Sumith : Optimized Awesomeness

GSoC Progress - Week 8

Hello. Short time since my last post. Here's my report since then.


I have continued my work on the Polynomial wrappers.

Constructors from hash_set and Basic have been developed and pushed up. Printing has also been pushed. I'm currently writing tests for both, they'll be ready soon.

When hash_set_eq() and hash_set_compare() were developed, we realised that there were many functions in *_eq() and *_compare() form with repeated logic, the idea was to templatize them which Shivam did in his PR #533.

Solution to worry of slow compilation was chalked which I wish to try in the coming week, using std::unique_ptr to a hash_set, instead of a straight hash_set. Hence not necessary to know the full definition of hash_set in the header. I've been reading relevant material, known as PIMPL idiom.


* #511 - Polynomial Wrapper

Targets for Week 9

I wish to develop the Polynomial wrappers further in the following order.

  • Constructors and basic methods, add, mul, etc, working with proper tests.
  • Solve the problem of slow compilation times.
  • As mentioned previously, use standard library alternates to Piranha constructs so that we can have something even when there is no Piranha as dependency.

After the institute began, the times have been rough. Hoping everything falls in place.

Oh by the way, SymPy will be present (and represented heavily) at PyCon India 2015. We sent in the content and final proposal for review last week. Have a look at the website for our proposal here.

That's all this week.