deephaven.ParquetTools.readTable() doesn't support Parquet date32 or date64 data types

[mikeblas]

Description

Using readTable() I've found that I can't read files that contain columns with the date32 or date64 data types. An exception is thrown. If we don't support it that's cool -- let's document it, and try to express a workaround about it if possible. But it would be nice to support the types ...

Steps to reproduce

from deephaven.ParquetTools import readTable

s = readTable("/data/equity-quotes-2021-03-04.parquet")

Expected results

Table is imported successfully

Actual results

An exception is thrown; here's how it starts out:

r-Scheduler-Serial-1 | i.d.g.s.SessionState      | Internal Error 'c5ef3418-ec87-4e7d-9b81-c1d7da6bfd02' java.lang.RuntimeException: Error in Python interpreter:
Type: <class 'RuntimeError'>
Value: io.deephaven.UncheckedDeephavenException: Unable to read column [Date]: DateLogicalType not supported
	at io.deephaven.db.v2.parquet.ParquetSchemaReader.lambda$readParquetSchema$2(ParquetSchemaReader.java:272)
	at java.util.Optional.orElseThrow(Optional.java:290)
	at io.deephaven.db.v2.parquet.ParquetSchemaReader.readParquetSchema(ParquetSchemaReader.java:266)
	at io.deephaven.db.tables.utils.ParquetTools.convertSchema(ParquetTools.java:650)
	at io.deephaven.db.tables.utils.ParquetTools.readTableInternal(ParquetTools.java:386)
	at io.deephaven.db.tables.utils.ParquetTools.readTable(ParquetTools.java:110)
	at org.jpy.PyLib.executeCode(Native Method)
:
:

Additional details and attachments

parquet_date.zip

[jcferretti]

Fixed by #1254