Not sure if this should be filed as a feature request instead.
Environment
Delta-rs version: 0.15.0
Binding: Python
Environment:
- Cloud provider: Databricks / AWS
Bug
What happened: I am unable to open a Delta table that has liquid clustering active.
What you expected to happen: Open the table successfully.
How to reproduce it:
spark.range(100).createOrReplaceTempView("test")
CREATE OR REPLACE TABLE nchammas.deltalakers_test
CLUSTER BY (id)
AS
SELECT *
FROM test;
import deltalake
deltalake.DeltaTable("/dbfs/user/hive/warehouse/nchammas.db/deltalakers_test")
This yields the following error:
DeltaProtocolError: Invalid JSON in log record, version=0, line=`{"domainMetadata":{"domain":"delta.liquid",
"configuration": "{\"clusteringColumns\":[{\"physicalName\":[\"id\"]}],\"domainName\":\"delta.liquid\"}",
"removed":false}}`, err=`unknown variant `domainMetadata`, expected one of `metaData`, `cdc`, `add`,
`remove`, `txn`, `protocol`, `commitInfo` at line 1 column 17`
There are other variations on this error, depending on whether the table has been optimized or cached. But they all mention domainMetadata, which was presumably added as part of liquid clustering.
Not sure if this should be filed as a feature request instead.
Environment
Delta-rs version: 0.15.0
Binding: Python
Environment:
Bug
What happened: I am unable to open a Delta table that has liquid clustering active.
What you expected to happen: Open the table successfully.
How to reproduce it:
This yields the following error:
There are other variations on this error, depending on whether the table has been optimized or cached. But they all mention
domainMetadata, which was presumably added as part of liquid clustering.