pvary commented on a change in pull request #1612:
URL: https://github.com/apache/iceberg/pull/1612#discussion_r526683067
##########
File path: mr/src/main/java/org/apache/iceberg/mr/hive/HiveIcebergMetaHook.java
##########
@@ -80,6 +81,10 @@ public void
preCreateTable(org.apache.hadoop.hive.metastore.api.Table hmsTable)
Preconditions.checkArgument(catalogProperties.getProperty(InputFormatConfig.PARTITION_SPEC)
== null,
"Iceberg table already created - can not use provided partition
specification");
+ Schema hmsSchema = HiveSchemaUtil.schema(hmsTable.getSd().getCols());
+ Preconditions.checkArgument(HiveSchemaUtil.compatible(hmsSchema,
icebergTable.schema()),
+ "Iceberg table already created - with different specification");
Review comment:
After discussing offline with @marton-bod we agreed that we should have
stricter mapping. See:
https://github.com/apache/iceberg/pull/1612#discussion_r526065544
There is still one remaining important use-case:
- The table is created before trying to access from Hive and the Catalog is
HadoopTable/HadoopCatalog/CustomCatalog. When the user wants to create the Hive
table it wants to use only a few columns from the table. Either because it has
too many columns and wants to have a simplified table in Hive, or the Iceberg
table contains columns which are not supported by Hive, like `time` or `fixed`.
----------------------------------------------------------------
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]
