JosefinaArayaTapia opened a new issue, #5389:
URL: https://github.com/apache/hudi/issues/5389
When loading a hudi table in the AWS Glue data catalog and then sending a
data update via Spark, when reading the table again from spark, the history of
the data appears and not just one.
How can I solve that it brings me the last updated data.
Steps to reproduce the behavior:
1. Load Data:
`hudiOptions = {
'hoodie.table.name': table_name,
'hoodie.datasource.write.recordkey.field': 'id',
'hoodie.datasource.write.partitionpath.field': 'period',
'hoodie.datasource.write.precombine.field': 'last_update_time',
'hoodie.datasource.write.table.type':'COPY_ON_WRITE',
'hoodie.datasource.hive_sync.enable': 'true',
'hoodie.datasource.hive_sync.database':database_name,
'hoodie.datasource.hive_sync.table': table_name,
'hoodie.datasource.hive_sync.partition_fields': 'period',
'hoodie.datasource.hive_sync.support_timestamp': 'true'
}
`
`CLIENT.write \
.format('org.apache.hudi') \
.option('hoodie.datasource.write.operation', 'insert') \
.options(**hudiOptions) \
.mode('overwrite') \
.save('s3a://'+bucket_name+'/'+table_name)`
2. read and update one row in table hudi
`client= spark.sql("select * from table_name where id=59")
`
`updateDF = client.withColumn("cod_estado", when(client.cod_estado.isNull(),
lit('1')).otherwise(lit(None)))
`
`updateDF.write.format('org.apache.hudi').option('hoodie.datasource.write.operation',
'upsert').options(**hudiOptions) .mode('append')
.save('s3a://'+bucket_name+'/'+table_name)`
3. Query in Athena ---> OK
4. Read Parquet ---> OK
5. Query in EMR from Catalog Glue --> NOK
**Expected behavior**
Query in EMR from Catalog only show the last data.
**Environment Description**
* EMR: emr-6.4.0
* Hudi version : 0.8.0-amzn-0
* Spark version :3.1.2
* Hive version :3.1.2
* Hadoop version : 3.2.1
* Storage (HDFS/S3/GCS..) : S3
* Running on Docker? (yes/no) : no
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
