How to create a connection with Oracle using Spark Scala without loading data?

user_14992414 0
Viewed 363

Is there any standard Scala class/Object to create a simple connection with oracle database (using JDBC) but without loading a specific table ? I just want to create a simple connection then pass some query (create/update, etc.) and finally close the connection.

i have seen a code like :

val jdbcDF = spark.read
  .format("jdbc")
  .option("url", "jdbc:postgresql:dbserver")
  .option("dbtable", "schema.tablename")
  .option("user", "username")
  .option("password", "password")
  .load()

But it loads a specific table as mentionned in .option("dbtable", "schema.tablename").

apache-spark scala oracle sql
1 Answers

Here are all the JDBC examples. 'write' creates the table automatically, no need to explicitly create the table, if table already behavior depends on "mode" like overwrite, append, error, no-op.

If you want to update the table like delete selected rows via spark you could create a new DF with desired rows only and write it in overwrite mode, as suggested here

user_28461680
Related