Max Gekk created SPARK-57460:
--------------------------------
Summary: Support nanosecond-precision timestamp types in the JDBC
datasource
Key: SPARK-57460
URL: https://issues.apache.org/jira/browse/SPARK-57460
Project: Spark
Issue Type: Sub-task
Components: SQL
Affects Versions: 5.0.0
Reporter: Max Gekk
Umbrella: SPARK-56822 (Timestamps with nanosecond precision).
Add read and write support for the nanosecond-capable timestamp types
TIMESTAMP_NTZ(p) and TIMESTAMP_LTZ(p) (p in 7-9) so this datasource reaches
parity with the microsecond TimestampType / TimestampNTZType. Remove the
SPARK-57166 rejection guardrail (supportDataType / supportsDataType) once read
and write are implemented and tested, and update FileBasedDataSourceSuite
accordingly. Cover precisions 7-9 for both NTZ and LTZ.
Scope (java.sql.Timestamp already carries sub-microsecond getNanos):
- Spark->JDBC: add nanos cases to JdbcUtils.getCommonJDBCType (Types.TIMESTAMP;
dialects may emit TIMESTAMP(p) / WITH TIME ZONE).
- JDBC->Spark: getCatalystType / getTimestampType (map driver TIMESTAMP to
nanos only when requested via option/dialect, to preserve default microsecond
behavior).
- Read: makeGetter builds TimestampNanosVal preserving
rs.getTimestamp().getNanos().
- Write: makeSetter sets full nanos.
- Dialect hooks in JdbcDialects for DBs with explicit TIMESTAMP(p).
- Guardrail: CreatableRelationProvider.supportsDataType whitelist in
interfaces.scala.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
