Maarten Breddels

Proof of concept, requires this patch to plotly: ``` diff --git a/packages/python/plotly/_plotly_utils/basevalidators.py b/packages/python/plotly/_plotly_utils/basevalidators.py index 0ade83a95..0065125ec 100644 --- a/packages/python/plotly/_plotly_utils/basevalidators.py +++ b/packages/python/plotly/_plotly_utils/basevalidators.py @@ -75,6 +75,7 @@ def copy_to_readonly_numpy_array(v, kind=None, force_numeric=False): """ np...

With the changes in the way selections and filters are done, this needs a fresh look to see if we really want to do this at the server layer.

fix(core): file column should support nbytes property

1

This is the start for the support of strings for s3 files needed to fix #284. It requires more changes however. We assume the data is always locally available as...

When virtual columns are slow, you can persist them to file. I don't want to merge this yet, since I don't think it works nicely with the state/get/set system. Needs...

using VAEX_PROGRESS=rich, exporting would show too many tasks

✨ add unique aggregator

2

Split off from #1987

This makes is irrelevalt for a user if a column is dict encoded or not all dataframe operations on it will be the same. For backwards compatibility we still do...

Should help with #1819