM, AndyNovo <[EMAIL PROTECTED]> wrote:
>
> > Haven't heard back from Victor Shoup, I'm not sure if I will...
>
> Don't hold your breath. I think he's not working on NTL
> much anymore, if at all.
>
> William
>
>
>
>
>
> > On Oct 1, 2:18 am, Bill Hart <[EMAIL PROTECTED]> wrote:
> >> Yep the theory is sound. The only concern would be to check that the
> >> matrices weren't getting screwed up between NTL and fpLLL format, but
> >> it is kind of obvious that they aren't, else the patch wouldn't work
> >> at all.
>
> >> But if it is merged upstream in NTL, SAGE will have to use a different
> >> cutoff between Pari and NTL to take advantage of the patch.
>
> >> Note my student Richard Howell-Peak and I worked on factorisation in Z/
> >> pZ[x]. This is now done. Andy and I will try and get something out
> >> over the next year (or sooner if possible) for Z[x]. It's going to
> >> totally rock, as Andy's new ideas have not been fully implemented
> >> before and should be faster than anything out there.
>
> >> Bill.
>
> >> On 30 Sep, 20:37, Martin Albrecht <[EMAIL PROTECTED]>
> >> wrote:
>
> >> > On Tuesday 30 September 2008, David Harvey wrote:
>
> >> > > Yeah why the hell not.
>
> >> > > As long as someone who understands these algorithms (me: not yet)
> >> > > signs off on precision bounds etc.
>
> >> > > Not sure if you'll get it upstream, but you could try.
>
> >> > fpLLL has a wrapper class which does all that for you, i.e. it
> >> > guarantees the
> >> > result to be LLL reduced.
>
> >> > Cheers,
> >> > Martin
>
> >> > --
> >> > name: Martin Albrecht
> >> > _pgp:http://pgp.mit.edu:11371/pks/lookup?op=get&search=0x8EF0DC99
> >> > _www:http://www.informatik.uni-bremen.de/~malb
> >> > _jab: [EMAIL PROTECTED]
>
> --
> William Stein
> Associate Professor of Mathematics
> University of Washingtonhttp://wstein.org
--~--~---------~--~----~------------~-------~--~----~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://www.sagemath.org
-~----------~----~----~----~------~----~------~--~---
