Feedback for GSoC Proposal

[Sagar Bharadwaj]

I am interested in Implementing multiple coordinate systems for Vectors in SymPy.

I have completed the timeline and here is my proposal. Please Review it and give me you valuable suggestions.

GitHub Wiki Page (Better formatting)

Google Docs Link

[Alan Bromborsky]

You proposal is very good.  My only suggestion is that you may want a class for vector differential operators (gradient operator).  If the coordinate system constructor returns the gradient operator you could (in future) overload * and ^ operators so that if f is a scalar field and V a vector field and grad the vector operator then -

gradient of scalar field = grad * f
divergence of vector field = grad * V
curl of vector field = grad ^ V

then the Laplacian would be grad * (grad * f)

--
You received this message because you are subscribed to the Google Groups "sympy" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sympy+un...@googlegroups.com.
To post to this group, send email to sy...@googlegroups.com.
Visit this group at https://groups.google.com/group/sympy.
To view this discussion on the web visit https://groups.google.com/d/msgid/sympy/b940f17a-d895-4c50-bfd4-70ddd1489c93%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Sagar Bharadwaj]

Yes. There is a class called Del in sympy.vector.deloperator.py. But in current implementation, it does not return the del operator ( the grad operator). It only calculates gradient, curl and divergence.

Thank you for the suggestion.
Even I was thinking returning a del operator from del class would be better. I'll incorporate that in my proposal.

Thank you for reviewing my proposal.

On 1 Apr 2017 5:35 p.m., "Alan Bromborsky" <abrom...@gmail.com> wrote:

You proposal is very good.  My only suggestion is that you may want a class for vector differential operators (gradient operator).  If the coordinate system constructor returns the gradient operator you could (in future) overload * and ^ operators so that if f is a scalar field and V a vector field and grad the vector operator then -

gradient of scalar field = grad * f
divergence of vector field = grad * V
curl of vector field = grad ^ V

then the Laplacian would be grad * (grad * f)



On 04/01/2017 05:40 AM, Sagar Bharadwaj wrote:

I am interested in Implementing multiple coordinate systems for Vectors in SymPy.

I have completed the timeline and here is my proposal. Please Review it and give me you valuable suggestions.

GitHub Wiki Page (Better formatting)

Google Docs Link

--
You received this message because you are subscribed to the Google Groups "sympy" group.

To unsubscribe from this group and stop receiving emails from it, send an email to sympy+unsubscribe@googlegroups.com.

To post to this group, send email to sy...@googlegroups.com.
Visit this group at https://groups.google.com/group/sympy.
To view this discussion on the web visit https://groups.google.com/d/msgid/sympy/b940f17a-d895-4c50-bfd4-70ddd1489c93%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google Groups "sympy" group.

To unsubscribe from this group and stop receiving emails from it, send an email to sympy+unsubscribe@googlegroups.com.

To post to this group, send email to sy...@googlegroups.com.
Visit this group at https://groups.google.com/group/sympy.

To view this discussion on the web visit https://groups.google.com/d/msgid/sympy/f1dd0d91-e5f2-b094-9e4e-a3eb8df574e8%40gmail.com.

[Sagar Bharadwaj]

Any other suggestions? Anything more that needs to be done on the project other than what I mentioned?

[Sagar Bharadwaj]

I have made the required changes in both the wiki and the doc.

Please look into the 'Differential Geometry Functions' subsection in 'Implementation' section.

Also the 'June 15 - June 22' period in the Timeline.

[szymon.m...@gmail.com]

Hi Sagar,
I found mistake in your proposal. Lame coefficient are defined as you wrote, but you should also take a root from your formula.
See at wikipedia: https://en.wikipedia.org/wiki/Curvilinear_coordinates#Lam.C3.A9_coefficients
They are defined as a norm of vector Diff(r(q1,q2,q3), qi) 

Szymon

[Sagar Bharadwaj]

Hi Szymon,

Thank you for reviewing my proposal. 

In the Lame coefficients formula in my proposal, I have raised the whole expression to the power half, which is equivalent to taking it's root. I think you might have missed that because it's in small font.

Was that the mistake you were referring or have I committed a mistake elsewhere?

Sagar

[szymon.m...@gmail.com]

OK, I get it.

or have I committed a mistake elsewhere?

No, I just didn't see this power.

Szymon