vagetablechicken commented on issue #3468: URL: https://github.com/apache/incubator-doris/issues/3468#issuecomment-630563767
Users can load/update/delete data by load JSON format rows, which can be converted to PartialRows. e.g. routine load from Kafka(will soon support JSON format load). We support UPDATE, so JSON format source rows may fill different columns. ### Load case **denotations**: key column -> kc, value column -> vc We have a table, it has columns: kc1,kc2,vc1,vc2. `{"kc1":"0","kc2":"0","vc1":"0"} `- kc1=0,kc2=0 is not exsited, so this row will be INSERT. "vc2" will fill with default value. `{"kc1":"0","kc2":"0","vc2":"0"} `- inserted above. So this row will be UPDATE. `{"kc1":"0","vc2":"0"} `- "kc2" is missing. Current policy is to fill "kc2" with default value. `{"del_flag":"1", "kc1":"0","kc2":"0"}` - From a user perspective, del_flag column's name doesn't matter, We can support “customize the del_flag column name” when creating routine load. ---------------------------------------------------------------- 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. For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: commits-unsubscr...@doris.apache.org For additional commands, e-mail: commits-h...@doris.apache.org