Update for the Beginning of the Summer

May 26, 2011

So the Google Summer of Code coding period officially started on Monday, and in solidarity with the students, I will be blogging once a week about various things. Some of the posts will just be about what I have done that week. Others will be continuations of my Risch Algorithm series of blog posts (see parts 0, 1, 2, and 3).

This week, I will do the former. I have spend the past several weeks preparing for the release. The main thing right now is to clear out the issues that are blocking the release. I merged in a branch that included all of my polys related fixes from my integration3 branch. Along with similar branch from earlier that had some non-polys related fixes (like some fixes to the integrals), all of my fixes from integration3 not directly related to my implementation of the Risch Algorithm should no be in master.

Once those issues are fixed, I should be ready to make a release candidate for the release. The last release was over a year ago (March 2010), and we’ve racked up quite a few changes since then. A few big ones are:

  • The new polys. This is (in my opinion) the biggest change. Because of the new polys, everything is faster, and simplification is far more powerful than it was before. This is for a few reasons. The biggest reason is that the new polys allow polynomials in any kind of expression, not just Symbols. This means that you can do things like factor the expression \cos^2{x} + 2\cos{x} + 1. As you can imagine, many simplifications of complex expressions are nothing more than polynomial simplifications, where the polynomial is in some function.

    In addition to this, the new polys have a much faster implementation, and if you have gmpy installed, it will use that and be even faster. There are also several faster algorithms, like a faster algorithm for multivariate factorization, that have been implemented. These all lead to blazing fast simplification and polynomial minipulation in SymPy.

  • The Quantum Module. Unfortunatly, I can’t say much about this, since I don’t know anything about quantum physics. Furthermore, at the time of the writing of this blog post, that part of the release notes hasn’t been written yet. Suffice it say that thanks to two GSoC projects from last summer (see this and this page), we now have a quantum physics module. A lot of the stuff in that module, from my understanding, is unique to SymPy, which is very exciting. (By the way, if you’re interested in this, Brian Granger can tell you more about it).
  • Various backwards incompatible changes. We’ve taken advantage of the fact that this will be a point release (0.7.0) to clean up some old cruft.
    • We’ve renamed the functions abs() and sum() to Abs() and summation(), respectively, because they conflicted with built-in names (although thanks to __abs__ magic, abs(expr) will still work with the built-in abs() function).
    • This will be the last release to support Python 2.4. This will be a big benefit to not have to support Python 2.4 anymore after this release. There were a ton of features added in Python 2.5 that we have had to either manually re-implement (like any() and all()), or have had to do without (like the with statement). Also, this will make porting to Python 3 much easier (this is one of our GSoC projects).
    • We split the class Basic, which is the base class of all SymPy types, into Basic and a subclass Expr. Mathematical objects like cos(x) or x*y*z**2 are instances of Expr. Objects that do not make sense in mathematical expressions, but still want to have some of the standard SymPy methods like .args and .subs() are Basic. For example, a Set object is Basic, but not Expr.
  • Lots of little bug fixes and new features. See the release notes.

Once we have the release out, I plan to go back to work on the Risch Algorithm. I am very close to finishing the exponential case, which means that once I do, any transcendental elementary function built up of only exponential extensions could be integrated or proven not to have an elementary integral by my algorithm. I also want to start getting the code ready to merge with the main code base, so that it can go in the next release (0.7.1).

Finally, I want to announce that I have been selected for a student sponsorship to the SciPy 2011 conference in Austin, TX in the week of July 11. Mateusz and I will be presenting a tutorial on SymPy. This will be the first time I have ever attended a conference, and I am very excited.


Follow

Get every new post delivered to your Inbox.

Join 124 other followers

%d bloggers like this: