Skip to content

[fix][paimon] Fix Paimon date partition conversion failure #54043

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
morningman merged 5 commits into from
Aug 3, 2025
Merged

[fix][paimon] Fix Paimon date partition conversion failure #54043

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
b078da1
show partitions command for paimon
Jul 21, 2025
4d45f4e
Revert "show partitions command for paimon"
Jul 21, 2025
bc47b1e
add mvmt regression
Jul 23, 2025
8cd1ae6
fix can not convert paimon date partition
Jul 29, 2025
a50be83
optimize get column name logic
Jul 30, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
21 changes: 15 additions & 6 deletions fe/fe-core/src/main/java/org/apache/doris/datasource/paimon/PaimonUtil.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -59,6 +59,7 @@
import org.apache.paimon.types.MapType;
import org.apache.paimon.types.RowType;
import org.apache.paimon.types.VarCharType;
import org.apache.paimon.utils.DateTimeUtils;
import org.apache.paimon.utils.InstantiationUtil;
import org.apache.paimon.utils.Pair;
import org.apache.paimon.utils.Projection;
Expand Down Expand Up @@ -116,12 +117,23 @@ public static PaimonPartitionInfo generatePartitionInfo(List<Column> partitionCo
Map<String, PartitionItem> nameToPartitionItem = Maps.newHashMap();
Map<String, Partition> nameToPartition = Maps.newHashMap();
PaimonPartitionInfo partitionInfo = new PaimonPartitionInfo(nameToPartitionItem, nameToPartition);
List<Type> types = partitionColumns.stream()
.map(Column::getType)
.collect(Collectors.toList());
Map<String, Type> columnNameToType = partitionColumns.stream()
.collect(Collectors.toMap(Column::getName, Column::getType));

for (Partition partition : paimonPartitions) {
Map<String, String> spec = partition.spec();
StringBuilder sb = new StringBuilder();
for (Map.Entry<String, String> entry : spec.entrySet()) {
sb.append(entry.getKey()).append("=").append(entry.getValue()).append("/");
sb.append(entry.getKey()).append("=");
// Paimon stores DATE type as days since 1970-01-01 (epoch), so we convert the integer to a date string.
if (columnNameToType.getOrDefault(entry.getKey(), Type.NULL).isDateV2()) {
sb.append(DateTimeUtils.formatDate(Integer.parseInt(entry.getValue()))).append("/");
} else {
sb.append(entry.getValue()).append("/");
}
}
if (sb.length() > 0) {
sb.deleteCharAt(sb.length() - 1);
Expand All @@ -131,7 +143,7 @@ public static PaimonPartitionInfo generatePartitionInfo(List<Column> partitionCo
try {
// partition values return by paimon api, may have problem,
// to avoid affecting the query, we catch exceptions here
nameToPartitionItem.put(partitionName, toListPartitionItem(partitionName, partitionColumns));
nameToPartitionItem.put(partitionName, toListPartitionItem(partitionName, types));
} catch (Exception e) {
LOG.warn("toListPartitionItem failed, partitionColumns: {}, partitionValues: {}",
partitionColumns, partition.spec(), e);
Expand All @@ -140,11 +152,8 @@ public static PaimonPartitionInfo generatePartitionInfo(List<Column> partitionCo
return partitionInfo;
}

public static ListPartitionItem toListPartitionItem(String partitionName, List<Column> partitionColumns)
public static ListPartitionItem toListPartitionItem(String partitionName, List<Type> types)
throws AnalysisException {
List<Type> types = partitionColumns.stream()
.map(Column::getType)
.collect(Collectors.toList());
// Partition name will be in format: nation=cn/city=beijing
// parse it to get values "cn" and "beijing"
List<String> partitionValues = HiveUtil.toPartitionValues(partitionName);
Expand Down
3 changes: 3 additions & 0 deletions regression-test/data/mtmv_p0/test_paimon_mtmv.out
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -143,6 +143,9 @@ true
-- !date_partition_base_table --
1 2020-01-01

-- !date_type_partition --
1 2020-01-01

-- !date_partition --
1 2020-01-01

14 changes: 10 additions & 4 deletions regression-test/suites/mtmv_p0/test_paimon_mtmv.groovy
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -268,8 +268,7 @@ suite("test_paimon_mtmv", "p0,external,mtmv,external_docker,external_docker_dori

// date type will has problem
order_qt_date_partition_base_table "SELECT * FROM ${catalogName}.`test_paimon_spark`.date_partition"
test {
sql """
sql """
CREATE MATERIALIZED VIEW ${mvName}
BUILD DEFERRED REFRESH AUTO ON MANUAL
partition by (`create_date`)
Expand All @@ -278,8 +277,15 @@ suite("test_paimon_mtmv", "p0,external,mtmv,external_docker,external_docker_dori
AS
SELECT * FROM ${catalogName}.`test_paimon_spark`.date_partition;
"""
exception "Unable to find a suitable base table"
}
sql """
REFRESH MATERIALIZED VIEW ${mvName} auto;
"""
waitingMTMVTaskFinishedByMvName(mvName)
def showPaimonDateTypePartitionsResult = sql """show partitions from ${mvName}"""
logger.info("showPaimonDateTypePartitionsResult: " + showPaimonDateTypePartitionsResult.toString())
assertTrue(showPaimonDateTypePartitionsResult.toString().contains("p_20200101"))
order_qt_date_type_partition "select * FROM ${mvName}"
sql """drop materialized view if exists ${mvName};"""

sql """
CREATE MATERIALIZED VIEW ${mvName}
Expand Down