Write reference data in Arrow-specific JSON test dataset format

voltrondata-labs / parquet-lot

Generate Parquet files for Spark/Arrow cross-version integration testing

Apache License 2.0

0 stars 0 forks source link

Write reference data in Arrow-specific JSON test dataset format #1

Open ianmcook opened 3 years ago

ianmcook commented 3 years ago

Currently the reference JSON files are written in plain JSON with no type information. Write them instead in this JSON format used specifically for Arrow integration testing: https://arrow.apache.org/docs/format/Integration.html

ianmcook commented 3 years ago

See the discussion in this Zulip thread https://ursalabs.zulipchat.com/#narrow/stream/180245-dev