n() returns symbolic expression

[zimmerma]

from sage-support: ​http://groups.google.com/group/sage-support/browse_thread/thread/b36c90f1490eac19#

sage: a=(sqrt(4*(sqrt(3) - 5)*(sqrt(3) + 5) + 48) + 4*sqrt(3))/ (sqrt(3) + 5) 
sage: a.imag().n()
0.939469338708203*sin(0.500000000000000*pi)

[jason]

Even simpler:

sage: n(arctan2(0,-log(2)))
pi

[zimmerma]

Note also the strange tty output (look in the 2nd argument of arctan2):

sage: a=(sqrt(4*(sqrt(3) - 5)*(sqrt(3) + 5) + 48) + 4*sqrt(3))/ (sqrt(3) + 5)
sage: a.imag()
sin(1/2*arctan2(0, -88* + 48))*sqrt(abs(4*(sqrt(3) - 5)*(sqrt(3) + 5) + 48))/(sqrt(3) + 5)

Should I open a separate ticket for that? Paul

[burcin]

I'm changing the component to symbolics, since this is probably a bug in pynac.

Regarding the problem with the output Carl mentions in comment:2: This is also present in GiNaC, but the printing is better:

ginsh - GiNaC Interactive Shell (ginac V1.5.7)
  __,  _______  Copyright (C) 1999-2010 Johannes Gutenberg University Mainz,
 (__) *       | Germany.  This is free software with ABSOLUTELY NO WARRANTY.
  ._) i N a C | You are welcome to redistribute it under certain conditions.
<-------------' For details type `warranty;'.

Type ?? for a list of help topics.
> a=(sqrt(4*(sqrt(3) - 5)*(sqrt(3) + 5) + 48) + 4*sqrt(3))/ (sqrt(3) + 5);
(sqrt(48+4*(5+sqrt(3))*(-5+sqrt(3)))+4*sqrt(3))*(5+sqrt(3))^(-1)
> imag_part(a);
(5+sqrt(3))^(-1)*sqrt(abs(48+4*(5+sqrt(3))*(-5+sqrt(3))))*sin(1/2*atan2(0,48+4*(-22)))

Note the term 4*(-22) at the end of the last line.

We should open a new ticket for this and report it on the GiNaC list. I'm not sure if this has anything to do with this ticket ATM. Numeric evaluation seems to work fine in GiNaC:

> evalf(imag_part(a));
0.9394693387082032295

[zimmerma]

Burcin,

We should open a new ticket for this and report it on the GiNaC list.

I've reported a new ticket (#9947). I let you report it on the GiNaC list.

Paul

[burcin]

Replying to zimmerma:

Burcin,

We should open a new ticket for this and report it on the GiNaC list.

I've reported a new ticket (#9947). I let you report it on the GiNaC list.

This issue was fixed upstream by Richard Kreckel.

While the fix makes the original example on this ticket work, Jason's example from comment:1 or the one reported by Tian Wei on sage-support (below) still don't work.

sage: b = sqrt(-log(2))
sage: print b.imag().n()
0.832554611157698*sin(0.500000000000000*pi)

[burcin]

I uploaded a patch to fix this. The problem wasn't in pynac after all, it was the numeric approximation function for arctan2().

[zimmerma]

positive review, good work Burcin!

Paul