rdblue commented on a change in pull request #1394:
URL: https://github.com/apache/iceberg/pull/1394#discussion_r478760991
##########
File path: mr/src/main/java/org/apache/iceberg/mr/Catalogs.java
##########
@@ -65,13 +68,17 @@ public static Table loadTable(Configuration conf,
Properties props) {
private static Table loadTable(Configuration conf, String tableIdentifier,
String tableLocation) {
Optional<Catalog> catalog = loadCatalog(conf);
+ Table table;
if (catalog.isPresent()) {
Preconditions.checkArgument(tableIdentifier != null, "Table identifier
not set");
- return catalog.get().loadTable(TableIdentifier.parse(tableIdentifier));
+ table = catalog.get().loadTable(TableIdentifier.parse(tableIdentifier));
+ } else {
+ Preconditions.checkArgument(tableLocation != null, "Table location not
set");
+ table = new HadoopTables(conf).load(tableLocation);
}
- Preconditions.checkArgument(tableLocation != null, "Table location not
set");
- return new HadoopTables(conf).load(tableLocation);
+ LOG.info("Table loaded by catalog: [{}]", table);
Review comment:
I think this should be done in the catalog or `HadoopTables`, not in
`Catalogs`. Otherwise, we don't get the log message consistently when table
loads go directly through the catalog (as in Spark 3). And if we were to later
add a message to the catalog, then this would be a duplicate.
----------------------------------------------------------------
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]
