How does one define a "real" variable

[Mani chandra]

Hi,

How does one define a variable in sage and make sure the it's
conjugate is the same as the variable itself?

Thanks,
Mani chandra

[Mani chandra]

Hi,

Perhaps I wasn't specific.

sage: var('a')
sage: a.conjugate()
conjugate(a)

What I want is for SAGE to return a, instead of conjugate(a). Does "var"
take any arguments that forces a variable to be real?

This is implemented in the sympy package, which I tried to use inside
sage, but..

sage: from sympy import symbols
sage: a, b = symbols('a, b', real=True)
sage: a.conjugate()
a
sage: b.conjugate()
b
sage: x = symbols('x')
sage: x = a + I*b
sage: x.conjugate()
conjugate(a) - I*conjugate(b)

Shouldn't it return a - I*b, instead of conjugate(a) - I*conjugate(b) ?
It does however give the correct answer when I overwrite SAGE's
namespace with sympy's namespace:

sage: from sympy import *
sage: a, b = symbols('a, b', real=True)
sage: x = symbols('x')
sage: x = a + I*b
sage: a.conjugate()
a
sage: b.conjugate()
b
sage: x.conjugate()
a - I*b

[William Stein]

On Sat, Aug 22, 2009 at 3:13 AM, Mani chandra <mcha...@iitk.ac.in> wrote:

Mani chandra wrote:
> Hi,
>
>     How does one define a variable in sage and make sure the it's
> conjugate is the same as the variable itself?
>
> Thanks,
> Mani chandra
>
> >
>
>

Hi,

   Perhaps I wasn't specific.

sage: var('a')
sage: a.conjugate()
conjugate(a)

What I want is for SAGE to return a, instead of conjugate(a). Does "var"
take any arguments that forces a variable to be real?

Unfortunately, I don't think this is implemented yet.

 -- William 

--
William Stein
Associate Professor of Mathematics
University of Washington
http://wstein.org

[Mani chandra]

William Stein wrote:
>
>
>
> sage: var('a')
> sage: a.conjugate()
> conjugate(a)
>
> What I want is for SAGE to return a, instead of conjugate(a). Does
> "var"
> take any arguments that forces a variable to be real?
>
>
> Unfortunately, I don't think this is implemented yet.
>
> -- William
>
>

Hi,
Can this be some sort of a feature request for the next release?

Thanking you,
Mani chandra

[Mani chandra]

Minh Nguyen wrote:
> Hi Mani,

> See ticket #6802:
>
> http://trac.sagemath.org/sage_trac/ticket/6802
>
>
Oops, I should have checked before asking for a feature request. There
seems to be a ticket on this topic since about 5 weeks.

http://trac.sagemath.org/sage_trac/ticket/6559

Mani chandra

[Golam Mortuza Hossain]

Hi,

On Aug 22, 11:20 am, Minh Nguyen <nguyenmi...@gmail.com> wrote:

> On Sun, Aug 23, 2009 at 12:16 AM, Mani chandra<mchan...@iitk.ac.in> wrote:
> >>>       Can this be some sort of a feature request for the next release?
>
> >> See ticket #6802:
>
> >>http://trac.sagemath.org/sage_trac/ticket/6802
>
> > Oops, I should have checked before asking for a feature request. There
> > seems to be a ticket on this topic since about 5 weeks.
>
> >http://trac.sagemath.org/sage_trac/ticket/6559

Here are some related discussions and a pynac patch

http://groups.google.com/group/pynac-devel/browse_thread/thread/76868c8ad5ae5b70

Burcin suggested that to avoid potential issue of having
two different symbols of different domain, we need to move symbol
directory into sage.


Cheers,
Golam

[Robert Bradshaw]

On Aug 22, 2009, at 3:13 AM, Mani chandra wrote:

>
> Mani chandra wrote:
>> Hi,
>>
>> How does one define a variable in sage and make sure the it's
>> conjugate is the same as the variable itself?
>>
>> Thanks,
>> Mani chandra
>>
>>>
>>
>>
> Hi,
>
> Perhaps I wasn't specific.
>
> sage: var('a')
> sage: a.conjugate()
> conjugate(a)
>
> What I want is for SAGE to return a, instead of conjugate(a). Does
> "var"
> take any arguments that forces a variable to be real?

You have to invoke simplify for it to look at assumptions.

sage: var('a')
sage: assume(a, 'real')
sage: a.conjugate().simplify()
a

- Robert

[Mani chandra]

There seems to be something wrong with this implementation:
sage: var('a, b')
(a, b)
sage: assume(a, 'real')
sage: assume(b, 'real')
---------------------------------------------------------------------------
AttributeError Traceback (most recent call last)

/home/mc/<ipython console> in <module>()

/opt/sage/sage/local/lib/python2.6/site-packages/sage/symbolic/assumptions.pyc
in assume(*args)
194 else:
195 try:
--> 196 x.assume()
197 except KeyError:
198 raise TypeError, "assume not defined for objects
of type '%s'"%type(x)

/opt/sage/sage/local/lib/python2.6/site-packages/sage/symbolic/assumptions.pyc
in assume(self)
96 maxima.set("context", cur)
97
---> 98 if not self in _assumptions:
99 maxima.activate(self._context)
100 _assumptions.append(self)

/opt/sage/sage/local/lib/python2.6/site-packages/sage/symbolic/assumptions.pyc
in __cmp__(self, other)
70 if isinstance(self, GenericDeclaration) and
isinstance(other, GenericDeclaration):
71 return cmp( (self._var, self._assumption),
---> 72 (other._var, other._assumption) )
73 else:
74 return cmp(type(self), type(other))

/opt/sage/sage/local/lib/python2.6/site-packages/sage/symbolic/expression.so
in sage.symbolic.expression.Expression.__nonzero__
(sage/symbolic/expression.cpp:7506)()

AttributeError: 'GenericDeclaration' object has no attribute 'variables'


Looks like it can define only one variable in the namespace as real.
This is pretty strange.

--Mani chandra

[Golam Mortuza Hossain]

Hi,

On Sat, Aug 22, 2009 at 2:28 PM, Robert
Bradshaw<robe...@math.washington.edu> wrote:
> You have to invoke simplify for it to look at assumptions.
>
> sage: var('a')
> sage: assume(a, 'real')
> sage: a.conjugate().simplify()
> a

I guess, we should be careful here. When you invoke ".simplify()"
the expression is passed to maxima. For maxima, default domain
for variables is *real* where as for pynac it is *complex*.

So you will get above answer without "assume(a, 'real')".

I think, we should forward this issue to sage-devel as it seems
to be a dangerous mixing.

Cheers,
Golam

[Robert Bradshaw]

I thought we used to put maxima in complex mode, do we not do that
anymore? I guess more things got shook up in the symbolic switchover
than I thought.

- Robert

[William Stein]

Robert, we still do that.  The initialization code in calculus.py is:

maxima = Maxima(init_code = ['display2d:false; domain: complex; keepfloat: true; load(topoly_solver)'],
                script_subdirectory=None)

 

[Robert Bradshaw]

Hmm... guess its a bug (or missunderstanding) in domain: complex.

- Robert

[Mani chandra]

Minh Nguyen wrote:
> On Sun, Aug 23, 2009 at 12:16 AM, Mani chandra<mcha...@iitk.ac.in> wrote:
>
>> Minh Nguyen wrote:
>>
>
> Ticket #6802 has been closed as a duplicate of #6559.
>
>
Hi,

Just wanted to ask, since you're the release manager for the next
release, do you think ticket no. 6559 will make it?

Thanking you,
Mani chandra