Hibernate version: 3.1

Name and version of the database you are using: MS SQL 2000

Following tables are present in our database.

Table 1
Part_Id - int identity (10) - Primary key
Part_Code - Char (10)
Part_Name - Varchar (50)



Table 2
Sequence_Id - int identity (10) - Primary key
Part_Id - int(10) - Foreign key -->Table1.Part_Id
Replacing_Part_Id - int(10) - Foreign key-->Table1.Part_Id


Is this many-to-many association or many-to-one association? Please let me know how to create hibernate mapping file for the above tables?

Many Thanks!

judging by your table layout, I'd guess that one part can only have one replacement at a time, but one part could be a replacement for many others.

<hibernate-mapping>
<class name="Table1ClassName" table="Table1">
<id name="partId" >
<column name="Part_Id">
<generator class="increment" />
</column>
</id>
<property name="partCode" type="string">
<column name="Part_Code" length="10" />
</property>
<property name="partName" type="string">
<column name="Part_Name" length="50" />
</property>
</class>
</hibernate-mapping>

<hibernate-mapping>
<class name="Table2ClassName" table="Table2">
<id name="sequenceId" >
<column name="Sequence_Id">
<generator class="increment"/>
</column>
</id>
<many-to-one name="table1PartId" class="Table1ClassName">
<column name="Part_Id">
</many-to-one>
<many-to-one name="replacingPartId" class="Table1ClassName">
<column name="Replacing_Part_Id">
</many-to-one>
</class>
</hibernate-mapping>

Hope this gives you an initial insight.....


Regards
Ahmad

Many thanks for answering basha. The mapping needs to be bidrectional though.

Let me know the functionality, i mean functionally how do you want to relate the two tables and what is the table / object that you would wish to use to drive the queries..

Regards
Ahmad

I forgot to mention that we use another column in the tables. The column
is called "deleted" and it indicates whether that row is deleted or not. This is done as a standard in order to avoid physical deletion of rows.


Table 1
Part_Id - int identity (10) - Primary key
Part_Code - Char (10)
Part_Name - Varchar (50)
deleted - int(1) - value zero or one


Table 2
Sequence_Id - int identity (10) - Primary key
Part_Id - int(10) - Foreign key -->Table1.Part_Id
Replacing_Part_Id - int(10) - Foreign key-->Table1.Part_Id
deleted - int(1) - value zero or one


And I created my mappings like this.





Note that I didn't use the second many-to-one mapping in Table2ClassName that you had given (which might be the solution to the problem that i am facing right now).


This one-to-many mapping of parent-child relationship worked fine except for one condition.

If I want to delete a part in the parent table, I actually update the delete column value of that part to one. Under this condition, I should update(mark as delete) child rows in table 2 as well. I cud do that for those rows that have the part_id as foriegn key in table 2 (since they are of collections in Table1ClassName). However there could be reference to this part id in the Replacing_Part_Id columns of Table2.




Now If I mark part_id 1 as deleted in table 1 I can easily mark the 1001 and 1002 sequence rows of table2 as deleted. How do I find the fourth row (sequnce 1004) which has reference to part_id 1?

Please provide java class definition for correct mappings. Many thanks for yoor help!!