Github user fjh100456 commented on the issue:
https://github.com/apache/spark/pull/19218
Encounter two problems:
1. I tried to fix it in the order of 'compression' > 'parquet.compression'
> 'spark.sql.parquet.compression. codec', but found 'parquet.compression' may
come from a user 'set' command or from a table property which is specified on
creation, so which one should take precedence?
2. If I change the priorities, it seems partitioned tables and
non-partitioned tables are all need to be fixed, maybe I have to moves the code
forward, for example, to 'Insertintohadoopfsrelationcommand.scala'.
I would appreciate it if you could give me some advice. @gatorsmile
@dongjoon-hyun
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
