Bogumił Kamiński

https://github.com/JuliaData/DataFrames.jl/pull/2881 > need page(s) about indexing and select/transform/... which would be more detailed than Getting Started, but less technical that the API reference.

This would improve consistency with other methods

allow hcat with any table

1

Currently in `hcat(x, df)` and `hcat(df, x)` we require `x` to be an `AbstractVector`. We could allow it to also be any Tables.jl table, so that the following would work:...

Improve flatten (slightly breaking)

8

Users find it confusing that strings are iterable, see https://discourse.julialang.org/t/flatten-in-case-column-contains-string-and-array/61284/5. They are not iterable in broadcasting. On the other hand we should not follow the full semantics of `Base.broadcastable` as...

Use threading in joins

1

A separate issue to keep track of this specific issue.

If both tables in join-operations are fitting into `Int32` range (probably a common case) then we can use `Int32` instead of `Int` for indexing. This should speed up things.

Improwe workflows with filtered DataFrame

21

This has been discussed in several places I create a separate issue for this to keep track of it as it is an important functionality I think. What we want...

Post PrettyTables.jl housekeeping

4

@ronisbr - I open this issue to keep track of to-do related to PrettyTables.jl: 1. coverage of show.jl is quite poor now: https://codecov.io/gh/JuliaData/DataFrames.jl/tree/master/src/abstractdataframe; it would be great to add more...

A separate issue to keep track only this feature

A separate issue to keep track only of this feature.