(incubator-paimon) branch master updated: [format] Fix orc and parquet writer about timestamp not contains [local timezone] and [is_adjust_to_utc] (#2739)

[lzljs3620320]

This is an automated email from the ASF dual-hosted git repository.

lzljs3620320 pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-paimon.git


The following commit(s) were added to refs/heads/master by this push:
     new c6e182afa [format] Fix orc and parquet writer about timestamp not 
contains [local timezone] and [is_adjust_to_utc] (#2739)
c6e182afa is described below

commit c6e182afa121070d5c4d6c4dc353f39450ed2cf3
Author: leoyy0316 <105328124+leoyy0...@users.noreply.github.com>
AuthorDate: Sat Jan 20 23:14:25 2024 +0800

    [format] Fix orc and parquet writer about timestamp not contains [local 
timezone] and [is_adjust_to_utc] (#2739)
---
 .../paimon/format/orc/reader/OrcSplitReaderUtil.java   |  3 ++-
 .../paimon/format/parquet/ParquetSchemaConverter.java  | 18 +++++++++++++-----
 2 files changed, 15 insertions(+), 6 deletions(-)

diff --git 
a/paimon-format/src/main/java/org/apache/paimon/format/orc/reader/OrcSplitReaderUtil.java
 
b/paimon-format/src/main/java/org/apache/paimon/format/orc/reader/OrcSplitReaderUtil.java
index ae00821f3..882f1c753 100644
--- 
a/paimon-format/src/main/java/org/apache/paimon/format/orc/reader/OrcSplitReaderUtil.java
+++ 
b/paimon-format/src/main/java/org/apache/paimon/format/orc/reader/OrcSplitReaderUtil.java
@@ -74,8 +74,9 @@ public class OrcSplitReaderUtil {
             case DATE:
                 return TypeDescription.createDate();
             case TIMESTAMP_WITHOUT_TIME_ZONE:
-            case TIMESTAMP_WITH_LOCAL_TIME_ZONE:
                 return TypeDescription.createTimestamp();
+            case TIMESTAMP_WITH_LOCAL_TIME_ZONE:
+                return TypeDescription.createTimestampInstant();
             case ARRAY:
                 ArrayType arrayType = (ArrayType) type;
                 return 
TypeDescription.createList(toOrcType(arrayType.getElementType()));
diff --git 
a/paimon-format/src/main/java/org/apache/paimon/format/parquet/ParquetSchemaConverter.java
 
b/paimon-format/src/main/java/org/apache/paimon/format/parquet/ParquetSchemaConverter.java
index 24160df56..0e445f0b5 100644
--- 
a/paimon-format/src/main/java/org/apache/paimon/format/parquet/ParquetSchemaConverter.java
+++ 
b/paimon-format/src/main/java/org/apache/paimon/format/parquet/ParquetSchemaConverter.java
@@ -115,11 +115,11 @@ public class ParquetSchemaConverter {
             case TIMESTAMP_WITHOUT_TIME_ZONE:
                 TimestampType timestampType = (TimestampType) type;
                 return createTimestampWithLogicalType(
-                        name, timestampType.getPrecision(), repetition);
+                        name, timestampType.getPrecision(), repetition, false);
             case TIMESTAMP_WITH_LOCAL_TIME_ZONE:
                 LocalZonedTimestampType localZonedTimestampType = 
(LocalZonedTimestampType) type;
                 return createTimestampWithLogicalType(
-                        name, localZonedTimestampType.getPrecision(), 
repetition);
+                        name, localZonedTimestampType.getPrecision(), 
repetition, true);
             case ARRAY:
                 ArrayType arrayType = (ArrayType) type;
                 return ConversionPatterns.listOfElements(
@@ -151,13 +151,21 @@ public class ParquetSchemaConverter {
     }
 
     private static Type createTimestampWithLogicalType(
-            String name, int precision, Type.Repetition repetition) {
+            String name, int precision, Type.Repetition repetition, boolean 
isAdjustToUTC) {
         if (precision <= 3) {
-            return Types.primitive(INT64, 
repetition).as(OriginalType.TIMESTAMP_MILLIS).named(name);
+            return Types.primitive(INT64, repetition)
+                    .as(
+                            LogicalTypeAnnotation.timestampType(
+                                    isAdjustToUTC, 
LogicalTypeAnnotation.TimeUnit.MILLIS))
+                    .named(name);
         } else if (precision > 6) {
             return Types.primitive(PrimitiveType.PrimitiveTypeName.INT96, 
repetition).named(name);
         } else {
-            return Types.primitive(INT64, 
repetition).as(OriginalType.TIMESTAMP_MICROS).named(name);
+            return Types.primitive(INT64, repetition)
+                    .as(
+                            LogicalTypeAnnotation.timestampType(
+                                    isAdjustToUTC, 
LogicalTypeAnnotation.TimeUnit.MICROS))
+                    .named(name);
         }
     }