rdblue commented on a change in pull request #2898:
URL: https://github.com/apache/iceberg/pull/2898#discussion_r780843476
##########
File path:
flink/v1.14/flink/src/test/java/org/apache/iceberg/flink/sink/TestFlinkIcebergSinkV2.java
##########
@@ -84,36 +84,63 @@
private final FileFormat format;
private final int parallelism;
private final boolean partitioned;
+ private final String distributionMode;
private StreamExecutionEnvironment env;
private TestTableLoader tableLoader;
- @Parameterized.Parameters(name = "FileFormat = {0}, Parallelism = {1},
Partitioned={2}")
+ @Parameterized.Parameters(name = "FileFormat = {0}, Parallelism = {1},
Partitioned={2}, Distribution={3}")
public static Object[][] parameters() {
return new Object[][] {
- new Object[] {"avro", 1, true},
- new Object[] {"avro", 1, false},
- new Object[] {"avro", 2, true},
- new Object[] {"avro", 2, false},
- new Object[] {"orc", 1, true},
- new Object[] {"orc", 1, false},
- new Object[] {"orc", 2, true},
- new Object[] {"orc", 2, false},
- new Object[] {"parquet", 1, true},
- new Object[] {"parquet", 1, false},
- new Object[] {"parquet", 2, true},
- new Object[] {"parquet", 2, false}
+ new Object[] {"avro", 1, true,
TableProperties.WRITE_DISTRIBUTION_MODE_NONE},
+ new Object[] {"avro", 1, false,
TableProperties.WRITE_DISTRIBUTION_MODE_NONE},
+ new Object[] {"avro", 4, true,
TableProperties.WRITE_DISTRIBUTION_MODE_NONE},
+ new Object[] {"avro", 4, false,
TableProperties.WRITE_DISTRIBUTION_MODE_NONE},
+ new Object[] {"orc", 1, true,
TableProperties.WRITE_DISTRIBUTION_MODE_NONE},
+ new Object[] {"orc", 1, false,
TableProperties.WRITE_DISTRIBUTION_MODE_NONE},
+ new Object[] {"orc", 4, true,
TableProperties.WRITE_DISTRIBUTION_MODE_NONE},
+ new Object[] {"orc", 4, false,
TableProperties.WRITE_DISTRIBUTION_MODE_NONE},
+ new Object[] {"parquet", 1, true,
TableProperties.WRITE_DISTRIBUTION_MODE_NONE},
+ new Object[] {"parquet", 1, false,
TableProperties.WRITE_DISTRIBUTION_MODE_NONE},
+ new Object[] {"parquet", 4, true,
TableProperties.WRITE_DISTRIBUTION_MODE_NONE},
+ new Object[] {"parquet", 4, false,
TableProperties.WRITE_DISTRIBUTION_MODE_NONE},
+
+ new Object[] {"avro", 1, true,
TableProperties.WRITE_DISTRIBUTION_MODE_HASH},
+ new Object[] {"avro", 1, false,
TableProperties.WRITE_DISTRIBUTION_MODE_HASH},
+ new Object[] {"avro", 4, true,
TableProperties.WRITE_DISTRIBUTION_MODE_HASH},
+ new Object[] {"avro", 4, false,
TableProperties.WRITE_DISTRIBUTION_MODE_HASH},
+ new Object[] {"orc", 1, true,
TableProperties.WRITE_DISTRIBUTION_MODE_HASH},
+ new Object[] {"orc", 1, false,
TableProperties.WRITE_DISTRIBUTION_MODE_HASH},
+ new Object[] {"orc", 4, true,
TableProperties.WRITE_DISTRIBUTION_MODE_HASH},
+ new Object[] {"orc", 4, false,
TableProperties.WRITE_DISTRIBUTION_MODE_HASH},
+ new Object[] {"parquet", 1, true,
TableProperties.WRITE_DISTRIBUTION_MODE_HASH},
+ new Object[] {"parquet", 1, false,
TableProperties.WRITE_DISTRIBUTION_MODE_HASH},
+ new Object[] {"parquet", 4, true,
TableProperties.WRITE_DISTRIBUTION_MODE_HASH},
+ new Object[] {"parquet", 4, false,
TableProperties.WRITE_DISTRIBUTION_MODE_HASH},
+
+ new Object[] {"avro", 1, true,
TableProperties.WRITE_DISTRIBUTION_MODE_RANGE},
+ new Object[] {"avro", 1, false,
TableProperties.WRITE_DISTRIBUTION_MODE_RANGE},
+ new Object[] {"avro", 4, true,
TableProperties.WRITE_DISTRIBUTION_MODE_RANGE},
+ new Object[] {"avro", 4, false,
TableProperties.WRITE_DISTRIBUTION_MODE_RANGE},
+ new Object[] {"orc", 1, true,
TableProperties.WRITE_DISTRIBUTION_MODE_RANGE},
+ new Object[] {"orc", 1, false,
TableProperties.WRITE_DISTRIBUTION_MODE_RANGE},
+ new Object[] {"orc", 4, true,
TableProperties.WRITE_DISTRIBUTION_MODE_RANGE},
+ new Object[] {"orc", 4, false,
TableProperties.WRITE_DISTRIBUTION_MODE_RANGE},
+ new Object[] {"parquet", 1, true,
TableProperties.WRITE_DISTRIBUTION_MODE_RANGE},
+ new Object[] {"parquet", 1, false,
TableProperties.WRITE_DISTRIBUTION_MODE_RANGE},
+ new Object[] {"parquet", 4, true,
TableProperties.WRITE_DISTRIBUTION_MODE_RANGE},
+ new Object[] {"parquet", 4, false,
TableProperties.WRITE_DISTRIBUTION_MODE_RANGE}
Review comment:
The number of tests doesn't need to be increased by 3x. Instead, can you
keep the same test cases and add distribution modes that make sense? For
example, have Avro use NONE, Parquet use HASH, and ORC use RANGE.
--
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.
To unsubscribe, e-mail: issues-unsubscr...@iceberg.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@iceberg.apache.org
For additional commands, e-mail: issues-h...@iceberg.apache.org
