Suggestion: Summarise column value distribution

[jamiekt]

I just discovered datanomy and it looks cool, I just shared with the data engineering community in my company.

It would be nice to be have a visualisation of the distribution of values in a column, just like visidata frequency tables.

Great start, looking forward to seeing what comes next (and perhaps contributing).

[raulcd]

That's a great idea, thanks for the feature request.
I will have to think about how to tackle it though as initially the project scope wasn't to build a full parquet reader but a format visualization. This will require to read the full content and build the the value distribution on the page as histograms are not accessible via pyarrow and not all the writers support writing those. This has memory implications so I'll have to think about what would be the best way of implementing without increasing memory consumption that much. Nonetheless is a great idea and will think on how to implement it.