Github user sureshthalamati commented on the pull request:
https://github.com/apache/spark/pull/8676#issuecomment-139410536
@rxin
Even if spark is running on jdk1.7, customers using older version of
drivers will run into AbstractMethodError exception. I think adding
requirement for customers to use new drivers that implement getSchema()
function will be unnecessary.
After implementing the current approach I got curious on how the jdbc read
functionality finds the meta data and learned
org.apache.spark.sql.execution.datasources.jdbc.JDBCRDD.resolveTable also uses
s"SELECT * FROM $table WHERE 1=0" to get column information.
Alternative approach is to add getMetadataQuery(table:string) to the
JdbcDialect interface that helps to determine if table exists for write case ,
and column type information in the case of read instead of
getTableExistsQuery() as implemented in the current pull request. It might be
a milli second slower in the case of write call for dialects that specify
âselect 1 from $table limit 1", instead of âselect * from $table limit
1â. Advantage is one method to the interface will address both the cases.
Any comments ?
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
