hudi use partition path field as hive partition field error in flink

[onlywangyh]

To Reproduce

Steps to reproduce the behavior:

1. create a mysql table like :

CREATE TABLE `timeTypeTest` (
  `id` int(11) NOT NULL AUTO_INCREMENT,
  `datetime1` datetime DEFAULT NULL,
  `date1` date DEFAULT NULL,
  `datetime16` datetime(6) DEFAULT NULL,
  `time16` time DEFAULT NULL,
  `timestamp16` timestamp(6) NULL DEFAULT NULL,
  `timestamp16Partition` varchar(45) DEFAULT NULL,
  PRIMARY KEY (`id`),
  UNIQUE KEY `id_UNIQUE` (`id`)
) ENGINE=InnoDB AUTO_INCREMENT=2 DEFAULT CHARSET=latin1

2. insert a data
   insert into mydb.timeTypeTest values ('2', '2020-07-30 10:08:22', '2020-07-30', '2020-07-30 10:08:22.000000', '10:08:22', '2020-07-30 10:08:22.000000', '2020-07-30')

3. start a flink cdc to sink hudi with my config properties:

--hive-sync-enable=ture
--hive-sync-jdbc-url=jdbc:hive2://localhost:10000
--hive-sync-db=testDb
--hive-sync-table=testTable
--record-key-field=id
--partition-path-field=timestamp16
--hive-sync-partition-fields=inc_day
--hive-style-partitioning=true
--hive-sync-mode=jdbc
--hive-sync-username=hive
--hive-sync-password=hive

hoodie.deltastreamer.keygen.timebased.timestamp.type=EPOCHMILLISECONDS
hoodie.deltastreamer.keygen.timebased.output.dateformat=yyyy-MM-dd
hoodie.datasource.write.keygenerator.consistent.logical.timestamp.enabled=true
hive_sync.partition_extractor_class=org.apache.hudi.keygen.TimestampBasedAvroKeyGenerator

Expected behavior
create a hive table testTable with string partition field inc_day and add a partition "2020-07-30". But actually the partition field is timestamp16 with bigint type.

show partitions testTable;  ---- "2020-07-30"
select timestamp16 from testTable; ----- null

[danny0405]

Hello, i think this is expected because these two params:

--partition-path-field=timestamp16
--hive-sync-partition-fields=inc_day

should keep the same.

[onlywangyh]

If i keep the same params like --partition-path-field=timestamp16, --hive-sync-partition-fields=timestamp16. There will be some question:
1、In the schema the timestamp16 is a bigint type. When we use timestamp16 as a partition field. It will be a string type in hive schema. The bigint type can't convert to a string. So that select timestamp16 from testTable; will also return null.

2、In KeyGenerator we use the PARTITIONPATH_FIELD_NAME to get a partition path, we use the HIVE_SYNC_PARTITION_FIELDS as a partition field sync to hive . Sometimes these two params keep same will be good when the field is string type . But The TimestampBasedAvroKeyGenerator relies on timestamps for the partition field. The field values are interpreted as timestamps and not just converted to string while generating partition path value for records . When use TimestampBasedAvroKeyGenerator the 1596074902000L will converted to a string partition path like 2020-07-30 . In this case the hive partition field should use another field as a partition field instead of use the partition path field which the field values is another values. So the PARTITIONPATH_FIELD_NAME、HIVE_SYNC_PARTITION_FIELDS should diff to avoid the origin partition path field as a hive partition field cause some loss of precision、converted err

[codope]

HUDI-3978 to track the issue. Closing it as we have a patch in review.