I am using the simon_two_descent() method for EllipticCurve(), and
have some curves where the default pari precision is in sufficient,
for example


e=EllipticCurve([0,0,0,-10164,409444]);
e.simon_two_descent()

fails withe the (gp) run-time error "  *** bnfsunit: precision too low
in get_arch."

Now in a normal gp session I can fix this after either
\p56
or
default(realprecision, 56)

but when running in sage, doing
pari.set_real_precision(120)
seems to have no effect.

I guess this is because the latter call effects the precision of the
pari library functions, whereas simon_two_descent() is a gp script.

If this is the case, and there is no better method, I guess we could
add an extra input parameter "prec" to the simon_two_descent()
function and then in that function call default(realprecision,prec).
The default would be 28.

I should be capable of doing this myself!  But
(1) I'm not sure which file to edit: looks like
/home/jec/sage-2.7/devel/sage-main/build/sage/schemes/elliptic_curves/gp_simon.py
but there is at least one other clone of that file;
(2) What should I do then to rebuild sage and test it?
(3) When I am happy that it works, exactly what should I do to upload
the change to that it gets incorporated into sage?

If this is documented, someone please point me in the right direction.
 By the way, whoever it was who did the "SAGE search" thing deserves a
prize -- it is permanently in my firefox top right corner and is how I
look up anything about SAGE now!

John

PS I also have some fixes for the mwrank C++ code which I should
upload, so the answers to my questions above should ideally answer
that too.


-- 
John Cremona

--~--~---------~--~----~------------~-------~--~----~
To post to this group, send email to [email protected]
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://sage.scipy.org/sage/ and http://modular.math.washington.edu/sage/
-~----------~----~----~----~------~----~------~--~---

Reply via email to