openinx commented on a change in pull request #2389:
URL: https://github.com/apache/iceberg/pull/2389#discussion_r603231454
##########
File path: site/docs/flink.md
##########
@@ -312,17 +312,47 @@ INSERT OVERWRITE hive_catalog.default.sample
PARTITION(data='a') SELECT 6;
For a partitioned iceberg table, when all the partition columns are set a
value in `PARTITION` clause, it is inserting into a static partition, otherwise
if partial partition columns (prefix part of all partition columns) are set a
value in `PARTITION` clause, it is writing the query result into a dynamic
partition.
For an unpartitioned iceberg table, its data will be completely overwritten by
`INSERT OVERWRITE`.
-## Reading with DataStream
+## Iceberg Operation with DataStream API
+### Load Iceberg Catalog
Review comment:
Here I think we should introduce the background why we introduce the
`CatalogLoader` & `TableLoader`: We flink operator want to access the iceberg
table while Catalog & Table are not serializable because they depends on some
resources that could not be serializable (such as Connection). So we have to
introduce the `Loader` to maintain the configurations which are required to
initialize the `Catalog` and `Table`.
Also we'd better to list the general `CatalogLoader`s and explain what do
they mean:
a. `HiveCatalogLoader` ;
b. `HadoopCatalogLoader` ;
c. `CustomCatalogLoader`.
Ditto for `TableLoader`.
--
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.
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
