Let arrow schemas specify the ENUM type from parquet schemas

[the80srobot]

Is your feature request related to a problem or challenge? Please describe what you are trying to do.

I am exporting data from an application using an Arrow schema - the data is then written into a parquet file. Some fields are logically enums, and I would like to export them as the ENUM type in Parquet, but there is currently no way to specify that.

Describe the solution you'd like

In parquet, ENUM is an annotation of a BYTE_ARRAY, similar to STRING. Unlike String, it's parametrized with the allowable values.

It sounds like it might be an extension type, as in #5822 .

[tustvold]

Does pyarrow or similar support this, as that might provide inspiration for how to support this?

Otherwise #4702 is probably the closest thing we have to this, in terms of supporting parquet specific metadata.

[jhorstmann]

Unlike String, it's parametrized with the allowable values.

I don't think that is the case inside the parquet format itself, but it might be a useful part of the api to ensure the values get dictionary encoded

[the80srobot]

Sorry for the delay. I checked the C++ and Python versions and neither one of them provides a similar API.

I still think it might be useful to provide this, but as I learned more about the representation of enums in Parquet, I can see how it might be superfluous.