This week's book giveaway is in the Other Open Source APIs forum. We're giving away four copies of Storm Applied and have Sean Allen, Peter Pathirana & Matthew Jankowski on-line! See this thread for details.
I have a PL/SQL procedure that updates the parent table, and the child table. The procedure receives a string containing a delimited list of database table and column names and values that are used to locate and update a single row in table PARENT as well as any associated rows in table CHILD. If a row doesn't exist in table CHILD, a new row is created and if the row does exist, it is updated.
The string parameter (that is passed to the procedure) is parsed and SQL statements are generated according to which tables, columns and values are found. Native dynamic SQL is used to execute the generated SQL statements.
There is a one-to-many relationship between PARENT and CHILD. In other words, there may be zero, one or several (but never more than three) rows in table CHILD with the same PARENT_ID column value.
I am looking for the most efficient way to determine whether I need to insert into or update table CHILD. Currently (and this is code I inherited -- I didn't write it originally) I have this: [NOTE: This is a very simplified version of my code, for illustration only.]
Avi, That is the best way to do it. The only thing I would change is to make the query "select count(*) into L_COUNT". This could execute slightly faster depending on which is column 1. If column 1 is in the index, the execution speed would be the same, but the wildcard still protects you from future schema changes.