|Bill Allombert on Mon, 02 Apr 2012 10:59:51 +0200|
[Date Prev] [Date Next] [Thread Prev] [Thread Next] [Date Index] [Thread Index]
On Wed, Mar 28, 2012 at 09:52:36AM -0500, Ariel Pacetti wrote: > > Dear pari developers, > > sometime ago Bill wrote a routime called forqfvec which is a mix of > qfminim and somehow forvec (see > http://pari.math.u-bordeaux.fr/archives/pari-dev-1005/msg00003.html). > > The advantage of such routine is that instead of giving the whole > output of qfminim (which might blow up the memory) it lets you > manage each vector at a time and then discard it. The first > application of it is the routine qfrep writen by Karim. > > In such routine, one takes each vector of qfrep and compute its > norm, and then puts a 1 in the right place of a vector. This gives > the q-expansion of modular forms (of integral weight or half > integral weight) which is quite useful. I was asked by Bill to make > some lobby on this routine, so let me just mention some other > applications of it so as to consider including it in the future > releases. Hello Ariel, Karim has commited my patch in rev 29e443c6fb79037e2ccd65084c5180fbc4568e21. Here the documentation: forqfvec(v,q,b,expr): q being a square and symmetric matrix representing a positive definite quadratic form, evaluate expr for all vector v such that q(v) <= b. The formal variable v runs through all such vectors in turn. ? forqfvec(v, [3,2;2,3], 3, print(v)) [0, 1]~ [1, 0]~ [-1, 1]~ Cheers, Bill.