Consider support the PETL library

[blais]

Is your feature request related to a problem? Please describe.
The petl library is an alternative to Pandas that confers several benefits.
It's also much more regular and predictable to do transformations.
It does not have schema declarations.

Describe the solution you'd like
A common schema solution that works for both Pandas and PETL would be a great idea.

Describe alternatives you've considered
I found Pandera as I was just about to write the same thing myself.
I'm going to try using it instead.

[cosmicBboy]

hi @blais, thanks for pointing out PETL, I wasn't aware of it!

As you may or may not know, pandera is figuring out how to scalably and maintainably support other data manipulation libraries. PETL seems to be a good candidate for another pandera-support validation backend.

The main blocker for this is #381, which will abstract out all the pandas-specific stuff into a "pandas validation backend". The architecture of this is still in the works, but if you can make a comment on that issue with:

• a high-level description of the similarities and differences between pandas/PETL
• how pandera's schema and schema component classes map onto PETL data structures

that would be very helpful to make pandera more easily extensible.

Please let me know if you have any questions, and feel free to pop into the discord if you want to chat!