How to insert an n:m-relationship with technical primary keys generated by a sequence?

Posted by bitschnau on Stack Overflow See other posts from Stack Overflow or by bitschnau
Published on 2010-05-05T08:18:20Z Indexed on 2010/05/05 8:38 UTC
Read the original article Hit count: 291

Let's say I have two tables with several fields and in every table there is a primary key which is a technical id generated by a database sequence:

table1             table2
-------------      -------------
field11  <pk>      field21  <pk>
field12            field22

field11 and field21 are generated by sequences.

Also there is a n:m-relationship between table1 und table2, designed in table3:

table3
-------------
field11  <fk>
field21  <fk>

The ids in table1 und table2 are generated during the insert statement:

INSERT INTO table1 VALUES (table1_seq1.NEXTVAL, ...
INSERT INTO table2 VALUES (table2_seq1.NEXTVAL, ...

Therefore I don't know the primary key of the added row in the data-access-layer of my program, because the generation of the pk happens completely in the database.

What's the best practice to update table3 now? How can I gain access to the primary key of the rows I just inserted?

© Stack Overflow or respective owner

Related posts about database

Related posts about database-queries