Writing to iceberg table with sort order after coalesce

explain_so

== Parsed Logical Plan ==
OverwriteByExpression RelationV2[id#61L] prodhive.jzhuge.wso_244_so_1633558130, true, Map(matchByName -> true), true
+- Repartition 4, false
   +- Sort [id#3L ASC NULLS FIRST], true
      +- Project [id#0L AS id#3L]
         +- Range (0, 100000, step=1, splits=Some(6))

== Analyzed Logical Plan ==
OverwriteByExpression RelationV2[id#61L] prodhive.jzhuge.wso_244_so_1633558130, true, Map(matchByName -> true), true
+- Repartition 4, false
   +- Sort [id#3L ASC NULLS FIRST], true
      +- Project [id#0L AS id#3L]
         +- Range (0, 100000, step=1, splits=Some(6))

== Optimized Logical Plan ==
OverwriteByExpression RelationV2[id#61L] prodhive.jzhuge.wso_244_so_1633558130, true, Map(matchByName -> true), true, IcebergWrite schema=struct<id: bigint not null>, format=parquet
+- Sort [id#0L ASC NULLS FIRST], false
   +- RepartitionByExpression [id#0L ASC NULLS FIRST], 500
      +- Repartition 4, false
         +- Range (0, 100000, step=1, splits=Some(6))

== Physical Plan ==
OverwriteByExpression prodhive.jzhuge.wso_244_so_1633558130, [AlwaysTrue()], org.apache.spark.sql.util.CaseInsensitiveStringMap@7bdc4108, true, IcebergOverwriteByFilter(table=prodhive.jzhuge.wso_244_so_1633558130, format=PARQUET, filter=true)
+- *(2) Sort [id#0L ASC NULLS FIRST], false, 0
   +- Exchange rangepartitioning(id#0L ASC NULLS FIRST, 500)
      +- Coalesce 4
         +- *(1) Range (0, 100000, step=1, splits=6)

explain_so_part

== Parsed Logical Plan ==
OverwritePartitionsDynamic RelationV2[id#77L, part#78L] prodhive.jzhuge.wso_244_so_part_1633558130, Map(matchByName -> true), true
+- Repartition 4, false
   +- Project [id#3L, (id#3L % cast(2 as bigint)) AS part#65L]
      +- Sort [id#3L ASC NULLS FIRST], true
         +- Project [id#0L AS id#3L]
            +- Range (0, 100000, step=1, splits=Some(6))

== Analyzed Logical Plan ==
OverwritePartitionsDynamic RelationV2[id#77L, part#78L] prodhive.jzhuge.wso_244_so_part_1633558130, Map(matchByName -> true), true
+- Repartition 4, false
   +- Project [id#3L, (id#3L % cast(2 as bigint)) AS part#65L]
      +- Sort [id#3L ASC NULLS FIRST], true
         +- Project [id#0L AS id#3L]
            +- Range (0, 100000, step=1, splits=Some(6))

== Optimized Logical Plan ==
OverwritePartitionsDynamic RelationV2[id#77L, part#78L] prodhive.jzhuge.wso_244_so_part_1633558130, Map(matchByName -> true), true, IcebergWrite schema=struct<id: bigint not null,part: bigint>, format=parquet
+- Sort [part#65L ASC NULLS FIRST, id#0L ASC NULLS FIRST], false
   +- RepartitionByExpression [part#65L ASC NULLS FIRST, id#0L ASC NULLS FIRST], 500
      +- Repartition 4, false
         +- Project [id#0L, (id#0L % 2) AS part#65L]
            +- Range (0, 100000, step=1, splits=Some(6))

== Physical Plan ==
OverwritePartitionsDynamic prodhive.jzhuge.wso_244_so_part_1633558130, org.apache.spark.sql.util.CaseInsensitiveStringMap@7bdc4108, IcebergDynamicOverwrite(table=prodhive.jzhuge.wso_244_so_part_1633558130, format=PARQUET)
+- *(2) Sort [part#65L ASC NULLS FIRST, id#0L ASC NULLS FIRST], false, 0
   +- Exchange rangepartitioning(part#65L ASC NULLS FIRST, id#0L ASC NULLS FIRST, 500)
      +- Coalesce 4
         +- *(1) Project [id#0L, (id#0L % 2) AS part#65L]
            +- *(1) Range (0, 100000, step=1, splits=6)

iceberg_sort_order_and_coalesce.ipynb

tblproperties_so

spark-sql> SHOW TBLPROPERTIES jzhuge.wso_244_so_1633558130;
sort-order      id ASC NULLS FIRST
format  iceberg/parquet
current-snapshot-id     1949824784145854272
provider        iceberg
write.distribution-mode none

tblproperties_so_part

spark-sql> SHOW TBLPROPERTIES jzhuge.wso_244_so_part_1633558130;
current-snapshot-id     6079782482913875055
format  iceberg/parquet
sort-order      part ASC NULLS FIRST, id ASC NULLS FIRST
provider        iceberg
write.distribution-mode none