Oh, one more thing. About actually representing your objects. Sympy roughly follows the following requirement `type(obj)(*obj.args)==obj`. It is rather important detail.
Check out this https://github.com/sympy/sympy/wiki/Args-Invariant and please learn why this invariant is important (tree traversal and rebuilding, etc). At the moment we have problems deciding how exactly we should deal with the corner cases, but hopefully this will be fixed before you start your projects. I can not stress how important of a detail this is. On 28 May 2013 12:23, Stefan Krastanov <krastanov.ste...@gmail.com> wrote: > Hi Sachin and Prasoon, > > I have also CCed Aaron, Gilbert and the mailing list as a whole. If you > have questions send them publicly to the mailing list, not privately to the > mentors (unless you consider the issue to really be a private matter) > > First of all, congratulations! > > Be sure to setup blogs for gsoc (or a gsoc category with a dedicated rss > feed if you want to reuse a personal blog). > > We also encourage you to help with reviews during your projects. For > instance, set aside a few hours or even a whole afternoon each week > dedicated to pull request reviews (checking the code for antipatterns, > pulling it and playing around with the new functionality, checking > coverage, etc). In your case this would be easier as anyway you will need > to review each others work, but checking other pull request from time to > time would be beneficial. > > During the application period we also asked you to comment on each-others > applications. It would be great if you can share those comments now. You > are probably going to get as much help from each other as you will get from > Gilbert, me and the community. > > I suggest that we set up a wiki page where you can work together and > propose the api for the vectors. A possible way to start would be for > Prasoon to copy the examples given by Sachin in his application and > complement/fix them as he sees fit and for Sachin to do the same with > Prasoon's examples (only the vector relate examples in both cases). > > Also, it would be great if you can provide a mock SymPy session for the > following problem (it is from electromagnetism because of Sachin's project, > but focus for now only on the vectors api). > > 1. (the space) In 3D flat space. > 2. (scalar fields) Create a scalar field `rho` which is to represent a > planar charge density with a Gaussian cross-section. For instance in > Cartesian or cylindrical coordinates rho=exp(-z^2). > 3. (vector fields) Create a vector field `B` which is constant in time, > homogeneous in space, pointing in the z direction. > 4. (vectors that are not vector fields) Set up a particle at some > arbitrary coordinates with some arbitrary initial velocity. > 5. (trajectories, parametrized lines) Imagine that you have a black box > which takes the world as input and outputs a trajectory for the particle. > Give the api for that black box. > > > And the following is more related to the Sachin's project, but after you > finish 1-5 I hope you can both comment on it: > > 6. (parametrized fields) how will 2 and 3 look if they depend on time > 7. (Maxwell) given rho, E and B derive the PDEs governing them. Assume > there are NO point particles. > 7a. (Maxwell) given 2 and 3 (or 6) derive the electric field (again assume > you have a black box for the algorithm and provide only an api) > > And this one is rather advanced, you can leave it out for now: > > 8. (no particles) Instead of defining 4 and 5, just use DiracDelta fields. > This is done in Jackson. > 9. What about other degenerate fields besides point particles in the form > of DiracDelta. What about line and surface charges. > > In any case, do not just to questions 6-9. Take your time and respond in > details to 1-5. And please do it in a well formated wikipage, not here on > the mailing list. > > Stefan > -- 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+unsubscr...@googlegroups.com. To post to this group, send email to sympy@googlegroups.com. Visit this group at http://groups.google.com/group/sympy?hl=en-US. For more options, visit https://groups.google.com/groups/opt_out.
