Re: [sage-devel] maxima replacing floats with rationals

2010-02-08 Thread John Cremona
On 8 February 2010 13:29, Burcin Erocal wrote: > Hi Nils, > > I'll leave the maxima floats question to the experts. > > On Mon, 8 Feb 2010 00:48:47 -0800 (PST) > Nils Bruin wrote: > >> Incidentally, >> sage: S.operands()[0].operands()[0].operands()[3].pyobject() >> 5*I + 5 >> sage: type(S.operand

Re: [sage-devel] maxima replacing floats with rationals

2010-02-08 Thread Burcin Erocal
Hi Nils, I'll leave the maxima floats question to the experts. On Mon, 8 Feb 2010 00:48:47 -0800 (PST) Nils Bruin wrote: > Incidentally, > sage: S.operands()[0].operands()[0].operands()[3].pyobject() > 5*I + 5 > sage: type(S.operands()[0].operands()[0].operands()[3].pyobject()) > 'sage.rings.n

[sage-devel] maxima replacing floats with rationals

2010-02-08 Thread Nils Bruin
I noticed that "keepfloat: true" does not get always get honoured by maxima. A symptom: sage: integrate(cos(1.17*x),x) 0.854700854701*sin(1.17*x) (works as expected) sage: S=integrate(cos(1.17*x^2),x) sage: S -1/156*((5*I + 5)*sqrt(2)*sqrt(13)*erf((3/20*I - 3/20)*sqrt(2)*sqrt(13)*x) + (5*I - 5)*