jprdonnelly/impala
1
0
Fork 0
mirror of https://github.com/apache/impala.git synced 2025-12-25 02:03:09 -05:00
Code Issues Projects Releases Wiki Activity

IMPALA-14290: Make Iceberg partitioning column names case insensitive

Browse Source 
When creating or altering partitions of Iceberg tables, Impala only
accepts column names if they are in lowercase and throws
ImpalaRuntimeException otherwise.

This patch allows the usage of other cases as well in PARTITION SPEC
clauses. IcebergPartitionField converts field names to lower case in its
constructor, similar to ColumnDef and PartitionKeyValue.

Testing:
 * ran existing tests
 * add new test with mixed letter case columns

Change-Id: I4080a6b7468fff940435e2e780322d4ba1f0de49
Reviewed-on: http://gerrit.cloudera.org:8080/23334
Reviewed-by: Daniel Becker <daniel.becker@cloudera.com>
Tested-by: Impala Public Jenkins <impala-public-jenkins@cloudera.com>
This commit is contained in:
Daniel Vanko
2025-08-21 21:17:17 +02:00
committed by Daniel Becker
parent db92c88a4c
commit edd5ff6e2a
2 changed files with 36 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
fe/src/main/java/org/apache/impala/analysis/IcebergPartitionField.java
View File

@@ -54,8 +54,8 @@ public class IcebergPartitionField extends StmtNode {
Preconditions.checkState(type.isScalarType());
sourceId_ = sourceId;
fieldId_ = fieldId;
origFieldName_ = origFieldName;
fieldName_ = fieldName;
origFieldName_ = origFieldName.toLowerCase();
fieldName_ = fieldName.toLowerCase();
transform_ = transform;
type_ = (ScalarType)type;
}

34
testdata/workloads/functional-query/queries/QueryTest/iceberg-column-case-sensitivity-issue.test vendored
View File

@@ -18,3 +18,37 @@ STRING, STRING
---- RUNTIME_PROFILE
aggregation(SUM, NumRowGroups): 1
====
---- QUERY
# IMPALA-14290: check partition columns case insensitivity
CREATE TABLE partition_case_insensitive ( START_TIME TIMESTAMP, End_Time TIMESTAMP )
PARTITIONED BY SPEC (DAY(START_TIME), MONTH(enD_tIMe))
STORED AS ICEBERG;
---- RESULTS
'Table has been created.'
====
---- QUERY
ALTER TABLE partition_case_insensitive SET PARTITION SPEC (MONTH(start_TIME), DAY(end_time));
---- RESULTS
'Updated partition spec.'
====
---- QUERY
INSERT INTO partition_case_insensitive VALUES ("2023-11-27 00:00:00", "2024-08-12 00:00:00");
ALTER TABLE partition_case_insensitive DROP PARTITION (MONTH(sTARt_timE) = "2023-11");
---- RESULTS
'Dropped 1 partition(s)'
====
---- QUERY
INSERT INTO partition_case_insensitive VALUES ("2023-12-27 00:00:00", "2024-08-12 00:00:00");
INSERT INTO partition_case_insensitive VALUES ("2023-12-25 00:00:00", "2024-08-12 00:00:00");
INSERT INTO partition_case_insensitive VALUES ("2023-12-25 00:00:00", "2024-07-12 00:00:00");
INSERT INTO partition_case_insensitive VALUES ("2023-11-27 00:00:00", "2024-07-12 00:00:00");
SHOW PARTITIONS partition_case_insensitive;
---- LABELS
Partition, Number Of Rows, Number Of Files
---- RESULTS
'{"start_time_day":null,"end_time_month":null,"start_time_month":"646","end_time_day":"19916"}',1,1
'{"start_time_day":null,"end_time_month":null,"start_time_month":"647","end_time_day":"19916"}',1,1
'{"start_time_day":null,"end_time_month":null,"start_time_month":"647","end_time_day":"19947"}',2,2
---- TYPES
STRING, BIGINT, BIGINT
====