@xhochy I'm still having issues with `pyarrow`. I have no issues writing the parquet files but when I try to read them back using Dask I get an error saying that the schema is different between the common metadata and the other partitions. When I individually read in each parquet partition, however, the resulting pyarrow table has the correct schema (I am specifying a schema when I write the files, but the string vs null column type is being ignored).
[ Full content available at: https://github.com/apache/arrow/issues/2110 ] This message was relayed via gitbox.apache.org for [email protected]
