Hi
I have a scanario where i have to populate 7 tables in the database . The Interface will start execution when the data is invoked for all the tables simultaneously . And at the end i am calling a stored procedure to mark the end of the entire cycle .
But even if the data for ony some of the tables arrive still i have to carry out the entire cycle.I mean even if data for only 3 tables come form the source i have to populate those 3 tables and call the stored procedure and then end the interface.
At design time it is not known data for how many tables will arrive . And some tables are small and some are very large . Say Out of 7 tables data might be inserted in 6 of them while the population for the 7th might still be going on . The objective is to call the last step only when data for all the tables which has arrived has been populated.
How do i implement the last step .
Can anybody tell me design for this scanarion.
You can use a BPM to achieve this. Steps involved in the BPM are
1. Receive step (to receive the source)
2. send step (here update all the 7 tables, use a nested data type).
3. send step(to execute the stored procedure)
Ensure the data type for updating 7 tables has cardinality 0..unbounded like this
<MT_TARGET>
<stmt1> 0..unbounded
<table>tab1</table>
<action>INSERT</action>
<access>
<node1>
</access>
</stmt1>
<stmt2> 0..unbounded
<table>tab1</table>
<action>INSERT</action>
<access>
<node1>
</access>
</stmt2>
.
.
.
</stmt7>
You can go for excetion handling if required....
Regards,
Jai Shankar.
Message was edited by: Jai Shankar
Add a comment