If I want to get the first row (or any row, but not a mix of values from multiple rows) of every group how can I implement that? Alternatively, what's the best way to implement distinct/drop_duplicates based only on a subset of columns?

Does

any_value

work for your usecase? https://www.getdaft.io/projects/docs/en/stable/api_docs/groupby.html#daft.dataframe.GroupedDataFrame.any_value

@Sammy Sidhu Empirically I tried that and it seems to grab the first value but I also noticed that in one of the docs it says that the function does not guarantee that it grabs from the same row across columns. Is this the right understanding?

@Kevin Wang maybe we should add an any_value wrapper that does a row-wise any_value across columns?

we could probably enforce that it grabs the same row. i think it already does but we just don’t promise that it will always do that

Yeah I was thinking we can maybe expose a

.groupby().any_row()

which would actually give that guarantee In the future I’d love to have also

.groupby().first(sort_group_by=…)

which seems like a pretty common ask