Some utility functions for Parquet.jl that parses parquet files generated by Pandas (mainly with the PyArrow engine), taking into account Pandas metadata.
The main improvement is that the current method should parse most numpy-encoded dates correctly.
It should be installable with:
import Pkg
Pkg.add(url="https://github.com/oskarr/PdParquet.jl", rev="dev")For an overview of the Pandas parquet metadata, see the docs.
I'm probably not going to be actively maintaining this beyond what is necessary for my use-case, and it is likely that the code will reside mostly on the dev branch for a long time.