Hello,
I was able to reproduce this in Sage 3.3
var('a b c')
first = a + b + c
first._operands[0]._operands[0] is a
True
second = loads(dumps(first))
second._operands[0]._operands[0] is a
False
Should anyone open a bug in trac? I am not sure of how to do that and
which function to address this bug, but it seems definitely something
worth it, don't you think?
Thank you
Maurizio
On 27 Feb, 10:35, Kenny <masso.ke...@gmail.com> wrote:
> Sorry... i forgot to display the results:
>
> + True
> + False
>
> That means thebugis in thedumpor the load function!
>
> On Feb 27, 10:33 am, Kenny <masso.ke...@gmail.com> wrote:
>
> > I've took a look a the example code, and so I think that thebugis
> > in the dumps and load function:
>
> > var('Rs')
> > mio = Rs + Rs*Rs + sqrt(10)
> > tuo=loads(dumps(mio))
> > print mio._operands[0]._operands[0] is Rs
> > print tuo._operands[0]._operands[0] is Rs
>
> > IAt this time I've not looked inside the code, but I think that it
> > will be simple for the developer of
> > this functions to find the solution! :)
>
> > If there is not an intrepid volunteer I will look at that on
> > Saturday...
>
> > On Feb 27, 9:51 am, Maurizio <maurizio.gran...@gmail.com> wrote:
>
> > > Carl,
> > > you were totally right.
>
> > > The results are matching your previsions. I get:
> > > var('Cb')
> > > G_igr_d._operands[1]._operands[1] is Cb
>
> > > False
>
> > > G_works = SR(repr(G_igr_d))
> > > time G_works.subs(paramsd)
>
> > > Time: CPU 0.16 s, Wall: 0.86 s
>
> > > I add now another aspect: the G_igr_d is symbolically evaluated in
> > > another worksheet, then I save it, and I load it back in the "numeric"
> > > worksheet, where I do all the numerical maths.
>
> > > If I go to the "symbolic" worksheet, I get:
>
> > > G_igr_d._operands[1]._operands[1] is Cb
> > > True
>
> > > What does this mean?
> > > Please note that in the symbolic worksheet, the Cb is declared only
> > > once (on top) as:
> > > var('Cb rCb Cf rCf Lf rLf Rs rQ')
>
> > > in the same way it is done in the numeric worksheet:
> > > var('rQ rCb rLf rCf Duty')
> > > var('Cb Lf Cf')
>
> > > What am I missing? I don't seem to catch any particular interaction
> > > with other symbolic system. The functions I use are:
> > > solve and subs, symplify and expand, matrix operations ( + * **(-1) ),
> > > matrix augmentation (augment), and then matrix extraction of elements
> > > (by column and row index)
>
> > > Any other info is needed? Hopefully this can fix an undiscoveredbug,
> > > or at least fix my mistake (which maybe can be helpful as well to the
> > > community!)
>
> > > Maurizio
>
> > > On 27 Feb, 02:58, Carl Witty <carl.wi...@gmail.com> wrote:
>
> > > > On Wed, Feb 25, 2009 at 2:33 PM, Maurizio <maurizio.gran...@gmail.com>
> > > > wrote:
> > > > > Is it a problem interfacing to any symbolic package different from the
> > > > > ordinary maxima one? I don't exactly recall if I took advantage of
> > > > > others.
>
> > > > OK, I spent some time looking at this, and this may indeed be the
> > > > problem. I have a workaround for your case, but I may need more
> > > > information to get somebody to actually fix the problem.
>
> > > > Can you try this test:
>
> > > > var('Cb')
> > > > G_igr_d._operands[1]._operands[1] is Cb
>
> > > > and report what the last line prints?
>
> > > > If it prints False, then you're seeing what I'm seeing: somehow you
> > > > have two different synbolic variable objects, both named Cb.
> > > > Ordinarily you would only ever have one such object. With two
> > > > separate objects, Python can't immediately answer the question whether
> > > > Cb_1==Cb_2, so it asks Maxima (which says yes, they're equal).
>
> > > > In that case, the workaround is to do:
> > > > G_works = SR(repr(G_igr_d))
> > > > This is an ugly hack that prints out G_igr_d, and then parses it back
> > > > in; doing so makes it use the standard variable objects.
>
> > > > If this is the problem, can you try to remember how you actually got
> > > > G_igr_d? It seems like the problem actually might be the integration
> > > > with some other symbolic system, and it would be very helpful to know
> > > > which one.
>
> > > > Carl
--~--~---------~--~----~------------~-------~--~----~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to
sage-devel-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://www.sagemath.org
-~----------~----~----~----~------~----~------~--~---
