-
Notifications
You must be signed in to change notification settings - Fork 8
Open
Description
Hi,
I'm interested in something like this.
My goal is: Define rules to check and correct data in dataframes.
So far I've made a start implemented something like this on https://github.com/lverweijen/pymodify
But I'm not very far yet.
Similarities:
- I have two kinds of rules:
CheckandCorrection. MyCorrectionseems somewhat similar to what this project calls aRule. This project seems to have no equivalent to what I call aCheck. - I also like to write rules both in python or in some kind of yaml-like configuration format. I'm not completely object to using
eval, although. My rules in python look a bit different from yours. I'm not opposed to add class-based rules to my own project in a similar way as this project does.
Differences:
- In the short term, I'm mostly interesting in applying rules on dataframes, so I've already made pandas a dependency. I'm willing to decouple my project from pandas in the future if I find a practical way to do so and if there is interest in doing so.
- It's useful it the rules written as code can be vectorized (better performance on big datasets), so I've already implemented this, although it could be improved. This is perhaps mostly useful if the target are dataframes.
- I'm not very interested in writing rules in "natural language", although I do want my rules to be somewhat readable, preferably by non-programmers as well.
If this project is completely dead, I'm willing to take over the name of this project.
If it gets revived, maybe I can use it as a dependency and avoid reimplementing the wheels myself, although it will probably require some work to make it practical for my own purposes.
Reactions are currently unavailable
Metadata
Metadata
Assignees
Labels
No labels