suntectec commented on PR #4533:
URL: https://github.com/apache/flink-cdc/pull/4533#issuecomment-6491515580

   Re: the review above (@Ho1ySh1tC0d1ng)
   
   Thanks for the detailed review. All five points are addressed in the 
reworked branch (rebased on master, single commit `7c54546`):
   
   1. **DECFLOAT** — instead of advertising a mapping Debezium 1.9.8 cannot 
materialise, `Db2TypeUtils` now fails fast during schema inference for 
`DECFLOAT`, `XML`, `CLOB`, `DBCLOB` and `BLOB`. 
`Db2PipelineITCase#testUnsupportedColumnTypesFailFast` reads real 
`DECFLOAT(16)` / `DECFLOAT(34)` values from DB2 and asserts the rejection, so 
the schema/record mismatch cannot occur silently.
   2. **Runtime schema-change discovery** — not enabled. We documented the 
boundary instead: `GET_LIST_OF_NEW_CDC_ENABLED_TABLES` still contains `AND 
1=0`, `schema-change.enabled` only forwards events reported by the underlying 
source, the initial `CreateTableEvent` is still emitted, and 
checkpoint/savepoint restore does not refresh an outdated schema.
   3. **`jdbc.properties.*`** — now forwarded to Debezium as `database.*`, with 
an explicit `debezium.*` property taking precedence. 
`Db2DataSourceFactoryITCase#testJdbcPropertiesAffectConnection` reads `VALUES 
CURRENT SCHEMA` back from a real JDBC connection to prove the settings arrive.
   4. **Shaded driver** — `com.ibm.db2.jcc:*` is excluded from the shaded 
artifact; the built jar contains no `com/ibm/db2` classes and no IBM Maven 
metadata. The docs explain installing the driver into `FLINK_HOME/lib` of the 
submitting client and every JM/TM.
   5. **UPDATE prerequisites** — documented: capture tables require 
`CHG_UPD_TO_DEL_INS='Y'`, and direct `IBMSNAP_OPERATION='U'` records are not 
supported. The IT cases use the paired D/I setup you mentioned.
   
   Test status: 26 cases per generation (8 unit + 13 factory IT + 5 pipeline 
IT) pass on Java 11 / Flink 1.20.3 and Java 17 / Flink 2.2.0, plus Spotless and 
Checkstyle; the documentation builds for both locales.


-- 
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]

Reply via email to