lirui-apache commented on a change in pull request #10618: [FLINK-15287][hive]
Fix orc optimization for version less than 2.3 by introducing orc shim
URL: https://github.com/apache/flink/pull/10618#discussion_r360305976
##########
File path:
flink-connectors/flink-connector-hive/src/test/java/org/apache/flink/connectors/hive/TableEnvHiveConnectorTest.java
##########
@@ -124,9 +124,15 @@ public void testGetNonExistingFunction() throws Exception
{
public void testDifferentFormats() throws Exception {
String[] formats = new String[]{"orc", "parquet",
"sequencefile", "csv"};
for (String format : formats) {
- if (format.equals("orc") &&
HiveShimLoader.getHiveVersion().startsWith("2.0")) {
- // Ignore orc test for Hive version 2.0.x for
now due to FLINK-13998
- continue;
+ if (format.equals("orc")) {
+ String version =
HiveShimLoader.getHiveVersion();
+ // hive 2.0.x with orc can not run in local
mode, because Hadoop23Shims only
+ // supported for DFS, but can work in cluster.
+ // hive 2.2.x with orc can not run, because
hive 2.2.x exec lack orc classes,
Review comment:
Why could 2.2.x run ORC test before this PR? It doesn't feel right that
we're adding shim and our test covers fewer versions.
----------------------------------------------------------------
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]
With regards,
Apache Git Services
