arina-ielchiieva commented on a change in pull request #1298: DRILL-5796:
Filter pruning for multi rowgroup parquet file
URL: https://github.com/apache/drill/pull/1298#discussion_r202052538
##########
File path:
exec/java-exec/src/test/java/org/apache/drill/exec/store/parquet/TestParquetFilterPushDown.java
##########
@@ -243,24 +248,98 @@ public void testTimeStampPredicateWithEval() throws
Exception {
.toFile();
ParquetMetadata footer = getParquetMetaData(file);
- testParquetRowGroupFilterEval(footer, "o_ordertimestamp = cast('1992-01-01
10:20:30' as timestamp)", false);
- testParquetRowGroupFilterEval(footer, "o_ordertimestamp = cast('1992-01-01
10:20:29' as timestamp)", true);
+ testParquetRowGroupFilterEval(footer, "o_ordertimestamp = cast('1992-01-01
10:20:30' as timestamp)", RowsMatch.SOME);
+ testParquetRowGroupFilterEval(footer, "o_ordertimestamp = cast('1992-01-01
10:20:29' as timestamp)", RowsMatch.NONE);
- testParquetRowGroupFilterEval(footer, "o_ordertimestamp >=
cast('1992-01-01 10:20:29' as timestamp)", false);
- testParquetRowGroupFilterEval(footer, "o_ordertimestamp >=
cast('1992-01-03 10:20:30' as timestamp)", false);
- testParquetRowGroupFilterEval(footer, "o_ordertimestamp >=
cast('1992-01-03 10:20:31' as timestamp)", true);
+ testParquetRowGroupFilterEval(footer, "o_ordertimestamp >=
cast('1992-01-01 10:20:29' as timestamp)", RowsMatch.ALL);
+ testParquetRowGroupFilterEval(footer, "o_ordertimestamp >=
cast('1992-01-03 10:20:30' as timestamp)", RowsMatch.SOME);
+ testParquetRowGroupFilterEval(footer, "o_ordertimestamp >=
cast('1992-01-03 10:20:31' as timestamp)", RowsMatch.NONE);
- testParquetRowGroupFilterEval(footer, "o_ordertimestamp > cast('1992-01-03
10:20:29' as timestamp)", false);
- testParquetRowGroupFilterEval(footer, "o_ordertimestamp > cast('1992-01-03
10:20:30' as timestamp)", true);
+ testParquetRowGroupFilterEval(footer, "o_ordertimestamp > cast('1992-01-03
10:20:29' as timestamp)", RowsMatch.SOME);
+ testParquetRowGroupFilterEval(footer, "o_ordertimestamp > cast('1992-01-03
10:20:30' as timestamp)", RowsMatch.NONE);
- testParquetRowGroupFilterEval(footer, "o_ordertimestamp <=
cast('1992-01-01 10:20:30' as timestamp)", false);
- testParquetRowGroupFilterEval(footer, "o_ordertimestamp <=
cast('1992-01-01 10:20:29' as timestamp)", true);
+ testParquetRowGroupFilterEval(footer, "o_ordertimestamp <=
cast('1992-01-01 10:20:30' as timestamp)", RowsMatch.SOME);
+ testParquetRowGroupFilterEval(footer, "o_ordertimestamp <=
cast('1992-01-01 10:20:29' as timestamp)", RowsMatch.NONE);
- testParquetRowGroupFilterEval(footer, "o_ordertimestamp < cast('1992-01-01
10:20:31' as timestamp)", false);
- testParquetRowGroupFilterEval(footer, "o_ordertimestamp < cast('1992-01-01
10:20:30' as timestamp)", true);
+ testParquetRowGroupFilterEval(footer, "o_ordertimestamp < cast('1992-01-01
10:20:31' as timestamp)", RowsMatch.SOME);
+ testParquetRowGroupFilterEval(footer, "o_ordertimestamp < cast('1992-01-01
10:20:30' as timestamp)", RowsMatch.NONE);
}
+ @Test
+ public void testFilterPruning() throws Exception {
+ // multirowgroup2 is a parquet file with 3 rowgroups inside. One with a=0,
another with a=1 and a=2, and the last with a=3 and a=4;
+ // FilterPushDown should be able to prune the filter from the scan
operator according to the rowgroup statistics.
+ final String sql = "select * from dfs.`parquet/multirowgroup2.parquet`
where ";
+ PlanTestBase.testPlanMatchingPatterns(sql + "a > 1", new
String[]{"numRowGroups=2"}, null); //No filter pruning
Review comment:
You may use the below method instead of passing the null:
```
/**
* The same as above, but without excludedPatterns
*/
public static void testPlanMatchingPatterns(String query, String[]
expectedPatterns) throws Exception {
testPlanMatchingPatterns(query, expectedPatterns, null);
}
```
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
With regards,
Apache Git Services
