Cleanup

[kieferk]

This is version 0.3.0. It's been a very long time!

• Pandas-ply is no longer required. The symbolic functionality has been improved, though there are probably still some bugs.
• Huge amounts of rewrites/improvements to the core code. Most of this is gonna be invisible to the user.
• Some decorators have changed and selection functions are overhauled so I doubt this is backwards compatible (but most other stuff should be pretty much the same...)
• Symbolic functions now work properly (like desc(X.col) for ordering and contains('c") in selection, to name a couple). This was pretty tough and honestly the reason I had to ditch the pandas-ply symbolic stuff.
• You can use negations on symbolics in selection functions like in dplyr. For example select(~X.cut) would select everything but the cut column.
• Python 3 is now all i'm going to support, so it may break in python 2.
• Original tests run, but obviously there may still be bugs here and there.

Cheers

[sharpe5]

Nice, nice work! I am very familiar with R, and I really missed how nice it was to work with dplyr. Your work on this is making the world a better place :)

[kieferk]

Thanks! I feel exactly the same way. The dplyr syntax is just so vastly superior. Let me know if you find any bugs or issues with the new version.