[GitHub] [iceberg] rdblue commented on a change in pull request #1920: Hive: Serialize metadata location so split generation does not need to load the table

[GitBox <gi...@apache.org>]

rdblue commented on a change in pull request #1920:
URL: https://github.com/apache/iceberg/pull/1920#discussion_r542915762



##########
File path: mr/src/main/java/org/apache/iceberg/mr/hive/HiveIcebergStorageHandler.java
##########
@@ -85,7 +86,11 @@ public void configureInputJobProperties(TableDesc tableDesc, Map<String, String>
     map.put(InputFormatConfig.TABLE_IDENTIFIER, props.getProperty(Catalogs.NAME));
     map.put(InputFormatConfig.TABLE_LOCATION, table.location());
     map.put(InputFormatConfig.TABLE_SCHEMA, schemaJson);
+    if (table instanceof BaseTable) {

Review comment:
       I like using `BaseTable` here. I'm not sure that we're going to keep `HasTableOperations` around because it doesn't provide much value over `BaseTable`. If a table uses `TableOperations`, then it is just a thin wrapper so there isn't much of a need to customize.
   
   And metadata tables technically have a `TableOperations` that could be accessible, but replacing the table with a static one based on the metadata location in the table ops would read the table and not the metadata table. So I think it is better to be more specific here.




----------------------------------------------------------------
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:
users@infra.apache.org



---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@iceberg.apache.org
For additional commands, e-mail: issues-help@iceberg.apache.org