search

apache / hudi

Upserts, Deletes And Incremental Processing on Big Data.
https://hudi.apache.org/
Apache License 2.0
5.39k stars 2.42k forks source link

[SUPPORT] use flinksql write hudi0.15 timestamp(3), use sql-client for hive catalog search timestamp(9),why? #12156

Open screamedddd opened 2 days ago

screamedddd commented 2 days ago

flinksql api execute sql CREATE TABLE hudi_table( ts timestamp(3), uuid VARCHAR(40) PRIMARY KEY NOT ENFORCED, rider VARCHAR(20), driver VARCHAR(20), fare DOUBLE, city VARCHAR(20) ) PARTITIONED BY (city) WITH ( 'connector' = 'hudi', 'path' = 'file:///tmp/hudi_table', 'table.type' = 'MERGE_ON_READ', 'hive_sync.enabled' = 'true', 'hive_sync.db' = 'hudi', 'hive_sync.table' = 'hudi_table', 'hive_sync.jdbc_url' = '', 'hive_sync.metastore.uris' = '', 'hive_sync.mode' = 'hms' );

after insert into data, hive db=hudi has three tables, hudi_table_ro and hudi_table_rt and hudi_table

this is a question: three tables column ts type is timestamp(9) , why?

danny0405 commented 2 days ago

Hive does not support timestamp precision yet, did you use the Hudi hive catalog then?