uros-b commented on code in PR #56683:
URL: https://github.com/apache/spark/pull/56683#discussion_r3461780806
##########
common/utils/src/main/resources/error/error-conditions.json:
##########
@@ -308,12 +308,6 @@
],
"sqlState" : "0A000"
},
- "AUTOCDC_TARGET_DOES_NOT_SUPPORT_MERGE" : {
Review Comment:
Main concern: error-message UX regression for the genuinely-incompatible
case. The old path gave a precise, actionable error: "the target table X does
not support row-level operations. AutoCDC requires a target backed by a
connector that supports MERGE." After this change, a user who points AutoCDC at
a truly MERGE-incapable target (e.g. plain parquet) now gets a generic
UNSUPPORTED_FEATURE.TABLE_OPERATION ("MERGE INTO TABLE"), and the table named
in the error is the internal auxiliary table (__autocdc_aux…), not the target
the user declared. That's a confusing diagnostic: the user never created that
table. Consider whether the execution-time failure should be caught and
re-wrapped to point back at the user-facing target, or whether a lighter
capability probe (without the DSv2-specific assertion) is still worth keeping.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
